How people screw up on their product demos

ryanleask.wordpress.com

29 points·rmason··6 comments

Good list. To complement this article, I highly recommend "Demoing Software for Fun & Profit" by Dave Winer. There are pretty good tips in it.

http://scripting.com/davenet/1995/01/04/demoingsoftwareforfu...

I disagree with #9. Maybe Bill was on vacation, broke his arm and stayed in hospital for a full month. Of course he's underperforming.

The data is there for you to query. There should be pointers to begin with, but the software should never make business decisions for you.

If the business analyst needs an hour to find the data, then the interface is too complicated, or the analyst did not learn how to use the software in the first place.

Please don't make me write software that can replace your business analyst.

I was thinking along the same lines. What he is asking for is quite close to general AI!

There are many ways to get around this issue. Alerts/notifications based on thresholds being exceeded is one. Another is to start with a dashboard that highlights the major issues for you to look at. Another is to have a marker/icon next to any figure that needs your attention (ala Stephen Few style).

There is nothing wrong with investigative analytics. I'm not saying slice and dice is a bad thing - just there are much better ways to open your BI demo.

Having spent some years in business intelligence, point #9 especially hits home.

Just about everyone I know who has delivered a canned Cognos demo has done that exact drill down example.

Bottomline: Developers don't use their own product in the same context that the end user will. Which means the end user won't get the right picture if the developer does the demo.

Reply on news.ycombinator.com