When you’re talking about conversion, a good practice is to draw a conversion funnel to represent each step your lead should go through until your ultimate goal (revenue).

A conversion funnel often looks like this:

Typical conversion funnel
Typical conversion funnel (image from Kissmetrics)

If you look at the chart, your intuition will tell you to increase the conversion rate at each step to maximize your final metric.

It’s comfortable for marketers to focus on a few figures and work on them. What most forget is that these intermediate conversion rates are generally negatively correlated.

Let’s take a typical example. You decide to build a minimal landing page (or minimal homepage) to maximize the number of free signups. You shorten the page, hide navigation and add contrast on your call to action. These efforts result in a 100% increase in signups. Yeah!

If you follow the funnel’s logic, this improvement should ultimately result in a 100% increase in revenue. But it isn’t. Your revenue is only increasing by 10%. Why?

Before your optimization, people who signed up were essentially people interested in your solution. Now many of them are just curious people who gave their email because the solution could hypothetically fit their needs and they had to give their email to know more. Some of them even entered a disposable email address because they aren’t engaged enough. In one word, your user base became much less qualified.

Not surprising if this impressive increase in conversion didn’t impact your revenue with the same panache.

Focusing on one step can even be counter-productive.

I experienced it a few weeks ago. On one of my websites, I sell a product in two steps:

  • First, the visitors arrive on a sales page. Here I describe the problem, explain in details the benefits of the solution and display social proofs. It’s a quite long page and the visitors don’t know yet they’re being sold something.
  • At the very bottom of the page, there is a form the visitors can fill with their email address if they’re interested. Then they arrive on the checkout page where they can see the price of the product and finally buy it.

I made a test in order to increase the number of users arriving at the checkout. My sales dramatically dropped.

Conversion funnel A
The original version, with only one form at the very bottom of the sales page.
Conversion funnel B
The new version with two forms to increase the number of email addresses collected.

Let’s do the math to get the revenue conversion rate:

  • Original version: 24% * 5% = 1,2%
  • New version: 27% * 2% = 0,5%

I literally destroyed my sales by wanting to collect more leads. In the second version, most visitors filled the first form to continue. They only saw the first part of the landing page and spent less time on it. Their engagement wasn’t strong enough to take the next step. I hadn’t “prepared” them enough to purchase.

Two lessons learned

I learnt two lessons from this.

  • You need to follow the metric directly concerned by your experiment, but also the impact further in the conversion funnel. Make sure to follow these several goals simultaneously to make the best decisions.

  • It’s essential to have a less mechanical and more psychological vision of conversion. You have to take a step back and ask yourself: what is the state of mind of my lead at this point? What are exactly their motivation?