Re: Rules of thumb in estimating

[Bill DuBay <bill_dubay -at- PHOENIX -dot- COM>]

This is a problem not just for contractors, but for everybody. It is very
difficult if you

1. Don't know the technology
2. Don't know how much in the way of engineering documents is available, how
much is already on paper.
3. Don't know about the other resources required for the task, engineers'
time, available tools, testing, editing,  etc.

Once you know those things, and you can come up with the number of pages for
the document, you can start estimating on the basis of days required per
page. I often spend two days per page. This includes research, writing,
indexing, illustrations, formatting, reviews, etc.

The biggest problem I face is the scheduling of engineer's time. We like to
produce documents in tandem with the product, but prior to release, there is
no engineering time scheduled for documentation. That means I have to wait
for the code to freeze before I can finish the job. The engineer's time is
required for technical edits.


Bill DuBay
Technical Writer
Phoenix Technologies Ltd.
email: bill_dubay -at- phoenix -dot- com
(714)790-2049 FAX: (714)790-2001
http://www.phoenix.com
-------------
Original Text
From: Kathy Marshall <kmarshall -at- MODACAD -dot- COM>, on 7/8/97 7:47 AM:
Matt,
I'm interested in hearing how everyone estimates their doc projects too.
I remember hearing or reading somewhere that the industry standard is
something like 2 pages per hour. I think that's what it was.  I used to
use that as a base and would then work around it -- projects that
required more research and were start-from-scratch things, I allowed
more time. Projects that were rewrites, I estimated less time (maybe 4-6
pages per hour). Then it got all complicated because you had to factor
in reviews, rewriting, etc.

Nowadays, I just eyeball it and I'm usually right. It goes something
like this:
Boss: So how long will this take to complete?
Me: 4 months, start to finish.
Boss: 4 months?! But it needs to be done in 2 months. That's when the
project is scheduled for release.
Me: Ok, let me rework the schedule...(a few hours later)...Ok, I think
we can do it in 2 months if we restructure a few things.
Boss: Great.

2 months later...the software is NOT released, nor has anyone reviewed
the manually thoroughly (which was completed on time). By the time the
doc has finallly been reviewed and revised, 4 months have passed.

-Kathy



> ----------
> From:         Matthew J Long[SMTP:mjl100z -at- MAIL -dot- ODU -dot- EDU]
> Reply To:     Matthew J Long
> Sent:         Tuesday, July 08, 1997 5:44 AM
> To:   TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU
> Subject:      Rules of thumb in estimating
>
> On Mon, 7 Jul 1997, Cramer, Kim wrote:
>
> > This is how it works here:
> >
> > 1. We estimate time to completion for each book for each release,
> then
> > we track actual time.  We're expected to estimate within +/- 10% of
> > actual.  This comparison helps us become better at estimating for
> future
> > projects so they are more likely to stay on track for the estimated
> > release date.
> >
>
> Kim,
>
> I landed a great job as a technical writer only 6 months out of
> college in
> which I am the only (and therefore Lead) tech writer. I call all the
> shots
> as far as design, usage, conventions, etc. I learned a lot in my
> internship and with the work I have done so far (I have now been doing
> this for 9 months) and my company likes my work and seems confident in
> my
> skills. However, I start to get concerned about the amount of time it
> should take to produce a user's manual. In my internship, things were
> very
> laid back--no deadlines. Well, the real world just ain't like that. I
> haven't had any major deadlines so far, but I think that is just
> because
> my bosses have no idea how long a user's manual should take.
>
> When I read your comment (above), I thought it would be a good idea to
> ask
> you, but also anyone else, how to estimate the amount of time it
> should
> take to produce a manual. I am aware that the answer is going to
> depend
> on the application and how many "features" is sports. But I would love
> to
> know how you get started. When I used to paint houses during my
> summers
> in college, the only way I could be accurate in my estimates to the
> customer was through trial and error, Sometimes I would underestimate
> and
> lose my butt, while other times it was the opposite. I would bet that
> the
> same is true in estimating the time it takes to complete a manual, but
> I
> would also bet that there are a  few rules of thumb. Can you advise?
>
> I welcome any comments!!
>
> Thanks!
>
>
> ////////////////////////////**************************************
>         Matthew J. Long
>         Technical Writer
>         mjl100z -at- mail -dot- odu -dot- edu
>         matt -dot- long -at- justice -dot- usdoj -dot- gov
>
>         -When you can't be eloquent, choose brevity!
> ********************************************************////////////
>
> ~~
>  TECHWR-L (Technical Communication) List Information: To send a
> message
> to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot-  Send
> commands
>    to LISTSERV -at- LISTSERV -dot- OKSTATE -dot- EDU (e.g. HELP or SIGNOFF TECHWR-L).
>   Search the archives at http://www.documentation.com/  or search and
> browse the archives at
> http://listserv.okstate.edu/archives/techwr-l.html
>  Send list questions or problems to the listowner at
>

 TECHWR-L (Technical Communication) List Information: To send a message
to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot-  Send commands
   to LISTSERV -at- LISTSERV -dot- OKSTATE -dot- EDU (e.g. HELP or SIGNOFF TECHWR-L).
  Search the archives at http://www.documentation.com/  or search and
browse the archives at http://listserv.okstate.edu/archives/techwr-l.html

 TECHWR-L (Technical Communication) List Information: To send a message
to 2500+ readers, e-mail to TECHWR-L -at- LISTSERV -dot- OKSTATE -dot- EDU -dot-  Send commands
   to LISTSERV -at- LISTSERV -dot- OKSTATE -dot- EDU (e.g. HELP or SIGNOFF TECHWR-L).
  Search the archives at http://www.documentation.com/  or search and
browse the archives at http://listserv.okstate.edu/archives/techwr-l.html