Bryan Eisenberg is the co-founder of Future Now and the creator of some devilishly clever techniques to help web businesses improve their conversion rates.
Along with his brother Jeffrey, Bryan co-authored the bestsellers 'Waiting For Your Cat to Bark?' and 'Call to Action'. He also publishes the GrokDotCom blog and was voted one of the world’s top 10 usability gurus in our recent User Experience Report.
Here, he talks about the ancient Greek philosophy behind persuasion and suggests why the roots of web design can be traced back to the fifth century BC. Oh yes...
Can you sum up your thinking about persuasion architecture and how that can affect abandonment rates?
The main challenge of abandonment is that people don’t feel confident enough to take an action. ComScore, for example, recently published a study that shows form abandonment doesn’t occur at the form and checkout abandonment doesn’t happen at the checkout.
If you have a 72% abandonment rate on your form, it’s not because the form is bad. Form abandonment tends to happen well before people reach it – when they are on product pages.
If you look at the hundreds of units people have on their website, you will see abandonment tends to rise for the more highly-seen product pages, because the category pages - whose responsibility it is to make sure customers find the right product - aren’t doing their job.
People don’t have confidence that they are getting all the answers they need and you can’t expect people to take an action when they aren’t confident.
Marketing in the past used to be about grabbing people’s attention – disrupting them - while they were passive. That’s one way of marketing and it won’t go away.
Yahoo! did a great study that showed the most popular activity people did while watching TV was sleeping. But you will never find someone sleeping while they are clicking on a website. They are on a mission and are investigating a scent to fulfil a need or desire, and we don’t leave them the right choices to click on.
It’s the same problem – which we are seeing with a whole bunch of companies right now – that happens when you have been testing and optimising a landing page for a couple of years and can’t budge the conversion rate. That’s because there are false assumptions.
One is that customers are ready to buy, fill out a form or download a whitepaper right now. People may or may not take an action right now.
Also, some companies look at the web from a branding or direct marketing perspective, but not everyone is average and at the same time, you can’t target everybody on an individual level because you don’t have enough data to do that.
Customers are experiencing your web page from 360 degrees and coming in from their own angles of approach. This is a self-service medium.
People make decisions for different reasons. We know that from Hippocrates, who was the first to describe four different personality-types. We call them perspectives: competitive, spontaneous, humanistic and methodical.
If you’re in a retail environment, you shouldn’t expect to sell to everyone in the exact same way. Any sales manager will tell you that you can’t do that.
Can you provide any real-life examples of how this can produce results?
We’ve based the process of persuasion architecture on three very simple questions. They’re so simple, people often pass over them. It’s like in basketball, where the foul shot is the most important shot – not the slam dunk or 360 or whatever. It’s the fundamentals that matter.
You need to ask which of those four 'perspectives' you want to take action and what action you want them to take? You need to plan their next click. You need to look at any page on your site and pick any sentence or hyperlink, and think about "for whom they are meant and why they should click on them?". The third question is: "What information could you provide to make that person take that action?"
By practising the basics of this on every single page, you develop an eye towards it. A perfect example is Overstock.com, where with one image change we were able to account for $25m in revenue – 5% of their overall margin.
What was happening was people would get to a movie category page, but the site would lose 92% of that traffic. When you think about it, what is so hard about selling movies? They aren’t complicated, expensive or a hugely considered purchase. So we did a simple test and thought about each perspective and tried to add a ‘contract’ for each one on that page.
With competitive people, who know exactly what they are looking for, we looked at the search box and there was an image next to it saying ‘Kids titles for learning and fun’. The competitive person was thinking ‘hey, I don’t want to search for kids’ titles’.
So we changed the image and 30% of the abandonment disappeared - $25m in annualised revenue.
It’s not magic – you just need to ask the simple questions.
We cheat. I like to say ‘cheat with process’. I live on a peninsula, with the ocean on one side and the bay on the other, and in the evenings an artist comes down and puts down paintings to sell.
All he is doing is fancy 'painting by numbers', copying paintings that people will have seen elsewhere, although he has a couple of his own works. He sells quite a few.
We tell people they can do the same thing – if you have a great talent in the organisation like a van Gogh, don’t use process, but if they leave, don’t complain. If you don’t have that, cheat by using a process.
Ultimately, do you expect networks to aggregate and sell profiles of users based on the perspectives they appear to have?
I think aggregation services will be a challenge because people’s cravings for privacy keep showing up religiously, so that will be difficult to do on a global scale.
But people’s keywords leave a tremendous amount of trails that indicate what kind of persona they are coming from. The person that types in ‘perfect diamond’ is not the same person that types in ‘diamond certification’.
We have built some software that we have plans to release early next year, that will take some of the software we use internally and open it up to the market.
When we are writing copy or images, we tag the image with who the link is supposed to be for – 60% competitive and 40% methodical, for example. We are working with web analytics vendors to plug it into web analytics so it can be reported as such. We’ll have more to report about how that integration works.
Do you think free tools like Google Analytics are encouraging people to look at the traffic on their site in the right way?
A good friend of mine, Avinash Kaushik, just wrote a wonderful post about data mining on the web - Data Mining And Predictive Analytics On Web Data Works? Nyet!
It’s true. What people are essentially doing with these great tools is trying to read tea leaves.
Behaviour on the web is influenced by the scent or lack of scent people experience on a page. That’s the reason 80% of the traffic on every site drops off in the first three clicks.
Every diagram I have seen has that quick drop-off and then the tail. The people that stick around longer are the ones that persevere or have some other knowledge that leads them to keep buying.
What happens is that the people that only get to the second or third click only clicked on what seemed likely to be the ‘least-worst’ link. That’s not a way to gather intelligence.
It’s the same in supermarkets and big, fancy retail outlets – they are very careful to create these flows of how people navigate around a store.
They can monitor with the cameras if the flow is moving correctly – Paco Underhill, in Why We Buy: The Science of Shopping, discovered this butt-brush effect where people were bumping into each other by the tie rack in Bloomingdales.
People are hitting these butt-brush effects on the web - without a big picture of how traffic is supposed to flow, it’s difficult to make judgements about whether it is working or not.
What we are trying to do is teach people to do analysis without the analysts. That’s a big issue because there aren’t many analysts out there.
How does that affect the KPIs online retailers should be monitoring?
It depends on the type of site but what we have started to do is get companies to focus on very simple KPIs.
Focus first on the tasks you want people to accomplish – for example, how many see a buying guide, then how many add it to their product wish-list, then how many purchase it. Then you can start looking at how you can get more people to add it to their wish-list.
If you focus on how people move through the buying phase rather than getting people to complete the transaction right away, you will start seeing results.
It’s not really about reports – it’s more of a holistic view. That’s where a lot of analytics fails today. Most analytics tools today are fancy hit counters. That’s why they are having issues around Web 2.0 and Ajax.
You have to plan what is going to happen up front, and then it is much easier to measure what you set out to accomplish.
There is a handful of analytics tools out there that focus on the visitor first and that’s the direction we need to go back to. We should never forget that clicks are people.
Too many people look at these big numbers of visitors and forget that these are their neighbour, cousin, father, daughter and so on, and went away dissatisfied.
Big numbers are hard for us to keep in our heads – the perfect example is if Chris in your office started looking faint or weak, everybody would come over and offer him a candy bar or drink and be very concerned. But then it’s hard to imagine that there are hundreds of thousands of people starving in Darfur.
Our minds are made up not to think about these big numbers. It’s the way the human operating system works, unfortunately.
Persuasion architecture, besides being focused on the results, metrics and so on, is focused on empathy - understanding what people need and want and how to take care of them properly.
How many start-ups are you dealing with nowadays? How many web companies are thinking about persuasion architecture from day one of the design process?
It’s interesting. What we’ve found is, in most cases, companies that are VC-backed tend not to have the discipline to do things like persuasion architecture. They are too impatient – they can take baby steps but they can’t take it as a whole discipline.
One of our favourite stories is we have a client that was featured at number 54 in the Inc 500 – the fastest growing companies. They have grown 2,250% over the last three years, and have been working with us for the last four years. They will directly attribute that to a lot of the work we’ve done.
The CEO recognised that this was an issue and focused on executing on conversion and customers. Focusing on your business goals and your customer goals is often counter-intuitive. We spent a good six or seven months doing the research and going through the content, and are starting to watch how traffic moves through their site.
Related stories: Interview with persuasion guru BJ Fogg