Why limit yourself to standard KPIs? Be agile, create your own KPIs, change them at a moment’s notice, and bring them to life!

We’ve written just about all there is to say about key performance indicators (KPIs). And we’re all very familiar with the KPIs offered by all analytics solution providers, which are generally very relevant.

But that doesn’t mean we should stop here: Many other custom KPIs, or meta-KPIs, can help us see things much more clearly!


ninja interogation


So what are these KPIs, and where can I find them?


First, we can use segmentation. Comparing two segments (for example, “buyers”/”non-buyers”, or “British”/”German”, or even “did”/”did not” take a certain action, etc.) or cohorts will add truly useful information to a great many of your analyses.
We can then create our own custom KPIs based on our needs.


ninja interogation


What can I learn from comparing “buyers” vs. “non-buyers”?


Let’s take the example of a basic performance indicator that’s very often overlooked: time spent per visit.
While everyone knows this indicator, few think of it as critically useful. And yet…
Example: An interface screenshot for entering visits (a consumer electronics site studied over a period of two weeks in July 2015):


screenshot of entering visits

Let’s compare with the same metric for the “buyers” segment:

visits and visits with purchases according to time spent on site

We can already see that purchases are made almost exclusively during long visits.
When we look at visits of significant length (greater than 2 minutes), we can even make out a probable negative correlation (indicated by the red arrows).

This is even more marked when we examine turnover:turnover/revenue broken down by length of visit

We see that just 22% of visits generate 86% of revenue.
(Well, well, the good old Pareto principle still holds true in the 21st century…)
Given this trend, let’s look at conversion rates, which should therefore give us some very useful information:

Conversion rate by time on site

Surprising, isn’t it? A strong correlation appears between visit length and conversion rate.
To go even further and improve ROI, we must determine several factors, like the number of categories or products viewed before a purchase decision is made, and the quantity of products purchased.

In conclusion, it would be wise in this case to orient your content strategy toward retention, in order to keep prospects on your site for a long period, and to determine calls-to-action, additional information, suggestions, and specific offers that are triggered according to visit length.

Visits shorter than 10 minutes convert much less. It’s during this timeframe that it’s profitable to motivate the potential customer. Simply increasing the visit length is not enough; you must create attractive, clear content with useful (and targeted, if possible) information which will make the user experience both easier and more enjoyable.

And for tracking, everyone can create custom KPIs and charts according to their needs.

For example, the issue can be summed up in a chart that clearly shows the share of each visit duration in terms of visits, conversions and turnover:

Share of visits, conversions and revenue for each visit length

Some people prefer to enhance the visualisation by reasoning based on the positive or negative impact of each visit duration (here, the KPI calculates the weight differences of conversions and revenue vs. number of visits):

Impact of conversions & revenue vs visits

ninja interogation


So which KPIs should I use for tracking?

As we mentioned, everyone can create custom KPIs based on needs and preferences. In the example shown in the two graphs above, we can choose percentages, or transform them into relative or absolute indices. It’s just a question of user preference.

Whatever you choose, the process is simple and quick, and it can be done within apps that are near and dear to analysts’ hearts: Custom Metrics, Segments, and Dashboard Manager:

  1. In the Custom Metrics app, I created a global conversion rate (not limited to conversion funnel) = conversions/entering visits.custom metrics interface
  2. In the Segments application, I create segments based on visit lengths that interest me: V5-10 for visits between 5 and 10min, then V10-20, V20-30 and V30+ for four segments total. Example of V10-20:Segments application interface
  3. We can now create specific KPIs by cross-calculating ① and ② in the Custom Metrics app, then integrating them in a dashboard with the Dashboard Manager.dashboard app interface

By using these three applications, it’s possible to create segments, metrics and custom KPIs in just a few clicks, and then integrate their data and charts into a dashboard.

It’s great news for your bottom line!

Note: In this example, I chose to focus on visit length, but of course, the same process can be applied to many different indicators.


You tell us…

We’re interested in your comments and experiences!


Brush up on your custom KPI best practices! Download KPIs: Define and Act, our free white paper on integrating KPIs into your company’s strategy.



Knowledge Manager After On-the-job and Off-the-job training in purchasing and management at Carrefour, and sales training at Procter & Gamble, JM evolved in the mass retail sector in top management positions for large hypermarket, central purchasing and logistics groups, with an expatriate experience in Africa as a Central Director. In late 1995, JM created an Internet start-up company and after three years (late 1998) he joined Alain Llorens and the AT Internet team where he took up his position in sales, and was also at the heart of the pioneering adventure in Web analytics. At 55 and after almost 13 years seniority in the company, JM has been Knowledge Manager since 2009.

Comments are closed.