Conversion rate optimisation : a 9 step process

An interview covering the process, tools and skills needed to increase conversion, leads and sales

Karl and Ben at the Googleplex

Many web consultants have relevant-but-similar backgrounds. They’re experts at design, or pay-per-click advertising, or another specialty.

But when I recently met Ben Jesson and Karl Blanks from Conversion Rate Experts their background didn’t fit the norm. Their company was founded when a real-life rocket scientist teamed up with a internet marketing specialist to look at websites through unconventional eyes.

This unusual perspective has turned out to be highly profitable for their clientele, which includes firms like Google, Vodafone, and Sony.

In this interview we explore how Conversion Rate Experts approaches optimisation since I think companies, agencies and consultants can learn a lot to their approach to returning the commercial returns from websites of all types.

I’ve asked them about the process they use to increase leads or sales from an online business and we’ve included links to a lot of the tools they use.

2012 update:

When I completed the original interview with Ben and Karl back in 2010, I asked whether they had a summary of their process to share. We did cover this in Question 3, but recently Karl has been in touch to say they now have an infographic to show this process. I thought I’d share it so you can review your processes against it. You can read more about their process on their blog where they expand upon each step.

What is Conversion rate optimisation?

Q1. We’re seeing a lot more companies working now on CRO. What is it? Is it more than landing page optimisation?

Yes, it should be. Landing page optimisation focuses on one page. We coined the term Conversion Rate Optimisation (“CRO”) in 2007 to describe the process of optimizing the business itself. It’s really commercial optimization

A proper job of CRO includes the review of the entire process from the initial lead-generation ad, all the way through to the post-sale follow-up. The real goal is to identify which parts of the sales funnel will yield the greatest wins with the least work.

That means it’s necessary to bring a lot of disciplines to the party, including understanding traffic sources, visitor psychology, and the company’s position in the marketplace, including its core strengths and weaknesses. On top of that there’s usability testing, copywriting, and web design factors to look at.

All these elements go into creating hypotheses for testing. We’re maniacal about testing, because we’ve seen too many businesses merely throw a series of “best practices” against the wall to see if anything sticks. Best practices should not be the answer to optimizing a website, but merely one starting point for formulating a test strategy.

Once we determine what truly works for a particular website, then we examine how our findings might be used in other media channels. For instance a better series of benefit statements might be transferrable to direct mail or email autoresponder campaigns””subject to testing in those media, of course.

The business case

Q2. How do you help companies persuade colleagues of the returns from CRO, the business case?

It’s easy; we explain that CRO allows companies to generate more revenue without spending more on advertising. It’s about getting a higher return from the existing ad spend.

Unlike certain industries like public relations, the entire foundation of CRO is based on data, measurement, and testing. You don’t need to present arguments when the data can do the talking for you. Once you measure the value of visitors, conversions, and sales, then it’s simple arithmetic to show how, say, a 10% boost in conversions would help the bottom line.

Here’s another powerful side-benefit: When you optimise your funnel and bring in more revenues, you then have earned a luxury: You get to decide whether to pocket those profits or plough them back into even more advertising, thus distancing yourself even further from your competitors. It’s a nice problem to have.

Identifying the biggest opportunities

Q3. What approaches do you use to decide which part of a site needs most urgent attention?

FORTUNE magazine called what we do “a combination of multivariate statistical analysis and good old-fashioned detective work” and that pretty well describes our approach.

It’s often very useful to map out your entire sales/conversion funnel and make sure it’s being comprehensively measured in whatever web analytics package you prefer.

Then you should look for the biggest drop-offs from one step to the next. We like to say that we look for the “blocked arteries” (that is, pages””or page elements””that get loads of visitors but are underperforming). How do you know if something is underperforming? Clues come from a range of feedback mechanisms: the analytics data, usability tests, surveys, customer support feedback “¦ and, of course, gut feel. Of course, we have the advantage of having been engaged by companies on several continents and in many industries, so we have a good knowledgebase of what’s good and what’s bad. See our list of effective tactics and strategies.

What mistakes limit conversion?

Q4. Give some examples of the most common “conversion rate killers” you see.

Killer #1: Not split testing. Many people think they’re done if they take action to make changes to their site. In reality they’re only “done” when tests show that the changes in fact improved conversions. Installing a “best practice” magic button that another site swears by might actually lower conversions. Despite the popularity of video, Google once discovered through tests that video reduced conversions on one of its pages. You simply must test to find out.

Not long ago, multivariate testing software cost more than £5,000 per month. Now you can use Google Website Optimizer and other software packages for free, so there’s really no excuse. We created a tool, called Which Multivariate, which helps you to select the best software for multivariate testing.

Killer #2 is “meek tweaking”””in other words, making changes that are never likely to have a significant effect.

Killer #3 is asking for the sale on the first visit. It’s often a good idea to test the creation of a multi-step conversion funnel, in which you provide great value before you ask for the order. Comparison charts, forums, special reports, and email marketing are examples of elements that allow you to provide good information, ask for names, cultivate a relationship and thereby improve the chances of a sale.

We recently gave a talk on 15 Common Causes of Conversion Death

What are the latest approaches?

Q5. Are you seeing any innovative design techniques that are helping conversions?

Excellent design is a prerequisite for conversion, but the biggest breakthroughs tend to be the new tools and techniques for gathering insights into the visitor’s mindset.

For determining how visitors interact with a site we often use both Clicktale and CrazyEgg.

KISSInsights and Ethnio are both good for asking your visitors to give you immediate feedback on your site. GazeHawk enables you to conduct an eye-tracking study on your site for a tiny fraction of the traditional cost.

Many of your readers will already know about how wireframing is important in order to get agreement on functional aspects before you take the time to make a site look good. We like Balsamiq for that purpose. [Editor - I use that too - a great simple wireframing tool for consultants and agencies]

Finally, your readers may want to get our free newsletter to see what a million-dollar landing page looks like, along with a graphical analysis. We call it that because it generated over a million dollars for one of our clients.

If you want even more examples of what’s possible with conversion improvement, watch the video in the header of our website.

  • http://www.keepitpersonal.co.uk Andy Smith

    Thanks for the newsletter, as a small business owner it’s tough to work on sourcing more traffic due to the huge amounts money our larger competitors are ploughing into SEO which we just cannot compete with. I am a frequent visitor to the conversion rate expert site and take a couple of days here and there when I have time to update the website and go through the list of tips and advice provided. Traffic is slowly increasing throughout the year but conversions have increased for sure. I currently have 3 websites in the UK one of which Keep It Personal, it is our flagship website however it was last redesigned 2 years ago and will need updating soon and offers our entire range of gifts.

    I am testing some new features on our new website selling justChristening Gifts, it is more relevant and should have higher conversion rates since it is a specialist and not a generalist site, does not require an account to make a purchase, has testimonials/reviews, a large “personalise this item!” button at the top of the product page and other tips which were gleaned from reading tips on increasing conversion rates. The design is very basic, but deliberately done so to make conversions as simple as possible, ie no drop down cascading menus, clearly defined categories and lots of photographs of the products. The testimonials show a big yellow smiley face, which I think makes the site friendlier, similar to the mascot on the conversion rate expert site. Analytics are showing high conversion rates for the small amount of gifts currently on the site so far.

    Keep up the good work, any more tips gladly appreciated!

  • http://www.emailsweets.com Goran

    Great post!

    I am currently doing the same thing (CRO) for quite complex sales process that takes around 25 stages and spans over 8 months. It includes all aspects of internet marketing (SEO, SEM, email marketing, branding), user behavior and understanding personas which are in top-level maslow pyramid, up to CRM and post-sales.

    My insights are that first thing to do is to identify and kill “glitches” in current process (bad pages, bad design, bad copywriting). When you get results from these obvious mistakes, than is time to get hands-on on the complex things. That’s why I think that LPO is a good start for complete conversion optimization.

    Regards!

  • http://www.makingyourowncandles.co.uk Kevin Partner

    It still amazes me how few people talk about conversion rate. My clients are all obsessed with their organic ranking in Google or creating a low cost Adwords campaign but traffic is utterly meaningless (and potentially very expensive) if it doesn’t convert. Increasing conversion rate is free money, after all. I’m bookmarking this post and I’ll also be investigating the links it contains: first on my list is KISSInsights which looks as though it may take the pain and hassle out of short surveys. It also has the benefit of striking while the iron’s hot.

    Excellent post, thanks very much – truly valuable information. How many will act on it though?

    • Dave Chaffey

      You’re right Kevin, my experience is that many many websites are static for a 2-3 year period between refreshes, expect perhaps landing pages for PPC if used. Interesting to think about what’s causing this, some ideas:

      – CMS limits changing of portlets/containers to refresh offers – in left or right sidebar or main body I always give Amazon as the example of a flexible if not elegant template layout
      – CSS not designed in such a way to be flexible or at least access to original CSS developer to document code is limited
      – Management doesn’t have a CRO mindset, or at least they don’t invest in resources/time to improve conversion, instead there is still a marketing campaign mindset rather than looking at continuous improvements to the site
      – Where there is dedicated resource, web analytics still focuses on improvement rather than optimisation in many organisations

    • Peter Bourne

      Dead on, we see the same effect. I think it comes to simple math, compound interest. Convert a visitor and that’s a customer for life, and potentially an ambassador for the brand. We tell our clients that 97% of their marketing budget is wasted (based on a 3% conversion rate average) and that tends to get their attention. Amazingly, we find that many have not fully grasped that going from a 3% CR to 4% is not a 1% improvement; its a 33% improvement in revenue and potentially profit.

      • http://www.smartinsights.com/ Dave Chaffey

        Thanks for your comment Peter,

        I like this part:

        3% CR to 4% is not a 1% improvement; its a 33% improvement in revenue and potentially profit.
        Stating the obvious is often useful. That said, change the dial to increase conversion by that amount could be a challenge depending on how optimised they are now.
        Dave

  • http://www.16i.co.uk 16 i

    Nice article Dave.

    Karl and Ben are definitely the pioneers of CRO, and this is a very useful resource to give peopke an informed background on the subject.

    As with all their articles, this one is well optimised to convert traffic of their own!

  • Pingback: Conversion rate optimisation – why is it important and how do you do it? — The Woof blog

  • http://www.impactmedia.co.uk Samantha Stratton

    A really good post, thank you!

    I have been asked to explain the difference between CRO and LPO on many occasions now and the first section of your post really sums it up.

    I think using both really helps you to get to grips with how your users interact with your site and CRO definitely reaps benefits when it is done right.

    We are going to start using KiSSinsights on our site soon and hoping to get some really insightful feedback into why visitors come to a site and do or don’t convert.

    • Dave Chaffey

      Glad that’s helpful Samantha,

      I think many see CRO = LPO which of course is massively limiting the potential for improving customer journeys, conversion and revenue, so it’s good to see a “holistic” approach explained here.

      As Matthew Todd at Logan Todd always points out, conversion rate is a meaningless metric in isolation, it needs to be segmented and placed in the context of revenue and profitability improvements.

  • http://twitter.com/BenGriffithsMe Ben Griffiths

    2011 is the year of CRO for me. I’ve been spending the last few months cleaning up the horribly antiquated code of the websites I inherited at my (relatively) new employer. I can’t wait to really dig in with some extensive testing!

  • http://www.firstfound-blog.co.uk/ Andy @ FirstFound

    I think I’ve always linked CRO and LPO in my head, so yeah, I agree with Dave. The “holistic” (not a fan of that word) approach is a breath of fresh air

  • Pingback: White Hat, Black Belt » Blog Archive » Friday Link Love: The Primordial Ooze Edition

  • http://profiles.google.com/q2results Sam Ingersoll

    But what if you are creating a new website :) Shouldn’t there just be a template to follow? hah hah. Actually, I took everything I could find on Conversion Rate Experts and UX Matters and some other sites/research and created a wireframe and description of “the perfect conversion page.” I need to slick it up but it’d be a fun share sometime soon.

    P.S. Of course, what you need to do first is (a) be better than your competition, (b) prove it, and then (c) communicate it effectively. The CRO design is most directly related to C.

    • http://www.smartinsights.com/ Dave Chaffey

      Hi Sam, thanks for the comment – let us know when you have your “perfect landing page wireframe”. Depends on product or service though, so maybe need a few variants for each market

  • Pingback: Click. Industry Insights from Red Door Interactive | Red Door Interactive

  • Dejeesh dev

    Great post, since I’m not a Conversion rate professional specialist. but I love this. I have been working online marketing industry for the last 2 years but I didn’t care more about conversions, but now I realizing as an Online marketer I have look on this and we need special attention to improve this. This post is really helps to get basic ideas we have to look on for more conversion with low budget.

    Many Thanks Team!

  • http://www.infiniteconversions.com/ Infinite Conversions

    Great article, Dave. We really feel that 2014 will be the year CRO gets more mainstream.

  • Pingback: 30 (New) SEO Terms You Have to Know in 2011 | White.net

  • acquireconvert

    Wow this is a really great post.

    I feel that there was not enough discussed about the fact that many people collect bad data or mis interpret data.

    I agree with Kevin though, few people are even considering CRO, so it is expected that these mistakes will be made

    This is my take on the process I wrote as a reply

    http://goo.gl/TpNVBW

Get FREE marketing planning templates

Start your Digital Marketing Plan today with our free Basic membership.

  • FREE fast start guides to review your approach
  • FREE digital marketing plan templates
  • FREE alerts on the latest developments

Get FREE marketing planning templates

Start your Digital Marketing Plan today with our free Basic membership.

  • FREE fast start guides to review your approach
  • FREE digital marketing plan templates
  • FREE alerts on the latest developments
Feedback Form
Feedback Form