Our last post highlighted an infographic on how to improve your conversion rate. Marketing people, myself included, love to geek out over conversion rate techniques. Make the button bigger. Add copy to the button. Demonstrate social proof…and so on. When you combine a bunch of these techniques the results can be pretty dramatic.

When our sister company, Rand Media Group, redesigned Tula Software they got a pretty nice bump.

We’re converting new visitors into trial users at 4 times the rate we were previously, and they’re far more engaged on the site when they do visit, visiting twice as many pages.

I tell you all this for two reasons.

1. To self-promote.

2. I do think conversion rate optimization can be very powerful

However, you can take it too far. Here’s how:

Numbers are just numbers

It’s easy to get fixated on a number. Typically we get fixated on numbers that are easy to understand. Click through rate (CTR), unique vistors and conversion rate are all numbers that are easy to understand. You’d think that the higher these numbers are the better. Not always. What if you apply a dollar sign to them? So cost per click (CPC) and cost per acquisition (CPA) seem pretty solid, right? Well you are getting warmer, but you’re not quite there yet.

The amount it costs you to acquire a new customer is pretty good, but there is a lot that number doesn’t give you. You may sign up customers at a low cost, but find out that those customers don’t stick around that long or cost more money to service. Depending on the nature of your business, you may not know that for several months or even years.

Standard deviation

Unless the data you are looking at is statistically significant, you can’t trust it. If you run two ads and one gets 12 clicks and the other gets 20, do you really know that one ad is better than the other? You need to let ads run for a while or invest a significant amount of money to really know which ad is likely to generate more clicks.

A lot of small businesses don’t have the cash or timeline to know with absolute certainty which ad or landing page is the most optimal. That is ok. Use your best judgement and do what you think is the best. Then forget about it for a while.

Opportunity cost

Proper testing takes knowledge, time, effort, tools and money. If you are a business owner who is into that stuff, watch out. You could spend a lot of your ultra valuable time trying to squeeze a percentage point here and there. If your website doesn’t have enough traffic, you should be spending time doing other things like getting traffic in the first place or creating a business that is worthy of attention.

Seth Godin has a brilliant blog post where he says:

But those that manage to capture the imagination, make sales and grow are doing it by perfecting the things that matter and ignoring the rest.

Optimization matters, but it’s often not the most important thing.

What’s the point?

I’m not trying to talk you out of testing and looking at data. But you can’t throw out common sense and intuition just because you have data. In fact, you’ll need it more than ever.