Great Ed Tufte Posts

Anyone interested in visual data analysis should stop by the Ask ET forum. Here are some great posts:

Election coverage
Megan Jaegerman’s News Graphics
Data Visualization Ramifications of the Retina Display
Sports Graphics

evidensity.js now on github

I’m open sourcing my javascript sparkline generator on github, mainly because I’m using it on several projects and this will help me keep things consistent.

New Version

Went live with a lot of new features tonight. I’ve been working on the sales pipeline and have been running into a lot of snags. Making a prediction of future sales based on history has turned out to be even more challenging than I anticipated. In the mean time, here’s a brief list of some of the changes.

  • Added tool tips to explain each item
  • Swapped colors for 3 month moving average and monthly total in sparklines
  • Added average monthly won and deal count to “How are we Doing Lately?”
  • Eliminated current month from average calculations
  • Added scatter plots for deal history.

Animated Scatter Plot

It’s not ready, but it’s quickly getting awesome.

Check this out:

Old Deals

I got a lot finished recently on the upcoming sales pipeline, but there’s still some figuring out to do. In the mean time I pushed a couple more features live including “What are my Oldest Deals?” and some other info on average closing times by customer. This information is essential for building a quantified sales pipeline.

Standard Deviation

Sometimes your brain gets a little fried while coding and I didn’t realize the answer to last night’s scaling problem was obvious has been solved since 1894. Looks like two Standard Deviations were enough to show detail in the scatter plot, but still show most of the values. I’m still scaling both the won and lost plots to whichever comes out to be larger, but it’s a big improvement. I’ve pushed those changes and a few more subtle ones out for you to enjoy.

scatter plot

Looks better doesn't it?

The solution made me think back to my statistics class in college, taught by a professor who refused to use a calculator. This would be impressive if he hadn’t constantly made mistakes. I asked him once why he didn’t use one and he said, “When I was a boy growing up in Palestine we only had one telephone in the village. Whenever it rang, everyone in town would run to it to see what was so important.”


Scatterplots and Deal Lengths

Now that I’ve launched the site and got all the boring stuff (API calls, authentication, billing, etc.) out of the way I can now again focus on the fun stuff.  I wrote a scatterplot function tonight that will help show how long the average deal is, a key piece of data for calculating a sales pipeline.  Here’s a quick taste of how it looks.

Scatter Plot

Deal Length vs Value

The only thing keeping me from pushing this live tonight is the scaling. If I use the max value, one outlying deal can throw the whole thing off. I’m going to make it scale to incorporate at least 70-90% of the data points.

I also added an average monthly Won value and average monthly new deals generated under “How are we Doing Lately.”

Deal Associations

Thanks everyone who has signed up for Evidensity and given me some great feedback.  One thing I’ve fixed is an issue where if you associate Highrise deals with people instead of companies, their names now show up in the reports.

Evidensity is now a Highrise Dashboard

“One of the fundamental cognitive tasks in analytical thinking is to reason about causality. Thus one of the fundamental principles of analytical design is to show causality.” – Ed Tufte, Association or Causation?

HighriseHighrise, the great customer resource management (CRM) tool by 37signals has a lot going for it (great user interface, fast performance, low price), but it’s reporting tools are non-existent. So I’ve built one.

The goal of analytical design is to prove causation, answer questions, and ultimately predict the future with enough time to alter your behavior. Evidensity takes this approach by asking a question and then trying to answer it  For example:

Who is my best customer?

Is it the customer who gave us the most money?  Is the customer who we like the best?  Evidensity answers this using the Highrise API and figuring out the amount of money they’ve spent on you and pitting that against the amount of effort you’ve spent working on their proposals. In that way, we define the best customer as one who pays, values your time and doesn’t interfere with satisfying the rest of your customers.  Flipping that on it’s head, do you want to know…

Who is my worst customer?

Evidensity is a quick way to make a quantitative analysis of where to spend your proposal resources.  In addition to these questions, it also answers:

How are we doing lately?
What are the top pending deals?
Who is my top sales person?
Who do we deal with most? 

Try Evidensity for free for 7 days.  After that it’s $29.99/month.

Why would you sign up for a brand new service like this?  There are other analytical tools out there like eazyBI and others.  As an early adopter you’ll have huge influence on Evidensity’s development.

My World View

  • Some people don’t want customizable line graphs
  • They want actionable intelligence about their data.
  • Sales pipelines are built on faulty assumptions and overly optimistic sales people
  • They should be built on historical data.
  • Your eyes and brain can handle it, so fit tons of data into one screen.

If this sounds like something you can get behind, I recommend signing up and trying it out for a week.  All I ask for is some feedback.  Ever since I became the top question answerer at 37signals, I’ve been getting a lot of requests to help with implementation and training for companies.  I’d rather build things for a slightly wider audience.

Money: A Chart of Almost All of It

xkcd may have made the best infographic of the year today.

Money: A Chart of Almost All of It, where it is and what it can do.


In other news, I’ve been working on a super secret project within this space. I hope to launch soon, but it’s been taking me longer than I expected. Here’s a taste: