Gotta be perfect

It is amazing when small things can cause problems.   Major major problems.  

Sure it is true in development of software.  But it is shocking how easy a miss-communication of something can cause tons of hard selling work to be derailed.  Quality of communication needs to be consistant and well structured all the time.  And you have to be perfect at determining who to talk to and when.  You have to consistantly make sure the appropriate people are bought in at the exact right time.  Gotta be perfect. 

Move the Deer Crossing

Happy to take product suggestions from everyone.   

Lucky

Just a quick comment. I continue to be lucky to work with the people I work with. They are smart and have drive to do well for our customers. The work is challenging and hard, and fortunately it is rarely repetitive. But it makes all the difference to be constantly challenged by those around you to do better.

Pie Charts - Great for Hiding Information

Part of the job of product managers (or anyone in business for that matter) is to communicate data.  "How good is this thing?"  Often, people use pie charts to communicate this data in dashboards or presentations. Why do people do this?  

Part of the job of product managers (or anyone in business for that matter) is to communicate data.  "How good is this thing?"  Often, people use pie charts to communicate this data in dashboards or presentations. Why do people do this?  

As Stephen Few (guru on dashboard design) says:

The truth is, I never recommend the use of pie charts.  ...  Pie charts don't display quantitative data very effectively.  Humans can't compare two-dimensional areas or angles very accurately.

And he's right.  This is an example I pulled from one of our recent internal dashboards.  The problem with this pie chart is that you can't really tell, without reading the numbers the relative size of the pies.  Is the 9.6% slice smaller than the 10.3%?  Is the 22.7% slice double the others or 3x the others?  

You should be able to look at a graph and QUICKLY see the relative comparison without having to read the numbers.  

A better of use of this would be a bar or column chart.  Without the % called out, I can much more quickly see relative size in the bar chart example below.  

Now, I actually disagree with Stephen Few in one respect.  There are cases where you SHOULD use pie charts.  You should use pie charts to hide information.  If you don't want your user to immediately know the differences in values because you want to minimize that particular data point then by all means, use pie charts.  And I'm being honest when there are cases where this can be useful. You may really want your audience to focus and understand one data point, and not pay attention to this other one (if the "other one" is required to be there).  In which case, use a pie chart for the "other one".  This will help make sure they focus on the metric you really want them to be looking at.  Of course, it is best if you can not include the "other one" metric in the first place.  But that isn't always feasible.  

Now, I actually disagree with Stephen Few in one respect.  There are cases where you SHOULD use pie charts.  You should use pie charts to hide information.  If you don't want your user to immediately know the differences in values because you want to minimize that particular data point then by all means, use pie charts.  And I'm being honest when there are cases where this can be useful. You may really want your audience to focus and understand one data point, and not pay attention to this other one (if the "other one" is required to be there).  In which case, use a pie chart for the "other one".  This will help make sure they focus on the metric you really want them to be looking at.  Of course, it is best if you can not include the "other one" metric in the first place.  But that isn't always feasible.  

I know that the topic of use of "Pie Charts" isn't world changing.  Really, this post is about communication.   The small stuff does matter in communication.  Is showing the value of your product important?  YES!  Get it right.  Spend time on it.  

A/B Testing Post from 37Signals

I found this post from 37 Signals to be an interesting viewpoint on testing.  It is about the process, not about the outcome.  

I only judge a test based on whether we designed and administered it properly.

As an industry, we don’t yet have a complete analytical model of how people make decisions, so we can’t know in advance what variations will work. This means that there’s no shame in running variations that don’t improve conversion. We also lack any real ability to understand why a variation may have succeeded, so I don’t care much whether or not we understood the results at a deeper level.

The only thing we can fully control is how we set up the experiment.....

See more at...  http://37signals.com/svn/posts/3169-ab-testing-its-not-about-the-results-and-its-definitely-not-about-the-why