Re: [ba-ohs-talk] What are we trying to accomplish?
At 15:19 -0700 2002/05/24, Eric Armstrong wrote:
>Johannes Ernst wrote:
>
>> I would like to add: the **benefits** of such a system. In dollars
>> saved per person per day, preferably. E.g. it will increase the
>> productivity of a user of class A by 5% (?) when performing task B.
>
>I'm curious. Have such numbers ever been adduced for word
>processors or spreadsheets? Databases? (01)
You bet they have. Every marketer worth their salt (fairly few are,
I'm afraid, but nevertheless), which means whose product sells, has
numbers like that for all sorts of different situations. Some you may
disagree with, etc. but they are there. Every single enterprise
software deal has an ROI calculation associated with it, and you
can't calculate ROI without having the numbers above. (02)
In case of when spreadsheets were new, this was simple. The following
is made up, but you get the essence of it:
user: department manager
application (one of several): monthly project budget updates
today: use mainframe application provided by MIS department
proposal for tomorrow: use newly bought "PC" with Visicalc
today's situation: formulas behind budget change approx. once every 3
months. Total effort to make one change: 3 days (aggregate work of
MIS people and department manager, including training for new
software)
proposed situation: department manager edits spreadsheet formulas
directly. After a training period, he can change the fomula in half a
day.
Savings every 3 months: 2.5 days * $1000 per day = $2500.
Purchase cost hardware, plus software: $2500.
Training required: $5000.
ROI: 9 months. Ergo, no-brainer. (03)
>How does one go about putting a dollar value on such things? (04)
Note the specific role, and specific activity, otherwise the
calculation does not work. In the context of the OHS, that's the gut
of the problem I'd think: we don't know. Assigning the numbers is
easy after that is done. So that's where the work needs to start I
would think (with my marketing / product management hat on). (05)