DB design calculate averages port from Excel

S

scratchtrax

I want to go from an Excel workbook to an Access DB application so that I can
add additional data and project items without compromising the formulas that
calculate the averages. I think if I persisted with Excel I could probably
get it to work, but I think it would be better to go Access. The purpose of
this DB will be:

"To develop average costs for project items and use these prices to develop
a reasonable overall average project cost. The average costs for project
items would be generated from criteria consisting of specified projects and
date ranges."

The average costs will ultimately be in the form of a report and will
consist of something like

Average Low Bid 2004|Average All Bidders 2004|Average Low Bid 2005|Average
All Bidders 2005|Average Low Bid 2006|Average All Bidders 2006|Average Low
Bid 2005-2006|Average All Bidders 2005-2006|Average Low Bid
2003-2006|Average All Bidders 2003-2006
obviously, the years will change.

ID ITEM
1 SIGNS $7.50 $16.00 |$12.00 $8.00
2 BARRICADES $115.00 $0.01 |$80.00 $125.00
3 DRUMS $75.00 $0.01 |$40.00 $70.00
4 WARNING DEVICES $60.00 $0.01 |$100.00 $5.00
14 MESSAGE SIGNS $4,050.00 $24,000.00|$6,000.00 $7,500.00
15 ASPHALT PRICE $15,000.00$15,000.00|$30,000.00$30,000.00
16 FUEL PRICE ADJUST $5,000.00 $5,000.00 |$8,000.00$8,000.00
17 SITE $20,000.00$96,000.00|$32,000.00$40,000.00

Project Number HWY-585M-09-2006-015|SGN-575-00-2001-036
Bidder Companies K Con Inc, Build It Co,|K Con Inc, Gen
Con Inc
 
J

Jeff Boyce

Think twice before leaving Excel...

Access is NOT a spreadsheet on steroids. To get the best use of Access'
relationally-oriented features/functions, you can't feed it 'sheet data.

You will need to learn about and use relational database design and
normalization. If these terms are new, you have a fairly steep learning
curve before you.

Is there a reason you wish to turn away from something you already know?

Is there a reason (other than "I think...") for selecting Access?

Good luck!

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/
 
L

Larry Daugherty

"Hold on there, Pardner..."

I usually agree with Jeff's posts and, in the main, do so here.
However, his (or my) assumption may be ahead of the mule:

By the time anyone seriously considers leaving Excel for Access they
are usually years late in making that decision. People use Excel as a
rudimentary data management facility long after the realization hits
that there have to be better ways. There's also an emotional
attachment to something in which they've invested so much time and
effort over the years. The glue that holds their Excel application
together is the user, who knows the desired outcome of every operation
and just what inputs must be massaged or provided to get there. There
is usually no documentation or guidance, other than verbal. Rarely
has anyone invested in a nice user interface with a menu of choices
and forms or even a few command buttons.

Depending on your existing knowledge of relational database design
there may be a big hump to get over to make the transition from Excel
to Access. Understanding normalization is the biggest key there.
Then, there is the Access interface and its universe of development
tools.

My impression of your data management requirements is that you have a
perfect candidate for an Access application.

For a professional Access developer, it would be a modest project.
For a neophyte starting from ground zero it could be a longish
project. Most of the neophyte's time and effort will be spend in
learning, applying new knowledge, learning some more, re-doing earlier
work, etc. The Access learning curve is long and steep.

Jeff's admonition that "Access is not a spreadsheet on steroids" is
absolutely, positive true. Their greatest similarity is that the
array of cells in a worksheet looks a lot like the datasheet view of a
table in Access. Even then, they aren't as close as they look. Also,
don't kid yourself that you'll "make an Access version of your Excel
workbook". It can't be done. The Excel workbook becomes one source
document that goes into the mix for creating the specifications and
subsequent design for the Access application.

HTH
 
J

Jeff Boyce

Larry

Not only will I suggest that folks stick with Excel, if that's working for
them... I'll even suggest they turn off their computer and resort to paper
and pencil, if what they describe needing doesn't seem to warrant the
investment of time/energy.

Of course, for job security purposes, I'm more than happy to design/build a
system after my users tell me they don't want to use paper/pencil <g>!

Regards

Jeff Boyce
Microsoft Office/Access MVP
 
S

scratchtrax

Thank you gentlemen. You both are correct. I don't believe that Access is
Excel on steroids and it has been a long time coming for this switch. For
me, I'm probably more familiar with Access than I am with Excel. I work with
relational databases, but I'll always think that I have a lot to learn. I
realized after I hit 'post' that I didn't explicitly ask the questions
(although you both provided helpful food for thought). Jeff maybe you're
right, maybe I should stay in Excel, I'm not sure. However, I think this
application will grow and I think it could grow nicely from Access. I do
realize that there will be a design hump (hence this post) and that this
beginning will later be further normalized and possibly repeated (the long
way). I have continued reading various posts and I think (although I know
that It'll need further work as it grows) I'll need to use one big table and
then query it for report generation. I was trying to resist this (main
thought for thinking I should stay in Excel) as there will be data that is
partly repeated. I can't figure how to come up with a design that will allow
it to be broken into peices without hiding data in the table names or not
being able to exclude specific items for the average calculation. Now I
think it will grow after this is accomplished and delivered, so that each and
every bid will be entered and not just the summary used for generalization
and estimate. If that were the case I think I could then come up with a
better design. Until then I guess I was just throwing it to you to see if I
was not considering one thing or another. Any further thoughts?
 
J

Jeff Boyce

If you can already envision "repeating" fields and fields with data embedded
in the fieldnames, then you are still in "spreadsheet" mode.

I usually recommend (and follow this myself) turning off the PC and grabbing
paper/pencil to rough out the entities (things about which you want to store
facts), attributes (the facts about the entities), and the relationships
among the entities (e.g., 1:1, 1:m, m:m).

When I have those laid out and the user(s) have signed off on the ideas (I
rarely expose them to the diagrams, but talk about "one customer can have
many orders"), I can then turn the PC back on and use my entities and
relationships to start setting up tables.

Good luck!

Regards

Jeff Boyce
Microsoft Office/Access MVP

scratchtrax said:
Thank you gentlemen. You both are correct. I don't believe that Access
is
Excel on steroids and it has been a long time coming for this switch. For
me, I'm probably more familiar with Access than I am with Excel. I work
with
relational databases, but I'll always think that I have a lot to learn. I
realized after I hit 'post' that I didn't explicitly ask the questions
(although you both provided helpful food for thought). Jeff maybe you're
right, maybe I should stay in Excel, I'm not sure. However, I think this
application will grow and I think it could grow nicely from Access. I do
realize that there will be a design hump (hence this post) and that this
beginning will later be further normalized and possibly repeated (the long
way). I have continued reading various posts and I think (although I know
that It'll need further work as it grows) I'll need to use one big table
and
then query it for report generation. I was trying to resist this (main
thought for thinking I should stay in Excel) as there will be data that is
partly repeated. I can't figure how to come up with a design that will
allow
it to be broken into peices without hiding data in the table names or not
being able to exclude specific items for the average calculation. Now I
think it will grow after this is accomplished and delivered, so that each
and
every bid will be entered and not just the summary used for generalization
and estimate. If that were the case I think I could then come up with a
better design. Until then I guess I was just throwing it to you to see if
I
was not considering one thing or another. Any further thoughts?
 
S

scratchtrax

This is where I remain for now, with pencil in hand. Thanks for the thoughts.
 
L

Larry Daugherty

A far better tool than Access, Excel, pencil & paper, stylus and clay
tablet and a precursor to all of them is the human brain! What I see
in the Original Post disposes me to believe that Access may be a
better solution paradigm than Excel. It is certainly worthy of
investigation and analysis: Just a few clues: there are different
entity types, there are many to many relationships, the data
population is expected to grow....

By the way, I like Excel. Some of my applications with which I'm most
pleased were done in Excel. While they all had data in them they were
really about importing data and calculating and producing new
workbooks, worksheets and charts. As a data management tool, Excel is
an also ran. For very simple data sets and a small data population
with extremely simple or no reporting requirements it's an easy, low
cost way to go.

Staying with a current paradigm on the sole justification that doing
so is *possible* is poor use of that foremost tool. As a race, we've
done a whole bunch more than just come down from the trees. When the
cost of doing business as you now do it exceeds the cost of
transforming to a new paradigm with greater efficiency and utility
then it's time to make a change. Have you noticed that production
(all those jobs) will go where the production cost is lowest for that
particular job? [Warren Buffet is now shifting the focus of his
investing toward the Orient].

The trick is in knowing all of those costs accurately. Determine what
is or should be the time for ROI within your organization. It will be
a rule of thumb rather than an absolute but the bean counters should
know it.

Most organizations are sloppy about monitoring how efficient they are
in their current operations so the metrics are blown at the start.

Try this: Every time you know you're going to do something using your
Excel (or other spreadsheet "database") application, take note of what
you're about to do and then note the time right now. Do the thing.
Record the time you finished with the task. Do that for a week. For
greater accuracy, do it for a longer period of time. Eventually
you'll be able to characterize the time to do something using that
tool. Also, you mentioned that your application will be getting more
use. That means that your monitoring should continue over time.
Maybe pick one day each week to monitor to track the increasing
utilizatization.

Next, determine the Loaded Labor Cost per hour for your organization.
This can be done even for government organizations. Gather the
information to determine all money or goods that are expended per
annum. That means the formal budget, resources consumed from other
organizations but not billed back (the value of real property used
such as buildings, furniture and equipment, parking lots ...) etc.
Divide that huge number by another one: the number of employees times
their paid hours per year.

With the above as the hourly rate, multiply the total hours per day
your current application is used by that rate times the number of
business days per year. That's your annual cost of using that
application to produce the results you get from it. What ever other
things you might have in the fire, doing the above is a very
worthwhile thing to do even if you only track the times used. If your
organization is doing things right then the bean counters can give you
an exact number for the current Loaded Labor Rate. If not, one of
them may even have the super-total cost of operations tucked away
somewhere. They'll either be pleased or shocked that you asked.

Next is to create some informal specifications for the replacement
application. While Access is a natural, I wouldn't even specify
Access unless you want to get your hand in and take over the
maintenance of an eventual Access application. Decide and define all
of the problems you'll solve and outputs required. Describe the
necessary workflows. Describe the necessary Forms and Reports. For
each critical workflow that you have measured in your current
application state a performance threshold. In most cases the
performance thresholds should be significantly higher. Also,
typographical errors should be significantly lower and accidental
damage to data due to mucking in the tables will be a thing of the
past.

Get a consultant to evaluate your prospective project and to work up
an estimate to complete it.

Now you have some things that you can compare. Some of the figures
may be pretty soft but you'll have an idea. If you already have your
standard for ROI then you'll also have an idea on which side of the
go/no-go threshold you stand.

Here's a thought: If you'll send me a populated copy of your Excel
workbook (which I would treat as confidential) and be willing to
collaborate with me via phone and email a few times then I can knock
together an Access demo application using your data. It would be done
in my free time over the next week or two. No charge. No, this is
not an offer of a free real live application. It's just to give you
an idea of the difference in performance and in results between two
ways of doing things.

That would help me discharge a New Year's Resolution.

If you're up for it, let's get started. In any case, the rest of the
post is intended for your guidance. You may know much or all of it
already.

HTH
--
-Larry-
--

scratchtrax said:
Thank you gentlemen. You both are correct. I don't believe that Access is
Excel on steroids and it has been a long time coming for this switch. For
me, I'm probably more familiar with Access than I am with Excel. I work with
relational databases, but I'll always think that I have a lot to learn. I
realized after I hit 'post' that I didn't explicitly ask the questions
(although you both provided helpful food for thought). Jeff maybe you're
right, maybe I should stay in Excel, I'm not sure. However, I think this
application will grow and I think it could grow nicely from Access. I do
realize that there will be a design hump (hence this post) and that this
beginning will later be further normalized and possibly repeated (the long
way). I have continued reading various posts and I think (although I know
that It'll need further work as it grows) I'll need to use one big table and
then query it for report generation. I was trying to resist this (main
thought for thinking I should stay in Excel) as there will be data that is
partly repeated. I can't figure how to come up with a design that will allow
it to be broken into peices without hiding data in the table names or not
being able to exclude specific items for the average calculation. Now I
think it will grow after this is accomplished and delivered, so that each and
every bid will be entered and not just the summary used for generalization
and estimate. If that were the case I think I could then come up with a
better design. Until then I guess I was just throwing it to you to see if I
was not considering one thing or another. Any further thoughts?
 
S

scratchtrax

Larry you're a gem and I thank you very much. This project is also as a
favor in my spare time and although it’s for a sister department at work,
it’s free of charge for me as well.

Although I don't have actual results for comparison, I've been considering
the hypothetical difference in results and performance between different ways
of doing things. No this hasn't gotten me too far yet, but it has made me
consider several different things I wasn't thinking about when I originally
posted this message. The rest of the post I will keep for my guidance in
future endeavors. As for this one, your last post has already proven to me
that it is probably not worth the time.

I say this because upon further examination the Excel tables that I have
really aren't as mature or refined as I once thought. The expansion of
different items that are attempting to be tracked and incorporated are just
now being entered and that data is not complete in its entirety. This may
also be another argument for why it should probably move from Excel.
However, I don't want to disappoint you if I send you these tables and you
see that the data management application isn't really all its cracked up to
be. You also made another point about calculating the time for the ROI with
the number of people that will use this application. As it was originally
requested, only a few people would use this application and the output of
this application would be utilized by several more. The data items that it
attempts to manage may grow, but the growth in the number of people that will
be reached by its product will be marginal. So really, I'm now treating this
as a learning exercise and selfishly attempting to develop this so that it
will help me more in the future when it does count (if you will).

So, with all of that ignorance in mind :) if you still want to help and if
this will still discharge a New Years Resolution that would be great. (I'd
love to know what that resolution is exactly by the way, unless of course its
classified :) I am still in the "interview the end users" phase and am
trying to figure out what all will be needed and how it could possibly be
done and what it could possibly grow into. I don't think one table is a good
idea at all and am no longer considering this method. I am a bit concerned
about relationships and how to get to the attributes I'll need, as I need
them, but I think it can be done with the way I'm considering doing it now.
Unless this no longer fits your parameters (which would be no problem and I
would perfectly understand), let me find out a few more things and attempt a
design that I think could work and then I'll send you that with the Excel
tables as a source and then you can review & decide from there.

However it works out, thank you very much for even considering such an offer.

Best Regards
--
http://njgin.aclink.org


Larry Daugherty said:
A far better tool than Access, Excel, pencil & paper, stylus and clay
tablet and a precursor to all of them is the human brain! What I see
in the Original Post disposes me to believe that Access may be a
better solution paradigm than Excel. It is certainly worthy of
investigation and analysis: Just a few clues: there are different
entity types, there are many to many relationships, the data
population is expected to grow....

By the way, I like Excel. Some of my applications with which I'm most
pleased were done in Excel. While they all had data in them they were
really about importing data and calculating and producing new
workbooks, worksheets and charts. As a data management tool, Excel is
an also ran. For very simple data sets and a small data population
with extremely simple or no reporting requirements it's an easy, low
cost way to go.

Staying with a current paradigm on the sole justification that doing
so is *possible* is poor use of that foremost tool. As a race, we've
done a whole bunch more than just come down from the trees. When the
cost of doing business as you now do it exceeds the cost of
transforming to a new paradigm with greater efficiency and utility
then it's time to make a change. Have you noticed that production
(all those jobs) will go where the production cost is lowest for that
particular job? [Warren Buffet is now shifting the focus of his
investing toward the Orient].

The trick is in knowing all of those costs accurately. Determine what
is or should be the time for ROI within your organization. It will be
a rule of thumb rather than an absolute but the bean counters should
know it.

Most organizations are sloppy about monitoring how efficient they are
in their current operations so the metrics are blown at the start.

Try this: Every time you know you're going to do something using your
Excel (or other spreadsheet "database") application, take note of what
you're about to do and then note the time right now. Do the thing.
Record the time you finished with the task. Do that for a week. For
greater accuracy, do it for a longer period of time. Eventually
you'll be able to characterize the time to do something using that
tool. Also, you mentioned that your application will be getting more
use. That means that your monitoring should continue over time.
Maybe pick one day each week to monitor to track the increasing
utilizatization.

Next, determine the Loaded Labor Cost per hour for your organization.
This can be done even for government organizations. Gather the
information to determine all money or goods that are expended per
annum. That means the formal budget, resources consumed from other
organizations but not billed back (the value of real property used
such as buildings, furniture and equipment, parking lots ...) etc.
Divide that huge number by another one: the number of employees times
their paid hours per year.

With the above as the hourly rate, multiply the total hours per day
your current application is used by that rate times the number of
business days per year. That's your annual cost of using that
application to produce the results you get from it. What ever other
things you might have in the fire, doing the above is a very
worthwhile thing to do even if you only track the times used. If your
organization is doing things right then the bean counters can give you
an exact number for the current Loaded Labor Rate. If not, one of
them may even have the super-total cost of operations tucked away
somewhere. They'll either be pleased or shocked that you asked.

Next is to create some informal specifications for the replacement
application. While Access is a natural, I wouldn't even specify
Access unless you want to get your hand in and take over the
maintenance of an eventual Access application. Decide and define all
of the problems you'll solve and outputs required. Describe the
necessary workflows. Describe the necessary Forms and Reports. For
each critical workflow that you have measured in your current
application state a performance threshold. In most cases the
performance thresholds should be significantly higher. Also,
typographical errors should be significantly lower and accidental
damage to data due to mucking in the tables will be a thing of the
past.

Get a consultant to evaluate your prospective project and to work up
an estimate to complete it.

Now you have some things that you can compare. Some of the figures
may be pretty soft but you'll have an idea. If you already have your
standard for ROI then you'll also have an idea on which side of the
go/no-go threshold you stand.

Here's a thought: If you'll send me a populated copy of your Excel
workbook (which I would treat as confidential) and be willing to
collaborate with me via phone and email a few times then I can knock
together an Access demo application using your data. It would be done
in my free time over the next week or two. No charge. No, this is
not an offer of a free real live application. It's just to give you
an idea of the difference in performance and in results between two
ways of doing things.

That would help me discharge a New Year's Resolution.

If you're up for it, let's get started. In any case, the rest of the
post is intended for your guidance. You may know much or all of it
already.

HTH
--
-Larry-
--

scratchtrax said:
Thank you gentlemen. You both are correct. I don't believe that Access is
Excel on steroids and it has been a long time coming for this switch. For
me, I'm probably more familiar with Access than I am with Excel. I work with
relational databases, but I'll always think that I have a lot to learn. I
realized after I hit 'post' that I didn't explicitly ask the questions
(although you both provided helpful food for thought). Jeff maybe you're
right, maybe I should stay in Excel, I'm not sure. However, I think this
application will grow and I think it could grow nicely from Access. I do
realize that there will be a design hump (hence this post) and that this
beginning will later be further normalized and possibly repeated (the long
way). I have continued reading various posts and I think (although I know
that It'll need further work as it grows) I'll need to use one big table and
then query it for report generation. I was trying to resist this (main
thought for thinking I should stay in Excel) as there will be data that is
partly repeated. I can't figure how to come up with a design that will allow
it to be broken into peices without hiding data in the table names or not
being able to exclude specific items for the average calculation. Now I
think it will grow after this is accomplished and delivered, so that each and
every bid will be entered and not just the summary used for generalization
and estimate. If that were the case I think I could then come up with a
better design. Until then I guess I was just throwing it to you to see if I
was not considering one thing or another. Any further thoughts?
 
L

Larry Daugherty

If we do continue this dialogue beyond this post we should take it off
line. You can easily decode my email address. I don't have yours.

I inferred from your first post that you were already using an Excel
spreadsheet to track project data and were abstracting information
from it. I assumed several tens of projects with all of the
constituent data.

If you don't have a base line for comparison and the pre-existing sets
of data then a demo for comparison purposes wouldn't have much impact.

As you start from ground zero, concepts are simple and Excel is the
evident low cost, low effort solution. There is no real design of the
information at the outset and you need only be sure that everything
you want to track is actually entered into the spreadsheet.
Organization would be on a "row per project" basis or a "worksheet per
project" basis. In the beginning, life is simple. As you begin to
relate different parts of the data to other parts then some thinking
is involved and some rudimentary design is employed. Some formulas
are embedded. Position often determines data significance. Usually
the sophistication of the solution increases with time and use.
"Programming" is usually ad-hoc. Needs aren't anticipated but are met
when they're encountered. The designer of the tool is usually it's
primary user so the tool relies on that knowledgeable user being the
user. There is no documentation. There is nothing sophisticated
about the tool whatever. [Note that Excel can be programmed to
provide a very sophisticated user interface and tome very powerful
results - it usually isn't done for informal projects], Iterative
development is the rule rather than the exception: often adding new
functionality requires re-working existing functionality. Excel is
intended to be a user level tool and ad hoc programming is often the
rule.

Getting going with Access requires a lot more thought and design
beforehand. It isn't rocket science it's just more difficult to learn
in the first place. People who regularly design Access applications
get into the habit of doing a bit more thinking at the front end of a
project. Also, experience causes Access developers to do some things
as 2nd nature. While Access can be developed iteratively, it isn't a
handy thing to do. Access is intended to be primarily a platform for
developers to develop applications which users will then use. It is
not really a good thing to be continually developing.

The crucial thing at the moment is whether you are actually using the
Excel tool to gather, track and, possibly, present information. I
didn't expect that there would be a sophisticated data management
application. Quite the contrary. Don't worry that I would fault what
you have. Things like that start out as just a handy means of
recording interesting data and then just grow. They are usually the
proactive creation of someone who needs some information on a regular
basis that no one else provides.

I don't really know what to tell you regarding going forward with a
demo. You're kind of hedging. My offer is based on there being a
fairly well populated Excel spreadsheet with real project data. In
addition I would need such specifications as might exist or could be
generated by you. In addition I would need to be able to talk to you
on the phone once or twice. Remember, I want to just generate a demo
Access application that would show the difference in performing
representative operations in Access. The demo isn't to be a free
application. That's contrary to the charter for these newsgroups.
Besides, that's what I do professionally and for which I charge money.
The demo will gracefully expire after you've had a change to run it
for a month or two.

I also don't want a "free coaching" job. I want to approach this thin
in a friendly but businesslike manner and give you and your colleagues
a utility to compare to the Excel tool.

If you want to go ahead with it then lets do so. Respond with your
workbook attached to an email that includes your name, your phone
number(s) and the best range of times to call. I'm in Southern
California and keep strange schedules.

What ever the case, do continue to interview your prospective users in
both a Q&A mode and request them to describe what they'd like to get
from an application like you're considering. You might raise the idea
of Excel and "other" platforms.

My resolution? At the first opportunity in 2008 that I recognize to
offer a service that I believe would be useful to the OP in one of the
Access newsgroups, do so. The offer and service have to be consistent
with newsgroup values. The offer is not a blank check for service but
exactly as I make the offer.

Believe me, I have received so much over the years from contributors
in these newsgroups and from their private web sites, especially the
MVPs.

Best regards,
--
-Larry-
--

scratchtrax said:
Larry you're a gem and I thank you very much. This project is also as a
favor in my spare time and although it's for a sister department at work,
it's free of charge for me as well.

Although I don't have actual results for comparison, I've been considering
the hypothetical difference in results and performance between different ways
of doing things. No this hasn't gotten me too far yet, but it has made me
consider several different things I wasn't thinking about when I originally
posted this message. The rest of the post I will keep for my guidance in
future endeavors. As for this one, your last post has already proven to me
that it is probably not worth the time.

I say this because upon further examination the Excel tables that I have
really aren't as mature or refined as I once thought. The expansion of
different items that are attempting to be tracked and incorporated are just
now being entered and that data is not complete in its entirety. This may
also be another argument for why it should probably move from Excel.
However, I don't want to disappoint you if I send you these tables and you
see that the data management application isn't really all its cracked up to
be. You also made another point about calculating the time for the ROI with
the number of people that will use this application. As it was originally
requested, only a few people would use this application and the output of
this application would be utilized by several more. The data items that it
attempts to manage may grow, but the growth in the number of people that will
be reached by its product will be marginal. So really, I'm now treating this
as a learning exercise and selfishly attempting to develop this so that it
will help me more in the future when it does count (if you will).

So, with all of that ignorance in mind :) if you still want to help and if
this will still discharge a New Years Resolution that would be great. (I'd
love to know what that resolution is exactly by the way, unless of course its
classified :) I am still in the "interview the end users" phase and am
trying to figure out what all will be needed and how it could possibly be
done and what it could possibly grow into. I don't think one table is a good
idea at all and am no longer considering this method. I am a bit concerned
about relationships and how to get to the attributes I'll need, as I need
them, but I think it can be done with the way I'm considering doing it now.
Unless this no longer fits your parameters (which would be no problem and I
would perfectly understand), let me find out a few more things and attempt a
design that I think could work and then I'll send you that with the Excel
tables as a source and then you can review & decide from there.

However it works out, thank you very much for even considering such an offer.

Best Regards
--
http://njgin.aclink.org


Larry Daugherty said:
A far better tool than Access, Excel, pencil & paper, stylus and clay
tablet and a precursor to all of them is the human brain! What I see
in the Original Post disposes me to believe that Access may be a
better solution paradigm than Excel. It is certainly worthy of
investigation and analysis: Just a few clues: there are different
entity types, there are many to many relationships, the data
population is expected to grow....

By the way, I like Excel. Some of my applications with which I'm most
pleased were done in Excel. While they all had data in them they were
really about importing data and calculating and producing new
workbooks, worksheets and charts. As a data management tool, Excel is
an also ran. For very simple data sets and a small data population
with extremely simple or no reporting requirements it's an easy, low
cost way to go.

Staying with a current paradigm on the sole justification that doing
so is *possible* is poor use of that foremost tool. As a race, we've
done a whole bunch more than just come down from the trees. When the
cost of doing business as you now do it exceeds the cost of
transforming to a new paradigm with greater efficiency and utility
then it's time to make a change. Have you noticed that production
(all those jobs) will go where the production cost is lowest for that
particular job? [Warren Buffet is now shifting the focus of his
investing toward the Orient].

The trick is in knowing all of those costs accurately. Determine what
is or should be the time for ROI within your organization. It will be
a rule of thumb rather than an absolute but the bean counters should
know it.

Most organizations are sloppy about monitoring how efficient they are
in their current operations so the metrics are blown at the start.

Try this: Every time you know you're going to do something using your
Excel (or other spreadsheet "database") application, take note of what
you're about to do and then note the time right now. Do the thing.
Record the time you finished with the task. Do that for a week. For
greater accuracy, do it for a longer period of time. Eventually
you'll be able to characterize the time to do something using that
tool. Also, you mentioned that your application will be getting more
use. That means that your monitoring should continue over time.
Maybe pick one day each week to monitor to track the increasing
utilizatization.

Next, determine the Loaded Labor Cost per hour for your organization.
This can be done even for government organizations. Gather the
information to determine all money or goods that are expended per
annum. That means the formal budget, resources consumed from other
organizations but not billed back (the value of real property used
such as buildings, furniture and equipment, parking lots ...) etc.
Divide that huge number by another one: the number of employees times
their paid hours per year.

With the above as the hourly rate, multiply the total hours per day
your current application is used by that rate times the number of
business days per year. That's your annual cost of using that
application to produce the results you get from it. What ever other
things you might have in the fire, doing the above is a very
worthwhile thing to do even if you only track the times used. If your
organization is doing things right then the bean counters can give you
an exact number for the current Loaded Labor Rate. If not, one of
them may even have the super-total cost of operations tucked away
somewhere. They'll either be pleased or shocked that you asked.

Next is to create some informal specifications for the replacement
application. While Access is a natural, I wouldn't even specify
Access unless you want to get your hand in and take over the
maintenance of an eventual Access application. Decide and define all
of the problems you'll solve and outputs required. Describe the
necessary workflows. Describe the necessary Forms and Reports. For
each critical workflow that you have measured in your current
application state a performance threshold. In most cases the
performance thresholds should be significantly higher. Also,
typographical errors should be significantly lower and accidental
damage to data due to mucking in the tables will be a thing of the
past.

Get a consultant to evaluate your prospective project and to work up
an estimate to complete it.

Now you have some things that you can compare. Some of the figures
may be pretty soft but you'll have an idea. If you already have your
standard for ROI then you'll also have an idea on which side of the
go/no-go threshold you stand.

Here's a thought: If you'll send me a populated copy of your Excel
workbook (which I would treat as confidential) and be willing to
collaborate with me via phone and email a few times then I can knock
together an Access demo application using your data. It would be done
in my free time over the next week or two. No charge. No, this is
not an offer of a free real live application. It's just to give you
an idea of the difference in performance and in results between two
ways of doing things.

That would help me discharge a New Year's Resolution.

If you're up for it, let's get started. In any case, the rest of the
post is intended for your guidance. You may know much or all of it
already.

HTH
--
-Larry-
--

Thank you gentlemen. You both are correct. I don't believe
that
Access is
Excel on steroids and it has been a long time coming for this switch. For
me, I'm probably more familiar with Access than I am with Excel.
I
work with
relational databases, but I'll always think that I have a lot to learn. I
realized after I hit 'post' that I didn't explicitly ask the questions
(although you both provided helpful food for thought). Jeff
maybe
you're
right, maybe I should stay in Excel, I'm not sure. However, I
think
this
application will grow and I think it could grow nicely from
Access.
I do
realize that there will be a design hump (hence this post) and
that
this
beginning will later be further normalized and possibly repeated (the long
way). I have continued reading various posts and I think
(although
I know
that It'll need further work as it grows) I'll need to use one
big
table and
then query it for report generation. I was trying to resist
this
(main
thought for thinking I should stay in Excel) as there will be
data
that is
partly repeated. I can't figure how to come up with a design
that
will allow
it to be broken into peices without hiding data in the table
names
or not
being able to exclude specific items for the average
calculation.
Now I
think it will grow after this is accomplished and delivered, so
that
each and
every bid will be entered and not just the summary used for generalization
and estimate. If that were the case I think I could then come
up
with a
better design. Until then I guess I was just throwing it to you
to
see if I
was not considering one thing or another. Any further thoughts?
--
http://njgin.aclink.org


:

"Hold on there, Pardner..."

I usually agree with Jeff's posts and, in the main, do so here.
However, his (or my) assumption may be ahead of the mule:

By the time anyone seriously considers leaving Excel for
Access
they
are usually years late in making that decision. People use
Excel
as a
rudimentary data management facility long after the
realization
hits
that there have to be better ways. There's also an emotional
attachment to something in which they've invested so much time and
effort over the years. The glue that holds their Excel application
together is the user, who knows the desired outcome of every operation
and just what inputs must be massaged or provided to get
there.
There
is usually no documentation or guidance, other than verbal. Rarely
has anyone invested in a nice user interface with a menu of choices
and forms or even a few command buttons.

Depending on your existing knowledge of relational database design
there may be a big hump to get over to make the transition
from
Excel
to Access. Understanding normalization is the biggest key there.
Then, there is the Access interface and its universe of development
tools.

My impression of your data management requirements is that you have a
perfect candidate for an Access application.

For a professional Access developer, it would be a modest project.
For a neophyte starting from ground zero it could be a longish
project. Most of the neophyte's time and effort will be spend in
learning, applying new knowledge, learning some more, re-doing earlier
work, etc. The Access learning curve is long and steep.

Jeff's admonition that "Access is not a spreadsheet on
steroids"
is
absolutely, positive true. Their greatest similarity is that the
array of cells in a worksheet looks a lot like the datasheet
view
of a
table in Access. Even then, they aren't as close as they
look.
Also,
don't kid yourself that you'll "make an Access version of your Excel
workbook". It can't be done. The Excel workbook becomes one source
document that goes into the mix for creating the
specifications
and
subsequent design for the Access application.

HTH
--
-Larry-
--

message Think twice before leaving Excel...

Access is NOT a spreadsheet on steroids. To get the best use of
Access'
relationally-oriented features/functions, you can't feed it 'sheet
data.

You will need to learn about and use relational database
design
and
normalization. If these terms are new, you have a fairly steep
learning
curve before you.

Is there a reason you wish to turn away from something you already
know?

Is there a reason (other than "I think...") for selecting Access?

Good luck!

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/


message
I want to go from an Excel workbook to an Access DB application so
that I
can
add additional data and project items without compromising the
formulas
that
calculate the averages. I think if I persisted with Excel
I
could
probably
get it to work, but I think it would be better to go
Access.
The
purpose
of
this DB will be:

"To develop average costs for project items and use these prices
to
develop
a reasonable overall average project cost. The average
costs
for
project
items would be generated from criteria consisting of specified
projects
and
date ranges."

The average costs will ultimately be in the form of a
report
and
will
consist of something like

Average Low Bid 2004|Average All Bidders 2004|Average Low Bid
2005|Average
All Bidders 2005|Average Low Bid 2006|Average All Bidders
2006|Average Low
Bid 2005-2006|Average All Bidders 2005-2006|Average Low Bid
2003-2006|Average All Bidders 2003-2006
 
J

Jeff Boyce

Larry

I'm with you most of the way, except for that part about Access not lending
itself to iterative development.

Maybe it has just been me and my personal approach, but iterative
development (using Access) is what I've been doing since '94, when Access
first came out.

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/

Larry Daugherty said:
If we do continue this dialogue beyond this post we should take it off
line. You can easily decode my email address. I don't have yours.

I inferred from your first post that you were already using an Excel
spreadsheet to track project data and were abstracting information
from it. I assumed several tens of projects with all of the
constituent data.

If you don't have a base line for comparison and the pre-existing sets
of data then a demo for comparison purposes wouldn't have much impact.

As you start from ground zero, concepts are simple and Excel is the
evident low cost, low effort solution. There is no real design of the
information at the outset and you need only be sure that everything
you want to track is actually entered into the spreadsheet.
Organization would be on a "row per project" basis or a "worksheet per
project" basis. In the beginning, life is simple. As you begin to
relate different parts of the data to other parts then some thinking
is involved and some rudimentary design is employed. Some formulas
are embedded. Position often determines data significance. Usually
the sophistication of the solution increases with time and use.
"Programming" is usually ad-hoc. Needs aren't anticipated but are met
when they're encountered. The designer of the tool is usually it's
primary user so the tool relies on that knowledgeable user being the
user. There is no documentation. There is nothing sophisticated
about the tool whatever. [Note that Excel can be programmed to
provide a very sophisticated user interface and tome very powerful
results - it usually isn't done for informal projects], Iterative
development is the rule rather than the exception: often adding new
functionality requires re-working existing functionality. Excel is
intended to be a user level tool and ad hoc programming is often the
rule.

Getting going with Access requires a lot more thought and design
beforehand. It isn't rocket science it's just more difficult to learn
in the first place. People who regularly design Access applications
get into the habit of doing a bit more thinking at the front end of a
project. Also, experience causes Access developers to do some things
as 2nd nature. While Access can be developed iteratively, it isn't a
handy thing to do. Access is intended to be primarily a platform for
developers to develop applications which users will then use. It is
not really a good thing to be continually developing.

The crucial thing at the moment is whether you are actually using the
Excel tool to gather, track and, possibly, present information. I
didn't expect that there would be a sophisticated data management
application. Quite the contrary. Don't worry that I would fault what
you have. Things like that start out as just a handy means of
recording interesting data and then just grow. They are usually the
proactive creation of someone who needs some information on a regular
basis that no one else provides.

I don't really know what to tell you regarding going forward with a
demo. You're kind of hedging. My offer is based on there being a
fairly well populated Excel spreadsheet with real project data. In
addition I would need such specifications as might exist or could be
generated by you. In addition I would need to be able to talk to you
on the phone once or twice. Remember, I want to just generate a demo
Access application that would show the difference in performing
representative operations in Access. The demo isn't to be a free
application. That's contrary to the charter for these newsgroups.
Besides, that's what I do professionally and for which I charge money.
The demo will gracefully expire after you've had a change to run it
for a month or two.

I also don't want a "free coaching" job. I want to approach this thin
in a friendly but businesslike manner and give you and your colleagues
a utility to compare to the Excel tool.

If you want to go ahead with it then lets do so. Respond with your
workbook attached to an email that includes your name, your phone
number(s) and the best range of times to call. I'm in Southern
California and keep strange schedules.

What ever the case, do continue to interview your prospective users in
both a Q&A mode and request them to describe what they'd like to get
from an application like you're considering. You might raise the idea
of Excel and "other" platforms.

My resolution? At the first opportunity in 2008 that I recognize to
offer a service that I believe would be useful to the OP in one of the
Access newsgroups, do so. The offer and service have to be consistent
with newsgroup values. The offer is not a blank check for service but
exactly as I make the offer.

Believe me, I have received so much over the years from contributors
in these newsgroups and from their private web sites, especially the
MVPs.

Best regards,
--
-Larry-
--

scratchtrax said:
Larry you're a gem and I thank you very much. This project is also as a
favor in my spare time and although it's for a sister department at work,
it's free of charge for me as well.

Although I don't have actual results for comparison, I've been considering
the hypothetical difference in results and performance between different ways
of doing things. No this hasn't gotten me too far yet, but it has made me
consider several different things I wasn't thinking about when I originally
posted this message. The rest of the post I will keep for my guidance in
future endeavors. As for this one, your last post has already proven to me
that it is probably not worth the time.

I say this because upon further examination the Excel tables that I have
really aren't as mature or refined as I once thought. The expansion of
different items that are attempting to be tracked and incorporated are just
now being entered and that data is not complete in its entirety. This may
also be another argument for why it should probably move from Excel.
However, I don't want to disappoint you if I send you these tables and you
see that the data management application isn't really all its cracked up to
be. You also made another point about calculating the time for the ROI with
the number of people that will use this application. As it was originally
requested, only a few people would use this application and the output of
this application would be utilized by several more. The data items that it
attempts to manage may grow, but the growth in the number of people that will
be reached by its product will be marginal. So really, I'm now treating this
as a learning exercise and selfishly attempting to develop this so that it
will help me more in the future when it does count (if you will).

So, with all of that ignorance in mind :) if you still want to help and if
this will still discharge a New Years Resolution that would be great. (I'd
love to know what that resolution is exactly by the way, unless of course its
classified :) I am still in the "interview the end users" phase and am
trying to figure out what all will be needed and how it could possibly be
done and what it could possibly grow into. I don't think one table is a good
idea at all and am no longer considering this method. I am a bit concerned
about relationships and how to get to the attributes I'll need, as I need
them, but I think it can be done with the way I'm considering doing it now.
Unless this no longer fits your parameters (which would be no problem and I
would perfectly understand), let me find out a few more things and attempt a
design that I think could work and then I'll send you that with the Excel
tables as a source and then you can review & decide from there.

However it works out, thank you very much for even considering such an offer.

Best Regards
--
http://njgin.aclink.org


Larry Daugherty said:
A far better tool than Access, Excel, pencil & paper, stylus and clay
tablet and a precursor to all of them is the human brain! What I see
in the Original Post disposes me to believe that Access may be a
better solution paradigm than Excel. It is certainly worthy of
investigation and analysis: Just a few clues: there are different
entity types, there are many to many relationships, the data
population is expected to grow....

By the way, I like Excel. Some of my applications with which I'm most
pleased were done in Excel. While they all had data in them they were
really about importing data and calculating and producing new
workbooks, worksheets and charts. As a data management tool, Excel is
an also ran. For very simple data sets and a small data population
with extremely simple or no reporting requirements it's an easy, low
cost way to go.

Staying with a current paradigm on the sole justification that doing
so is *possible* is poor use of that foremost tool. As a race, we've
done a whole bunch more than just come down from the trees. When the
cost of doing business as you now do it exceeds the cost of
transforming to a new paradigm with greater efficiency and utility
then it's time to make a change. Have you noticed that production
(all those jobs) will go where the production cost is lowest for that
particular job? [Warren Buffet is now shifting the focus of his
investing toward the Orient].

The trick is in knowing all of those costs accurately. Determine what
is or should be the time for ROI within your organization. It will be
a rule of thumb rather than an absolute but the bean counters should
know it.

Most organizations are sloppy about monitoring how efficient they are
in their current operations so the metrics are blown at the start.

Try this: Every time you know you're going to do something using your
Excel (or other spreadsheet "database") application, take note of what
you're about to do and then note the time right now. Do the thing.
Record the time you finished with the task. Do that for a week. For
greater accuracy, do it for a longer period of time. Eventually
you'll be able to characterize the time to do something using that
tool. Also, you mentioned that your application will be getting more
use. That means that your monitoring should continue over time.
Maybe pick one day each week to monitor to track the increasing
utilizatization.

Next, determine the Loaded Labor Cost per hour for your organization.
This can be done even for government organizations. Gather the
information to determine all money or goods that are expended per
annum. That means the formal budget, resources consumed from other
organizations but not billed back (the value of real property used
such as buildings, furniture and equipment, parking lots ...) etc.
Divide that huge number by another one: the number of employees times
their paid hours per year.

With the above as the hourly rate, multiply the total hours per day
your current application is used by that rate times the number of
business days per year. That's your annual cost of using that
application to produce the results you get from it. What ever other
things you might have in the fire, doing the above is a very
worthwhile thing to do even if you only track the times used. If your
organization is doing things right then the bean counters can give you
an exact number for the current Loaded Labor Rate. If not, one of
them may even have the super-total cost of operations tucked away
somewhere. They'll either be pleased or shocked that you asked.

Next is to create some informal specifications for the replacement
application. While Access is a natural, I wouldn't even specify
Access unless you want to get your hand in and take over the
maintenance of an eventual Access application. Decide and define all
of the problems you'll solve and outputs required. Describe the
necessary workflows. Describe the necessary Forms and Reports. For
each critical workflow that you have measured in your current
application state a performance threshold. In most cases the
performance thresholds should be significantly higher. Also,
typographical errors should be significantly lower and accidental
damage to data due to mucking in the tables will be a thing of the
past.

Get a consultant to evaluate your prospective project and to work up
an estimate to complete it.

Now you have some things that you can compare. Some of the figures
may be pretty soft but you'll have an idea. If you already have your
standard for ROI then you'll also have an idea on which side of the
go/no-go threshold you stand.

Here's a thought: If you'll send me a populated copy of your Excel
workbook (which I would treat as confidential) and be willing to
collaborate with me via phone and email a few times then I can knock
together an Access demo application using your data. It would be done
in my free time over the next week or two. No charge. No, this is
not an offer of a free real live application. It's just to give you
an idea of the difference in performance and in results between two
ways of doing things.

That would help me discharge a New Year's Resolution.

If you're up for it, let's get started. In any case, the rest of the
post is intended for your guidance. You may know much or all of it
already.

HTH
--
-Larry-
--

Thank you gentlemen. You both are correct. I don't believe that
Access is
Excel on steroids and it has been a long time coming for this
switch. For
me, I'm probably more familiar with Access than I am with Excel. I
work with
relational databases, but I'll always think that I have a lot to
learn. I
realized after I hit 'post' that I didn't explicitly ask the
questions
(although you both provided helpful food for thought). Jeff maybe
you're
right, maybe I should stay in Excel, I'm not sure. However, I think
this
application will grow and I think it could grow nicely from Access.
I do
realize that there will be a design hump (hence this post) and that
this
beginning will later be further normalized and possibly repeated
(the long
way). I have continued reading various posts and I think (although
I know
that It'll need further work as it grows) I'll need to use one big
table and
then query it for report generation. I was trying to resist this
(main
thought for thinking I should stay in Excel) as there will be data
that is
partly repeated. I can't figure how to come up with a design that
will allow
it to be broken into peices without hiding data in the table names
or not
being able to exclude specific items for the average calculation.
Now I
think it will grow after this is accomplished and delivered, so that
each and
every bid will be entered and not just the summary used for
generalization
and estimate. If that were the case I think I could then come up
with a
better design. Until then I guess I was just throwing it to you to
see if I
was not considering one thing or another. Any further thoughts?
--
http://njgin.aclink.org


:

"Hold on there, Pardner..."

I usually agree with Jeff's posts and, in the main, do so here.
However, his (or my) assumption may be ahead of the mule:

By the time anyone seriously considers leaving Excel for Access
they
are usually years late in making that decision. People use Excel
as a
rudimentary data management facility long after the realization
hits
that there have to be better ways. There's also an emotional
attachment to something in which they've invested so much time and
effort over the years. The glue that holds their Excel application
together is the user, who knows the desired outcome of every
operation
and just what inputs must be massaged or provided to get there.
There
is usually no documentation or guidance, other than verbal.
Rarely
has anyone invested in a nice user interface with a menu of
choices
and forms or even a few command buttons.

Depending on your existing knowledge of relational database design
there may be a big hump to get over to make the transition from
Excel
to Access. Understanding normalization is the biggest key there.
Then, there is the Access interface and its universe of
development
tools.

My impression of your data management requirements is that you
have a
perfect candidate for an Access application.

For a professional Access developer, it would be a modest project.
For a neophyte starting from ground zero it could be a longish
project. Most of the neophyte's time and effort will be spend in
learning, applying new knowledge, learning some more, re-doing
earlier
work, etc. The Access learning curve is long and steep.

Jeff's admonition that "Access is not a spreadsheet on steroids"
is
absolutely, positive true. Their greatest similarity is that the
array of cells in a worksheet looks a lot like the datasheet view
of a
table in Access. Even then, they aren't as close as they look.
Also,
don't kid yourself that you'll "make an Access version of your
Excel
workbook". It can't be done. The Excel workbook becomes one
source
document that goes into the mix for creating the specifications
and
subsequent design for the Access application.

HTH
--
-Larry-
--

message Think twice before leaving Excel...

Access is NOT a spreadsheet on steroids. To get the best use of
Access'
relationally-oriented features/functions, you can't feed it
'sheet
data.

You will need to learn about and use relational database design
and
normalization. If these terms are new, you have a fairly steep
learning
curve before you.

Is there a reason you wish to turn away from something you
already
know?

Is there a reason (other than "I think...") for selecting
Access?

Good luck!

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/


message
I want to go from an Excel workbook to an Access DB
application so
that I
can
add additional data and project items without compromising the
formulas
that
calculate the averages. I think if I persisted with Excel I
could
probably
get it to work, but I think it would be better to go Access.
The
purpose
of
this DB will be:

"To develop average costs for project items and use these
prices
to
develop
a reasonable overall average project cost. The average costs
for
project
items would be generated from criteria consisting of specified
projects
and
date ranges."

The average costs will ultimately be in the form of a report
and
will
consist of something like

Average Low Bid 2004|Average All Bidders 2004|Average Low Bid
2005|Average
All Bidders 2005|Average Low Bid 2006|Average All Bidders
2006|Average Low
Bid 2005-2006|Average All Bidders 2005-2006|Average Low Bid
2003-2006|Average All Bidders 2003-2006
 
L

Larry Daugherty

Jeff,

In the context in which I used the term, "Iterative development" is
development that is almost always ongoing and implies a lack of
discipline. The term "hacker" originally was used to describe people
who developed without a plan. If the development is ad-hoc and going
on at a high frequency such as daily, weekly, even monthly then it's
fair to characterize the practice as iterative. If the development is
done in pre-defined phases at a very low frequency then I would not
characterize it as "iterative". One pleasant period in my life was as
part of a development team providing the software for a dedicated work
processor. Our development schedule was roughly two releases each
calendar year. One release was for product enhancements. It was
always a large effort. The other release was for maintenance - bug
fixes and modest enhancements to the functionality. It was always a
smaller effort. In each case, we started the project by thrashing out
the project goals before we began designing and writing code. Source
code was rigorously controlled. Once the developer had completed
individual development and test, her or his code was incorporated into
the working base level of code. Our QA department vigorously tested
the software continuously. Discovered bugs were reported, evaluated,
and assigned a severity level. Before code freeze, any bug could be
addressed. After code freeze only the most severe bugs were
addressed. Once the code was frozen it was not checked out again
until the next project. At a high enough level that process can be
called iterative. The difference was that the whole process was
planned and scheduled.

Just about any application that solves real problems and gets out into
the world will require future enhancements and possibly bug fixes.
That's great, that's what we do. However, if each and every perceived
need is addressed at the moment of its recognition to the tune of a
complete development cycle for that one issue, someone has an
uninteresting career filled with busy work.

HTH
--
-Larry-
--

Jeff Boyce said:
Larry

I'm with you most of the way, except for that part about Access not lending
itself to iterative development.

Maybe it has just been me and my personal approach, but iterative
development (using Access) is what I've been doing since '94, when Access
first came out.

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/

If we do continue this dialogue beyond this post we should take it off
line. You can easily decode my email address. I don't have yours.

I inferred from your first post that you were already using an Excel
spreadsheet to track project data and were abstracting information
from it. I assumed several tens of projects with all of the
constituent data.

If you don't have a base line for comparison and the pre-existing sets
of data then a demo for comparison purposes wouldn't have much impact.

As you start from ground zero, concepts are simple and Excel is the
evident low cost, low effort solution. There is no real design of the
information at the outset and you need only be sure that everything
you want to track is actually entered into the spreadsheet.
Organization would be on a "row per project" basis or a "worksheet per
project" basis. In the beginning, life is simple. As you begin to
relate different parts of the data to other parts then some thinking
is involved and some rudimentary design is employed. Some formulas
are embedded. Position often determines data significance. Usually
the sophistication of the solution increases with time and use.
"Programming" is usually ad-hoc. Needs aren't anticipated but are met
when they're encountered. The designer of the tool is usually it's
primary user so the tool relies on that knowledgeable user being the
user. There is no documentation. There is nothing sophisticated
about the tool whatever. [Note that Excel can be programmed to
provide a very sophisticated user interface and tome very powerful
results - it usually isn't done for informal projects], Iterative
development is the rule rather than the exception: often adding new
functionality requires re-working existing functionality. Excel is
intended to be a user level tool and ad hoc programming is often the
rule.

Getting going with Access requires a lot more thought and design
beforehand. It isn't rocket science it's just more difficult to learn
in the first place. People who regularly design Access applications
get into the habit of doing a bit more thinking at the front end of a
project. Also, experience causes Access developers to do some things
as 2nd nature. While Access can be developed iteratively, it isn't a
handy thing to do. Access is intended to be primarily a platform for
developers to develop applications which users will then use. It is
not really a good thing to be continually developing.

The crucial thing at the moment is whether you are actually using the
Excel tool to gather, track and, possibly, present information. I
didn't expect that there would be a sophisticated data management
application. Quite the contrary. Don't worry that I would fault what
you have. Things like that start out as just a handy means of
recording interesting data and then just grow. They are usually the
proactive creation of someone who needs some information on a regular
basis that no one else provides.

I don't really know what to tell you regarding going forward with a
demo. You're kind of hedging. My offer is based on there being a
fairly well populated Excel spreadsheet with real project data. In
addition I would need such specifications as might exist or could be
generated by you. In addition I would need to be able to talk to you
on the phone once or twice. Remember, I want to just generate a demo
Access application that would show the difference in performing
representative operations in Access. The demo isn't to be a free
application. That's contrary to the charter for these newsgroups.
Besides, that's what I do professionally and for which I charge money.
The demo will gracefully expire after you've had a change to run it
for a month or two.

I also don't want a "free coaching" job. I want to approach this thin
in a friendly but businesslike manner and give you and your colleagues
a utility to compare to the Excel tool.

If you want to go ahead with it then lets do so. Respond with your
workbook attached to an email that includes your name, your phone
number(s) and the best range of times to call. I'm in Southern
California and keep strange schedules.

What ever the case, do continue to interview your prospective users in
both a Q&A mode and request them to describe what they'd like to get
from an application like you're considering. You might raise the idea
of Excel and "other" platforms.

My resolution? At the first opportunity in 2008 that I recognize to
offer a service that I believe would be useful to the OP in one of the
Access newsgroups, do so. The offer and service have to be consistent
with newsgroup values. The offer is not a blank check for service but
exactly as I make the offer.

Believe me, I have received so much over the years from contributors
in these newsgroups and from their private web sites, especially the
MVPs.

Best regards,
--
-Larry-
--

Larry you're a gem and I thank you very much. This project is
also
as a
favor in my spare time and although it's for a sister department
at
work,
it's free of charge for me as well.

Although I don't have actual results for comparison, I've been considering
the hypothetical difference in results and performance between different ways
of doing things. No this hasn't gotten me too far yet, but it
has
made me
consider several different things I wasn't thinking about when I originally
posted this message. The rest of the post I will keep for my guidance in
future endeavors. As for this one, your last post has already proven to me
that it is probably not worth the time.

I say this because upon further examination the Excel tables
that I
have
really aren't as mature or refined as I once thought. The
expansion
of
different items that are attempting to be tracked and
incorporated
are just
now being entered and that data is not complete in its entirety. This may
also be another argument for why it should probably move from Excel.
However, I don't want to disappoint you if I send you these
tables
and you
see that the data management application isn't really all its cracked up to
be. You also made another point about calculating the time for
the
ROI with
the number of people that will use this application. As it was originally
requested, only a few people would use this application and the output of
this application would be utilized by several more. The data
items
that it
attempts to manage may grow, but the growth in the number of
people
that will
be reached by its product will be marginal. So really, I'm now treating this
as a learning exercise and selfishly attempting to develop this
so
that it
will help me more in the future when it does count (if you will).

So, with all of that ignorance in mind :) if you still want to
help
and if
this will still discharge a New Years Resolution that would be great. (I'd
love to know what that resolution is exactly by the way, unless
of
course its
classified :) I am still in the "interview the end users" phase
and
am
trying to figure out what all will be needed and how it could possibly be
done and what it could possibly grow into. I don't think one
table
is a good
idea at all and am no longer considering this method. I am a
bit
concerned
about relationships and how to get to the attributes I'll need,
as I
need
them, but I think it can be done with the way I'm considering
doing
it now.
Unless this no longer fits your parameters (which would be no problem and I
would perfectly understand), let me find out a few more things
and
attempt a
design that I think could work and then I'll send you that with
the
Excel
tables as a source and then you can review & decide from there.

However it works out, thank you very much for even considering
such
an offer.
Best Regards
--
http://njgin.aclink.org


:

A far better tool than Access, Excel, pencil & paper, stylus
and
clay
tablet and a precursor to all of them is the human brain!
What I
see
in the Original Post disposes me to believe that Access may be a
better solution paradigm than Excel. It is certainly worthy of
investigation and analysis: Just a few clues: there are different
entity types, there are many to many relationships, the data
population is expected to grow....

By the way, I like Excel. Some of my applications with which
I'm
most
pleased were done in Excel. While they all had data in them
they
were
really about importing data and calculating and producing new
workbooks, worksheets and charts. As a data management tool, Excel is
an also ran. For very simple data sets and a small data population
with extremely simple or no reporting requirements it's an
easy,
low
cost way to go.

Staying with a current paradigm on the sole justification that doing
so is *possible* is poor use of that foremost tool. As a
race,
we've
done a whole bunch more than just come down from the trees.
When
the
cost of doing business as you now do it exceeds the cost of
transforming to a new paradigm with greater efficiency and utility
then it's time to make a change. Have you noticed that production
(all those jobs) will go where the production cost is lowest
for
that
particular job? [Warren Buffet is now shifting the focus of his
investing toward the Orient].

The trick is in knowing all of those costs accurately.
Determine
what
is or should be the time for ROI within your organization. It will be
a rule of thumb rather than an absolute but the bean counters should
know it.

Most organizations are sloppy about monitoring how efficient
they
are
in their current operations so the metrics are blown at the start.

Try this: Every time you know you're going to do something
using
your
Excel (or other spreadsheet "database") application, take note
of
what
you're about to do and then note the time right now. Do the thing.
Record the time you finished with the task. Do that for a
week.
For
greater accuracy, do it for a longer period of time. Eventually
you'll be able to characterize the time to do something using that
tool. Also, you mentioned that your application will be
getting
more
use. That means that your monitoring should continue over time.
Maybe pick one day each week to monitor to track the increasing
utilizatization.

Next, determine the Loaded Labor Cost per hour for your organization.
This can be done even for government organizations. Gather the
information to determine all money or goods that are expended per
annum. That means the formal budget, resources consumed from other
organizations but not billed back (the value of real property used
such as buildings, furniture and equipment, parking lots ...) etc.
Divide that huge number by another one: the number of
employees
times
their paid hours per year.

With the above as the hourly rate, multiply the total hours
per
day
your current application is used by that rate times the number of
business days per year. That's your annual cost of using that
application to produce the results you get from it. What ever other
things you might have in the fire, doing the above is a very
worthwhile thing to do even if you only track the times used.
If
your
organization is doing things right then the bean counters can
give
you
an exact number for the current Loaded Labor Rate. If not, one of
them may even have the super-total cost of operations tucked away
somewhere. They'll either be pleased or shocked that you asked.

Next is to create some informal specifications for the replacement
application. While Access is a natural, I wouldn't even specify
Access unless you want to get your hand in and take over the
maintenance of an eventual Access application. Decide and
define
all
of the problems you'll solve and outputs required. Describe the
necessary workflows. Describe the necessary Forms and
Reports.
For
each critical workflow that you have measured in your current
application state a performance threshold. In most cases the
performance thresholds should be significantly higher. Also,
typographical errors should be significantly lower and accidental
damage to data due to mucking in the tables will be a thing of the
past.

Get a consultant to evaluate your prospective project and to
work
up
an estimate to complete it.

Now you have some things that you can compare. Some of the figures
may be pretty soft but you'll have an idea. If you already
have
your
standard for ROI then you'll also have an idea on which side
of
the
go/no-go threshold you stand.

Here's a thought: If you'll send me a populated copy of your Excel
workbook (which I would treat as confidential) and be willing to
collaborate with me via phone and email a few times then I can knock
together an Access demo application using your data. It would
be
done
in my free time over the next week or two. No charge. No,
this
is
not an offer of a free real live application. It's just to
give
you
an idea of the difference in performance and in results
between
two
ways of doing things.

That would help me discharge a New Year's Resolution.

If you're up for it, let's get started. In any case, the rest
of
the
post is intended for your guidance. You may know much or all
of
it
already.

HTH
--
-Larry-
--

Thank you gentlemen. You both are correct. I don't believe that
Access is
Excel on steroids and it has been a long time coming for this
switch. For
me, I'm probably more familiar with Access than I am with
Excel.
I
work with
relational databases, but I'll always think that I have a lot to
learn. I
realized after I hit 'post' that I didn't explicitly ask the
questions
(although you both provided helpful food for thought). Jeff maybe
you're
right, maybe I should stay in Excel, I'm not sure. However,
I
think
this
application will grow and I think it could grow nicely from Access.
I do
realize that there will be a design hump (hence this post)
and
that
this
beginning will later be further normalized and possibly repeated
(the long
way). I have continued reading various posts and I think (although
I know
that It'll need further work as it grows) I'll need to use
one
big
table and
then query it for report generation. I was trying to resist this
(main
thought for thinking I should stay in Excel) as there will
be
data
that is
partly repeated. I can't figure how to come up with a
design
that
will allow
it to be broken into peices without hiding data in the table names
or not
being able to exclude specific items for the average calculation.
Now I
think it will grow after this is accomplished and delivered,
so
that
each and
every bid will be entered and not just the summary used for
generalization
and estimate. If that were the case I think I could then
come
up
with a
better design. Until then I guess I was just throwing it to
you
to
see if I
was not considering one thing or another. Any further thoughts?
--
http://njgin.aclink.org


:

"Hold on there, Pardner..."

I usually agree with Jeff's posts and, in the main, do so here.
However, his (or my) assumption may be ahead of the mule:

By the time anyone seriously considers leaving Excel for Access
they
are usually years late in making that decision. People
use
Excel
as a
rudimentary data management facility long after the realization
hits
that there have to be better ways. There's also an emotional
attachment to something in which they've invested so much
time
and
effort over the years. The glue that holds their Excel application
together is the user, who knows the desired outcome of every
operation
and just what inputs must be massaged or provided to get there.
There
is usually no documentation or guidance, other than verbal.
Rarely
has anyone invested in a nice user interface with a menu of
choices
and forms or even a few command buttons.

Depending on your existing knowledge of relational
database
design
there may be a big hump to get over to make the transition from
Excel
to Access. Understanding normalization is the biggest key there.
Then, there is the Access interface and its universe of
development
tools.

My impression of your data management requirements is that you
have a
perfect candidate for an Access application.

For a professional Access developer, it would be a modest project.
For a neophyte starting from ground zero it could be a longish
project. Most of the neophyte's time and effort will be
spend
in
learning, applying new knowledge, learning some more, re-doing
earlier
work, etc. The Access learning curve is long and steep.

Jeff's admonition that "Access is not a spreadsheet on steroids"
is
absolutely, positive true. Their greatest similarity is
that
the
array of cells in a worksheet looks a lot like the
datasheet
view
of a
table in Access. Even then, they aren't as close as they look.
Also,
don't kid yourself that you'll "make an Access version of your
Excel
workbook". It can't be done. The Excel workbook becomes one
source
document that goes into the mix for creating the specifications
and
subsequent design for the Access application.

HTH
--
-Larry-
--

message Think twice before leaving Excel...

Access is NOT a spreadsheet on steroids. To get the
best
use of
Access'
relationally-oriented features/functions, you can't feed it
'sheet
data.

You will need to learn about and use relational database design
and
normalization. If these terms are new, you have a
fairly
steep
learning
curve before you.

Is there a reason you wish to turn away from something you
already
know?

Is there a reason (other than "I think...") for selecting
Access?

Good luck!

--
Regards

Jeff Boyce
www.InformationFutures.net

Microsoft Office/Access MVP


Microsoft IT Academy Program Mentor
http://microsoftitacademy.com/


"scratchtrax" <[email protected]>
wrote
in
message
I want to go from an Excel workbook to an Access DB
application so
that I
can
add additional data and project items without
compromising
the
formulas
that
calculate the averages. I think if I persisted with
Excel
I
could
probably
get it to work, but I think it would be better to go Access.
The
purpose
of
this DB will be:

"To develop average costs for project items and use these
prices
to
develop
a reasonable overall average project cost. The
average
costs
for
project
items would be generated from criteria consisting of specified
projects
and
date ranges."

The average costs will ultimately be in the form of a report
and
will
consist of something like

Average Low Bid 2004|Average All Bidders 2004|Average
Low
Bid
2005|Average
All Bidders 2005|Average Low Bid 2006|Average All Bidders
2006|Average Low
Bid 2005-2006|Average All Bidders 2005-2006|Average
Low
Bid
2003-2006|Average All Bidders 2003-2006
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Similar Threads


Top