Dashboards are a key piece of managing your business. When we implemented dashboards at one client, we saw profit margins increase by ten points, because management became focused on the most important economic levers in the company.
Most people are familiar with executive dashboard best practices: link them to your strategy, have them cascade down the organization, have a mix of forward-looking variables (like leads in the sales pipeline) and backward-looking ones (like booked revenue).
This all sounds good, but the real learning comes from trying a few things, even if it means making some mistakes. To save you time, here are three lessons I’ve picked up over the years when implementing and using dashboards.
1.Seven Plus or Minus Two Measures, Not Twelve Pages
Robert Simons, HBS Professor and Author of Seven Strategy Questions, told me that in his executive education course on performance metrics, he asks all leaders to bring with them a copy of their existing dashboard. He awards a prize to the person who has the most metrics. It is, however, a consolation prize, because this executive is actually the loser. Simons advocates having only five to nine metrics on your dashboard.
At one company, we had a 12-page dashboard (I definitely would have won Professor Simons’ prize). You can imagine how difficult it was to detect issues.
There is a temptation to make sure every possible metric is on the dashboard so absolutely nothing can go wrong. The problem is when there are too many things to look at, nothing gets looked at with the right level of scrutiny, and it’s tough to get a “feel” for the numbers.
To narrow the list, follow some more of Simons’ advice: focus on failure. What five to nine things could sink your business if they went wrong?
2.It’s A Smoke Detector, Not a Problem Solver
One reason we ended up with 12 pages of metrics is we were trying too hard to make the dashboard more than a dashboard. The purpose of the dashboard should be to foster a disciplined review and understanding of the most important numbers in your business. When one of the numbers is off, that’s when you need to turn to analysis to diagnose the root cause. When you try to add all possible root causes to your dashboard, it gets too unwieldy. Get alignment with your team that the dashboard should function as a smoke detector – it alerts you when something is starting to smolder. But it doesn’t end there. When something in the numbers raises the alarm, then you need to do more analysis to find the root cause and develop a solution.
For example, for one client we noticed traffic was way down. We did a deep-dive analysis: we looked at sources of traffic (from search, email, social, etc.), time of day, and the device (desktop, mobile, tablet). It became immediately clear what the problem was – traffic was way down on mobile. Clearly our user experience was not satisfactory on the phone, so no one was engaging with that product. We launched an initiative to improve the mobile interface, and were able to get mobile traffic, and thus overall traffic, up to the target again.
3.Minimize Manual Work
To anyone who has helped me implement a dashboard project before: thanks for all your effort and sorry for making it so time consuming! I realized I put too much weight on getting the “ideal” metrics that perfectly matched our strategy and cascaded nicely across the organization, and not enough weight on implementation. As a result we put a lot more effort into creating the dashboard than using the dashboard.
Things got really complicated for my team once when we built a dashboard to support a new segmentation strategy. I wanted to look at performance by custom channel, by custom segment. This meant manual manipulation of the data. As a result, it took us the equivalent of two full-time employees just to create the dashboard each week.
Will it take you one full day each week to piece together different sources of information to calculate the number of new customers by your custom-defined sources (e.g., Facebook ads, house ads, Twitter ads, email campaigns, Pinterest posts, etc.)? Why not just use the sources that are pre-defined in Google Analytics to save you time (no time versus one full day per week), and perform deep-dive analyses when you notice there is an issue in one of the channels?
So What’s the Answer?
I have found the following are the best dimensions to monitor for a digital media membership (subscription) business:
- Number of active members
- Email capture rate (emails captured per new unique visitor)
- Email churn rate (percentage of email file that unsubscribed)
- Conversion rate by source (percentage of sign ups divided by unique visitors from that source). Hint: use the Google Analytics sources – direct, organic search, social, referral, email, paid search
- Renewal rate
- Average price
I’d look at each of these dimensions three ways: current period (month or week depending on the velocity of your business), comparative period (same period last year if you are an established company or prior period of you are in high growth mode), target (yes, you must have a target for each of these).
You can probably think of dozens of metrics that are missing – and maybe there should be a few substitutions for your business. If you have had success with five to nine metrics, please share in the blog comments or email me your success story!