December 17, 2008
Think you know how to test landing pages? Put your knowledge to the test and see if you’re on the right track with Tim Ash, author of Landing Page Optimization: The Definitive Guide to Testing and Tuning for Conversions, and President of Sitetuners.com. Tim is an entrepreneur, internet marketing leader and co-founder of several internet related start-ups. Currently Tim is the President of Site Tuners, which is a company that helps other companies test their landing page to optimize for higher traffic and conversions.
Tim has a very impressive resume, which is one of the main reasons I picked him for an interview. He’s worked with companies like American Express, Sony Music, Honda, Coach, COMP USA, Universal Studios and much more. But what really drove me to this interview was Tim’s presentation at the 2008 PPC summit in Los Angeles. His presentation opened up many ideas for me and others when it comes to testing your landing pages. What I walked away with was a copy of Tim’s book, Landing Page Optimization available on Amazon.com, and a plethora of ideas for testing my own landing pages to increase traffic and sales for my clients.
It would be wise for you to read this interview and write down as many ideas as you can for testing your own landing pages. And oh yea, don’t forget to buy Tim’s book!
1. Tim, for those who haven’t read your book yet, can you tell us what the number one mistake is people make when it comes to testing their landing pages?
Too much clutter – you have to have a singular focus and discipline and be very clear about what the desired conversion action is on your landing page. Everything else should be de-emphasized or ruthlessly eliminated.
2. What would you say is the appropriate time frame to test new content on your landing page is before deciding whether it worked or not?
Our tests usually run a few weeks. You should at least run one full week to get rid of time-of-day and day-of-week biases. But the real answer is that you should wait until you have a high statistical confidence in your answer. In other words decide how sure you want to be that you have found something better (95% for example) and wait until you reach that confidence level. It is hard to predict the exact amount of time because it depends on how much better your new version is than your original. If you double the conversion rate, you will not have to wait that long. But if you only find a 5% improvement you will have to collect data for much longer to make sure that it is not a short-term lucky streak (like flipping a coin five times and having it come up heads each time).
3. You mention that when testing your own landing pages your ideas or opinions can often be wrong. Do you feel there are best practices per business category or is every business different in its own way when it comes to what works and what doesn’t?
Each business is different because of it goals, cost structure, product offerings, brand strength, and online traffic sources. So nothing works universally. That is why you have to test on an ongoing basis and can’t simply copy your competitor’s landing pages.
4. I think a lot of people out there feel they don’t have time to test or it’s just not that important. In your opinion, just how important is it to constantly keep testing even if you feel things are going well?
You should always be testing something. It is possible to improve a page a lot and then run out of good ideas for what to test next. But you should then move on to another important page on your site.
5. We feel at times in order to see a large result you have to make a large change. Do you follow this way of thinking? Or do you recommend making smaller changes at one time?
Small changes can have a big impact. We increased conversions 58% on a lead form just by changing the title from “Free Quote Request” to “Instant Quote”. On the other hand we have tested completely redesigned pages against each other and gotten even bigger increases. So there is no right answer – it really depends on your risk tolerance and the amount of work you are willing to do upfront without any guarantee of results.
6. Is it best to only do A/B testing so you know exactly what worked and what didn’t?
I am not a big fan of what I call “after the fact meaning making”. It is really hard to tell why something worked – so drawing conclusions can be a dangerous business. Was it your button color or its contrast with everything else on the page? You may never know. It keeps people from testing the same ideas in another test – they assume that they already know “the answer” even though the context and audience are different.
7. In your presentation I thought it was interesting when you asked who should design your landing page…marketing, president of the company, programmers, web designers….the answer being none of the above. Your customers need to design your landing page. So how do you know exactly what your customers want or are looking for, or don’t like about your current landing page?
Try to test the impactful items on the page like headline, offer, call-to-action, sales copy, images, etc. You don’t know what your customers will prefer so have a wide range of ideas to make sure you do not test what YOU prefer.
8. When should one stop testing?
When you are out of good ideas and have not gotten meaningful improvement sin a few tests on the same page. At that point you should move on to another high-value page.
9. I wrote a blog post on trust symbols and credibility logos based off your presentation. How have you seen these items improve landing pages over time?
Transferring trust and credibility in the form of third party endorsements is always a good idea. This can mean same shopping seals, media mentions, client logos, or testimonials.
10. I think what is even more important than just testing elements on your landing page is being able to review the data and analyze it before making a decision on what worked and what didn’t. Is there a particular metric one should be looking for over another?
The metric depends on your conversion goal and could be revenue per visitor, form-fill rate, click-through to another page, or time-on-site depending on the conversion action. If you have multiple goals you have to assign a financial value to each one to make sure you are making the correct trade-off among them.
11. What is the second step one should take to begin testing their landing page? (First step is reading your book!)
After reading my book they should set up a free Google Website Optimizer account and test the headline on their highest volume landing page in a simple head-to-head A-B Split test. This can be done in about 15 minutes.
Thank you Tim, for taking the time to answer our questions. I’m sure readers will benefit from this and hopefully start testing their own landing pages more often and more effectively!