For this month’s series on PPC Hero, we’re taking a look back through our archives and revisiting some ideas we’ve talked about in the past and how the past year or more has changed our thinking about them. Things are always changing in PPC, and our views and opinions about it are no exception.
When it comes to conversion rate optimization (CRO), I would say that you might find it difficult to locate anyone that would fight against more of their website traffic converting. And I’m not just talking about the PPC traffic…ALL the traffic.
I wrote a post last August about how to get your clients to buy-in to CRO and CRO testing. I still fully stand by those general thoughts and ideas; however, as our team has worked more directly on CRO for our clients, now I’ve got some hard data facts on the kind of results and improvements and account can see from even ‘simple’ CRO tests and strategy implementations.
The original list contained the following tips:
- Research your competitors.
- Put together a plan and implementation strategy.
- Estimate the benefit to the client and report it to them.
- Pull data from previous CRO tests you’ve completed.
- Once you get a yes: you had better make it count!
Now the best part – the additions to that list!
- Don’t assume the test or change has to be large to be effective.
Some of the most successful CRO tests our team has been running for clients as of late have been somewhat minor in nature, but the return on those minimal changes has been more than impressive. Things like matching up PPC ad headlines with the headlines on the landing page and increasing font size for readability (yes – just those two changes) have resulted in 33% lift in conversion rates for an active client; and this is after the client-side team already had a CRO testing schedule in place. The very basic best practices can sometimes get looked over for bigger, sexier testing ideas, but they also can be responsible for the greatest improvements.
- Just because previous tests were less-than-stellar, doesn’t mean a new, fresh approach will have the same outcome.
Many of our clients now have done some version of CRO testing on their own or with another agency, or some other way…but they have. The deck tends to be harder to unstack against ourselves in these situations, because either the tests failed or the gains were minimal and didn’t leave an impressive CRO taste in the client’s mouth. You have to find a way to dig in to the tests they ran, what happened/went wrong, and then explain how you might do it differently (or better, of course). Additionally, maybe they ran too hard at the CRO side of things! Remember what I mentioned earlier about small tests being what we see truly add up? A lot of first time CRO testers go for the big guns and then don’t see big gun results, so they run off and say “who cares.” You should! Find some simple best practices they didn’t test in those previous attempts and push for those.
- It may be bright and shiny, but that may not be good for CRO.
There are aspects and elements of landing pages that when you think of them, certainly just feel like they would make sense to improve your conversion rates. Guess what? Sometimes our common sense thoughts are completely wrong! For example, we have a client who had an informational video on their PPC landing pages for potential customers to view and learn more about the client’s product. We had a goal set up to track the interaction with the video, and we saw that the percentage of site visitors who were pressing play was rather small. With that information, we decided to test pulling the video off the page entirely to see if its mere presence was driving away conversions. And it was! After removing the embedded video, conversion rate increased by 20%. The potential customers were still finding the information they wanted regarding the product and most importantly, now more of those people were moving to the step of signing up for a free trial. Moral of the story? We may think we know what our client’s customer’s want, but we may also be overloading them with too much opportunity to interact, therefore confusing them and driving them away.
- Unless your conversion rate is 100%, you need CRO testing.
I actually just posted about something similar to this, but in relation to CTR, just last week. I get the question “What’s a good conversion rate?” almost as often as the same question for CTR. My response is also now being duplicated for CRO as what it is for this question related to CTR – unless you’re converting every single one of your website visitors…you need to optimize your pages for more conversions. We’ve also encountered clients who are “happy” with where their conversion rates are at the time and don’t want to push tests. Wait…what? To that, I say this – you may be happy today, but a new competitor could be waiting in the bushes to pounce tomorrow and then when your conversion rates begin to slip, you won’t be happy anymore. Get ahead of the curve and keep on testing!
- Come up with a unique and/or specialized metric to analyze CRO successes and performance changes.
Back in August when I brought up reporting performance from CRO tests to your client, I was a little vague. I mentioned determining possible value of those lost conversion opportunities before you test, as well as reporting results, whether they be wins or losses in conversion volume. Since then, our own team has come up with a way to more specific about what possible results could come from an individual CRO test, and the same metric can be used to measure the results when it’s over. We call it Earned Average Conversion Rate (EACR) and it calculates the percentage of conversion rate lift expected or seen by any individual CRO test. Use something like this or one you come up with yourself, but find a way to measure each test and what you’re doing with it.
Tell us some of your experiences with CRO and how to manage that process with your clients alongside your PPC efforts in the comments section below – the good and bad, ups and downs! Thanks for reading and stay tuned for the rest of the series!
