Can you do too much conversion optimization? When should you stop?
Recently, an attendee in a webinar I was presenting asked an interesting question:
Is there such a thing as too much testing and/or optimizing?
It’s not an unusual question. In many of my conference presentations, someone asks me something similar and I’d like to answer it today in our ongoing series of conversion optimization FAQs.
There are two sides to this question.
The Law of Diminishing Returns
On one hand, the Law of Diminishing Returns suggests that your testing efforts will, at some point, produce lower value than your early testing. The early wins should give the biggest improvement as you harvest the “low-hanging fruit” by fixing the glaring problems. After that, large improvement opportunities should become more difficult to find.
Also, conversion rates can never reach over 100%, so that’s a hard stop.
If that’s true, then there must be a point where the cost of planning and producing tests outweighs the benefit of conversion rate improvement.
But, that’s not the way business works.
On the Other Hand, Should You Ever Stop Improving?
I used to think we’d see more impact from the law of diminishing returns, but we haven’t yet.
The question, “When am I finished with conversion optimization?” implies that there is an “optimized” point – that there’s an ideal page. In practice, you’ll never hit the 100% barrier.
There is always room for improvement. There are always competitors innovating. There are always questions to answer.
If you stop testing, the rest of the web pulls ahead. That’s why we need Evolutionary Site Redesign.
I believe we are still in the early days of testing new web experiences. Innovative designers and UX architects are coming up with new experiences all the time. Some are terrible ideas and others are going to work better than our current standards. The beauty of the scientific approach is that you can kill the losers much more quickly and definitively without having to worry about them turning into sacred cows. The pace of business is too fast today to allow them to develop.
The question we started with also implies that the goal of conversion optimization is only to find that perfect conversion funnel.
But, that’s not the only goal of conversion optimization.
I Hate the Term “Conversion Rate Optimization”
I want to step back for a moment and challenge the entire conversion rate optimization concept.
Part of the problem is in the name. “Conversion rate optimization” as a term is a misnomer. It puts the emphasis on the measurement of success rather than the outcome of success.
The conversion rate metric should be the method to answer your important business questions, not the end in itself. Thinking about it in a broader decision-making context re-orients your ideas from just tweaking and tuning minor page improvements to be a way of making important decisions.
What we’re talking about isn’t just “conversion rate optimization.” In my conversion optimization book, I introduce the term “Strategic Marketing Optimization,” which I think is closer to reality, but it still implies perfection as a goal.
When you think about the discipline of iterative testing, its true value comes from delivering a better process for decision-making. It should be a continuous improvement process. I think it should really be called “Scientific Marketing.”
The question to ask yourself, then, is really:
Is there such a thing as too much decision-making?
Is there such a thing as too much business improvement?
The answer is clear.
You should test when you have a question to answer; when you need to improve your business; when your competitors are making improvements.
There is no “too much testing” just as there is no “too much deciding.”
We won’t get away from using the term Conversion Rate Optimization in the near future, and I’ve made a certain peace with it. But it should be viewed as more; it should be your crucible for decision-making.
Add your voice to this year's "State of Experimentation" survey!
We just launched our annual "State of Experimentation" survey. If you're involved in digital experimentation, we would really appreciate your input. All respondents will receive the complete, global results ($5K value) for free. Survey closes June 23rd!Take the Survey >