BLOG

on Conversion Rate Optimization

Persuasion Design: A Conversion Rate Optimization Framework

“Everybody” knows that a website should have a purpose, that is, to fulfill both user goals and business goals. Likewise, “everybody” knows that improving usability greatly help to increase the percentage of users reaching these goals.

But first, I’d like to talk a little bit about the real value in higher conversion rates:

The fact is that it’s not the increased conversion rates that’s the biggest value here – far from it, actually. You see, your profit is incredibly sensitive to increased conversion rate. Let me illustrate:

In the scenario below, we’re assuming a company generating a 1 million per month revenue. But obviously, all that revenue isn’t the bottom line; every company has expenses, such as sales, marketing, staff salaries and other expenses. In this scenario, the company is left with a 100,000 per month profit.

If you are able to increase the conversion rates by 50%, then revenue would go from 1 million to 1.5 million. Most CEOs, sales executives and investors would jump with joy by that prospect alone – but look at what happens to the profit:

Profit increase because of Conversion Optimization

600% increase in profit. Now that is something to celebrate!

(and even though your other expenses should increase some due to higher sales, you are still left with a remarkably improved bottom line.)

Your organization becomes more robust:

You add a permanent increased value to your website because:

  • Your entire website becomes a more efficient “sales person”

  • You increase the ROI on your marketing budget (instead of throwing money into the bottomless pit of the marketing hole)

  • You’ll be able to exploit marketing channels that previously have not been economically viable, such as print and TV.

  • And this in turn makes your marketing strategy more sustainable because you’re no longer as susceptible to changes, for example algorithm updates in search engines.

Synergies from online to offline:

  • By testing online, you are able to identify what kind of communication is proven to be the most effective.

  • This allows you to use test winning communication concepts and messaging in offline media – where we all know measuring success is a lot harder.

It’ll save you a lot of time and a lot of frustration:

(and it’ll give you inspiration)

  • My method will save both us and our clients the eternal discussions about what each individual, with their personal preferences, think and feel is the best solution.

  • Instead, you can spend your time on what is actually fun and inspiring to work with, namely to create great ideas and concepts (perhaps even make it a contest to come up with the best test results.)

Case study: Allinclusive.no

Allinclusive.no is a holiday portal specializing in all inclusive travels. The company, founded and run by Mr. and Mrs. Carlstrøm, generated a revenue of 12 million NOK in 2011.

This is the story of how one test and 64 words increased their bookings by 53%.

allinclusive.no Original page

Vs.

allinclusive.no Variation

As we were approaching one of their top seasons, my biggest challenge was that I had to identify opportunities that didn’t require programming or lots of redesign work, simply to keep both cost and time spent to a minimum.

Throughout my analysis, I did identify several great opportunities in terms of usability and process optimization, but the response was (to quote MC Hammer) – You can’t touch dis’ (and I bet that you this instant got the song in your head – sorry!)

Anyway – By replacing the rotating campaign banner on the front page with a very deliberately crafted copy, I was able to increase their conversion rate from 0.50% to 0.68% – an increase of 52.3%.

Editor’s note: This test was conducted using Visual Website Optimizer.

 Graphs showing increase in conversion rate

This test is a testament to the power of a well crafted copy, and even with a very long sales funnel, the front page influences the visitor throughout the entire funnel. Indeed this influence becomes more and more pronounced the further in the visitor moves:

Front page ->

Search result

Hotel page

Upsell

Booking/Payment

Order confirmation

Improvement

+0.48%

+11.08%

+8.86%

+18.87%

+52.93%

Using the VWO/GA integration with eCommerce even revealed that while the conversion rate increased by over 50%, the revenue actually doubled.

How is this even possible?

Let’s start with the simple fact that your online revenue is determined by the following simple equation:

Visitors X Conversion rate X Average sales price

(Even if you don’t sell products or services on your website, this holds true – your revenue from your website, like lead gen etc, are still defined by these 3 factors)

  1. Visits are determined primarily by marketing, advertising, PR and customer loyalty

  2. Conversion Rate and Average Sale are determined mainly by:

    1. Motivation to perform the desired behavior (Persuation)

    2. Ability to perform the desired behavior (UX)

    3. Trigger to perform the desired behavior (Call to Action, like “add to cart”, “Contact”)

Looking at the above points, it’s starting to become very clear that there’s much more to improved conversion rates than just a more user friendly site – in the same way that you are able to design usability, it is indeed possible to design persuasion.

Over the years, I have developed frameworks for conversion rate optimization. In the beginning, I was very focused on the user experience and usability, relying on best practice and quantitative analysis.

The framework I’m giving to you today is a result of a cross-discipline approach drawing on fields such as UX, interaction design and prototyping (I love using Axure RP Pro for this), as well as web analytics, persuasive technology and social psychology.

Now, I must confess I am not an academic, but I have developed a profound interest in the areas of social psychology that are focused on why we do what we do, and what’s forming and triggering habits and routines. It’s incredibly rewarding both on a personal and, as you shall see in a bit, a professional level.

So without further ado, allow me to introduce you to what I call:

 Persuasion Design

The first step is to identify the Target Behavior, and this doesn’t have to be to increase sales – it might as well be to decrease unwanted behavior, such as reduce phone calls to a call center (they are costly, so it could save your company huge amounts of money).

This is also why I call it a framework for behavior management, not just for conversion rates; it gives you the ability to manage user behavior in general.

Start by identifying target customer behavior

The best way to start is by looking at your business and user objectives, as well as consider both your macro and micro conversions. In isolation, micro conversions may not seem like an obvious place to start, but due to some powerful psychological triggers, they can be used extremely efficiently to achieve macro conversions in a cleverly crafted multi-step conversion funnel (think a nested chain of micro conversions). But more about this in a minute 😉

Analyzing user behavior

Above all, the most important thing to Persuasion Design is to appreciate the fact that the website is used by real human beings. The second most important thing is to appreciate the fact that a page or a form never exist in a vacuum. There’s always a context in which the page exist on the site, and there’s always a context in which your visitor has arrived on your site. (This is what my friend Bryan Eisenberg refers to as “scent”)

Knowing both the mental and navigational context can be a hugely powerful tool to predict – and influence – user behavior and which questions or objections the user may have when arriving at your site and your pages.

And I dare say, the most powerful form of optimizations begins with redefining the most effective context by using the exactly right persuasion in the exactly right steps of the conversion process.

I usually divide Conversion Rate Optimization into 3 main categories:

  • Copy

  • GUI

  • Process (addressing friction & showstoppers)

Quantitative Analysis

Appreciate the fact that the users navigation is never a linear process. It never looks like the Goal Funnel you see in Google Analytics, for example. They with go back and forth, and in loops, stray away or exit.

To identify the navigational context, I use a quantitative approach using web analytics. In Google Analytics, you could use the “Visitors Flow” report to identify typical navigational patterns, and thus get a better understanding on how your visitors arrive at your page.

Find the first few occurrences of your “form conversion confirmation” page (like a “thank you” page or “order confirmation”), click on it and choose “Highlight traffic through here”. If there’s a lot of going back and forth between pages, there’s a very good chance you have failed to answer some critical questions earlier in the conversion funnel.

You can also use the “Visitors Flow” report to get a good idea about what your Goal Funnel should look like (In most cases, it should not to be just your form+confirmation page). You could then look at the “Goal Funnel” and “Goal Flow” to identify stages with high Exit Rates, then analyze each of these pages to form a hypothesis as to why people are exiting at these points.

If you haven’t already done so, do yourself a favor, and start reading Avinash Kaushik’s blog, and pick up his book “Web Analytics 2.0”, as well as Brian Clifton’s blog and book “Advanced Web Metrics”

And you definitely don’t want to miss out on the post “15 tools that reveal why your potential customers abandon your website”

Qualitative Analysis

To get a quick look into your customers brains, you could ask these 6 simple question via, for example, SurveyMonkey. Keep in mind these must be open text boxes so that the visitor can type in exactly when they think – not predefined answers (or you’ll lose a lot of valuable insight):

  1. How did you first learn about us?

  2. What made you decide to use us?

  3. How would you describe us to a friend?

  4. Would you recommend us to your friends or colleagues? Why?

  5. What do you think is our biggest advantage?

  6. If you could change just one thing on the website – what would it be?

(to get even more specific, you could use Quaraloo on individual pages to get exact intel on what they’re thinking in the specific stage in the purchase path they’re in.)

Even though these 6 questions should give you some very actionable information, this is barely scratching the surface, so make sure that you get your hands on as much information as possible. Other valuable sources for qualitative information are:

  • Personas

  • User testing

  • Focus groups

  • Staff interviews

  • Customer interviews

  • Media coverage

  • Social media

The list goes on and on.

Conversion Optimization Behavior Model

I have borrowed this model from BJ Fogg, who founded the Persuasive Technology Lab at Stanford University, and adapted it to use on the web. You can read more about the model on behaviormodel.org, but it is in essence a lense through which I analyse a web site or a web page to quickly identify its shortcomings.

The model states that for a desired behavior to occur (above the dotted “action line”), the user must be sufficiently motivated, be sufficiently able, and there must be a sufficiently powerful trigger.

  1. Motivation to perform the desired action (Persuade)

  2. Ability to perform the desired action (UX)

  3. Trigger to perform the desired action (“add to cart”, “Contact”)

Identify your test candidates

Through quantitative analysis with web analytics, you have now identified typical navigational patterns, and through the Behavior Model, you have analyzed the website and the pages involved in the navigational path towards your Target Behavior.

This should leave you with a number of pages that may be interesting to improve, and these are your Test Candidates. But how do you know which you really should improve? The answer is a process called Monetization Modeling:

Which are your pages with highest monetization potential?

If you have set up Goals in analytics, and you really ought to, and have assigned each of your macro & micro conversion with a Goal Value – or better yet implement eCommerce if you actually sell things online – then you can assign an actual value to each page.

Knowing both the value of the page (the “revenue”), the conversion rate (towards your Target Behavior) as well as Visits to the page, you can create a matrix like the illustration below:

Conversion rate monetization table

This part is not an exact science, but the goal is to guesstimate how much you are able to increase the conversion rate for each of your Test Candidates. My advice is to be conservative, though realistic. Practice doesn’t make perfect, but my experience it is now very rare that I miss the target, and when I do it’s usually me underestimating the opportunities.

Identify test finalists

What you are left with now is a prioritized list by revenue potential – these are your Test Finalists. It is up to you to decide how many pages should be optimized, or if you just want to begin from the top and work your way down.

Create your tests based on all research

In creating the tests, I make use of the Behavior model again, determining what are the shortcomings of each of the test finalists:

  1. Motivation to perform the desired action (Persuade)

  2. Ability to perform the desired action (UX)

  3. Trigger to perform the desired action (“add to cart”, “Contact”)

In this article I will not cover the UX-part – that’s an entire article on its own. Instead, I will focus on the Motivation aspect of things. But what exactly does increase Motivation?

We humans like to see ourselves as rational beings, and that we make our decisions based on facts and logic. But what happens in reality, even the most rational of us is that we use our emotions to make decisions and then justify them with logic – this is called rationalization.

The heart must be touched first before the facts are presented.

By “heart first” we mean scientifically proven principles of emotional persuasion and the importance of this can not be underestimated; If you’re able to convince the customer to make an emotional decision, he or she will automatically rationalize, thus cementing the decision. The sale is yours.

The turkey experiment

No, I haven’t lost my mind – the turkey is key to what I’m about to tell you.

Animal behaviorist M. W. Fox did a series of experiments involving a mother turkey and a polecat. For a mother turkey, the polecat is a natural enemy, and an approach by one would trigger a furious attack. Even a stuffed polecat would trigger this behavior when drawn towards the mother turkey in a string.

To quote Robert Cialdinis “Influence”

“When, however, the same stuffed replica carried inside it a small recorder that played the “cheep-cheep” sound of baby turkeys, the mother not only accepted the oncoming polecat, but gathered it underneath her. When the machine was turned off, the polecat model again drew a vicious attack.”

Yes, it sounds absolutely ridiculous, and we can all laugh at it. However, human beings have the same sort of automatic behavior that can be triggered under certain circumstances. And it does make sense – it’s actually very good for us. Imagine if we had to think hard every time we were to brush our teeth or tie our shoelaces? Or do an in depth analysis of what is proper behavior in a social setting? This is what Charles Duhigg in his book “The power of Habit” call a “Habit Loop”:

The power of habit

All we have to do is to get a Cue, identify the correct routine, Press Play on Tape, sit back, and relax:

Press play and sit back

The key to increase Motivation is therefore to use persuasion principles that make the visitor Press Play On Tape, and execute the desired behavior routine.

Even in this narrow segment of social psychology, there are a wealth of information and angles to approach the issue of increasing motivation, so for simplicity, I will only use Robert B. Cialdini’s 6 principles of persuasion. Even if you would solely rely on these 6 principles, it should produce tremendous results (Please do use responsibly. These techniques are devastatingly effective, but also help us cope with our hectic lives. Don’t ruin it by abusing it).

Below are a few examples of these 6 principles in action:

  1. Reciprocity

    1. If you give me something, I owe you

    2. “Free” sample or “free trial”

    3. We are obligated to give

    4. We are obligated to receive

    5. We are obligated to repay

  2. Commitment/Consistency

    1. Reliability is a highly valued personal trait in society

    2. If you say “A”, you must also say “B”

    3. If you’ve agreed to something, you are bound to keep your promise

    4. Ask for a small favor first, then it is much easier to ask for more later

    5. If you are defined as a loyal customer you have to behave like one

  3. Scarcity

    1. Only 3 left!

    2. Limited Time Offer!

    3. Exclusive / limited / secret information

    4. Deprivation of liberty, limited options

      1. Concorde was discontinued in 2003 – ticket sales boomed

      2. The Norwegian Sætres “Alphabet biscuits” was drawn in 2001 due to low sales, relaunched in 2007 after massive demand and petitions.

  4. Authority

    1. Perceived authority (Think the classic ad with a “doctor” in a white lab coat”)

    2. Genuine authority (often backed up by social proof)

  5. Liking

    1. Sympathetic or visually appealing

    2. Almost impossible to say no to a friend

      1. Send to a friend – recruiting campaign “gift certificates for you and your friends”

    3. We are like you, we understand you – mirroring behavior

    4. “Good cop – Bad cop” routine

  6. Social Proof

    1. Canned laughter. Everybody knows it fake, yet it works. (Imagine your least favorite sitcom without the canned laughter…)

    2. The effect of social proof is enhanced by the uncertainty (pluralistic ignorance)

      1. New or unknown brands can have extreme benefits of this principle

    3. When lots of people are doing it, it must be the right thing to do

    4. What does others say about you? Prove it!

      1. Testimonials

      2. Endorsements

      3. Expert opinions

      4. Diplomas, awards and nominations

      5. Security Certificates

Launch your test with Visual Website Optimizer

Now that you have your new page(s), it’s time to launch your test(s). Obviously, if you want to be serious about this kind of work, you do not want to use the built in features for SiteCatalyst, Google Analytics etc – they are simply much too weak. Visual Website Optimizer would be not your worst choice by far 😉

Identify your winning combinations

After a while you will reach statistical significance in your tests. Be aware, though that the algorithms calculating it may trigger too early – in fact, in the case of Allinclusive.no, it was triggered after only a day. I knew this wasn’t correct, so I let the test run for quite a bit longer, until it yet again gave a positive result for statistical significance.

And you have a conversion win!

Finally, I’d like to say that I’ve had some help with the graphics and illustrations from my fantastic work buddy Remi, one of my interaction designers, and you can download the infographic for the process here.

I truly wish this has been valuable to you, as it’s the first time I have given away (almost everything) of my process to achieve greatness with Persuasion Design.

Are you ready to win with your customers?

Comments (13)

Leave a Comment
  1. Hi Yassin, I’ve put a lot of effort into this post, so I’m really glad you liked it so much!

    Yeah, I do both coaching, speaking, workshops and consulting. Get in touch with me on LinkedIn, and we can take things from there 🙂

  2. Nice, very thorough.

    Having ran many tests similar to your allinclusive.no test, I can tell you that the improvement is not actually 53%. Your sample size is very small, and as you approach 200 conversions or so you’ll find it will go down significantly. Maybe 20% or less.

    I charge my clients on a pay-for-performance basis whereas what I charge is based on % increase in conversion, so the tests I run nowadays require me to have a large sample size because if I were to stop the test where you stopped it, the client would pay more than the actual effect on their bottom line.

    Either way, great article and I appreciate the time and effort you took to write this.

  3. Great article Sverre.

    You ought to take it up for a living.

    You might be quite good at it one day… 8^)

    Terence.

  4. Hi Sverre,
    Feels amazing to read this article. Why?

    Because I have been following a similar process in all the conversion optimization activities that I am leading! This article is a validation that my approach has been spot on!

    Thanks for sharing the article, this definitely helps me improve our company’s process for similar initiatives!

    Cheers,
    Harshil

  5. To identify the navigational context, I use a quantitative approach using web analytics. In Google Analytics, you could use the “Visitors Flow” report to identify typical navigational patterns, and thus get a better understanding on how your visitors arrive at your page.

    This and several more excerpts make this article one of the best I have ever read. Thank you for your time and effort. I am so intrigued by human psychology and hope to apply it more to our website redesigns.

  6. LOL@Lukáš – believe it or not, but I have cut, tweaked, and distilled to the very essence. I could easily have written twice this (not to mention a small book). I do realize that might would have been a little bit excessive for one blog post 😉

Leave Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes : <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Contact Us / Login

Product
Resources Home