Checkout optimization expert Dr Mike Baxter has produced his fourth stunning piece of e-commerce research for Econsultancy. The Checkout Optimization Guide is a must-read for all companies that accept payments online. We’re absolutely sure that it will increase your conversion rates.

In the report Mike suggested 70 ways to improve performance, and I caught up with him to talk more about the latest trends in e-commerce checkout design.

The new checkout report has been described as a workshop manual. Can you explain this? 

Fixing an e-commerce checkout is complicated. There are lots of things that can go wrong and many can go wrong in several different ways, just like a car.

So, one of the problems e-commerce managers have to overcome is where to start in improving their checkout. They need to ask themselves:

  • Is anything broken? 
  • If it is how can I tell?  
  • Once I’m sure it is broken, how do I fix it?
  • Even if nothing is actually broken, which bits of checkout should I tune-up and which should I leave alone? 

A workshop manual takes you though a systematic process for identifying problems, diagnosing the cause and then finding the cure. It seemed to me this was what was most needed for helping improve e-commerce checkout

How does the new guide differ from the last Checkout Special? What does the new report bring to the table? 

The last checkout report, published in 2007, was the third e-commerce benchmarking report I’d written for Econsultancy and it was the hardest by far.

They key difficulty in benchmarking e-commerce was that there was no framework for breaking down checkout into nice manageable chunks that could be analysed, measured and compared across sites.

The 2007 report created that framework and I was pleased with the outcome and what we’d managed to achieve. It was, however, such a big job that I didn’t have time to take a step back from the report to think through how a busy e-commerce manager might actually use the report.

If they sat down and read it from cover to cover, they got a lot out of it – and I’ve had lots of people saying how much value they got from it. But, considering it was a report about usability, the report itself wasn’t as useable as I’d have liked it to be!

So, this new report is written, from the first page to be a practical and accessible guide to improving e-commerce checkout.

The other main difference between the previous report and this one is that checkout design has moved on quite a lot in the past two years. So there is a load of new material in the new report that didn’t even get a mention in the previous report.

What has been happening in terms of checkout design since the last report? 

The big change is probably analytics. Every site I visited during the research for this report had at least one web analytics solution logging checkout performance. The extent to which this data is actually being used for insights remains uncertain but at least we have reached the stage where e-commerce managers can find out how well their checkout is working.

In terms of customer experience, there were no huge revolutionary trends emerging, and I wouldn’t really expect there to be. A good customer experience during checkout is all about getting the little things right – back to the workshop manual idea.

There were, however, a couple of interesting details that I feel could make a big impact over the next couple of years. One of these was auto-login for returning visitors.

The bad old days of required registration are largely (although not completely) behind us. Even without registration, however, we still want to make checkout as smooth as possible for all customers and that means not requiring visitors to type in information about themselves that we already know.

So, a few sites are starting to use auto-login. If you have bought before, you can be recognised via a cookie and your name and address details can be pre-populated. If you have had goods delivered to more than one address previously, these can be offered as options.

Even if you haven’t bought before but click through from a marketing email, at least your name and email address can be pre-populated.

These procedures raise some interesting security issues and need to be thought through carefully before being implemented but they can make a significant improvement to the customer experience.

You mention in the report that checkout abandonment is getting worse. Why do you think this is?

This was the biggest surprise of the whole report. Based on pretty strong data from around 300 etailers in the UK, the abandonment rate from checkout actually deteriorated at a rate of 0.1% per month over the past 23 months. This was a statistically significant result that we can put a lot of confidence in.

But it is pretty shocking. As to why it is happening, I suggest two possible reasons in the report. The first is that customer’s expectations are rising – there is evidence from other research, for example, that online shoppers are becoming substantially less tolerant of slow sites. Maybe they are also becoming less tolerant of poor checkouts.

The second reason, though, is that retailers aren’t keeping up with best practice. While a few checkouts are an absolute joy, most have got minor glitches and annoyances and a few are so confusing or hard to use, I’m surprised they ever sell anything.

Can you explain the three steps to an immediate boost in profits? 

The three steps are: 

Firstly, get your web analytics working properly. This may sound obvious but in all the analytics consultancy I do it is rare to find checkout analytics implemented to best practice.

A couple of common mistakes are having a single goal funnel but two or more paths through checkout; one for logged in returning customers and another for new customers.

Another common mistake is not having product categories logged properly, and this prevents you seeing if, for example, visitors are abandoning checkout more often with big-ticket items in the basket.

Step two is to work through the checkout checklist in the report; we have compiled a list of 70 issues that can be checked and even scored against a competitor to evaluate your current checkout process.

The third step is to make simple changes immediately and specify harder changes for the next site update.

For anyone who hasn’t done any optimisation of their checkout, going through these three steps WILL make real improvements in checkout performance and they shouldn’t take long nor cost a fortune to implement.

What can firms do to the checkout process to attract more repeat business? 

One we’ve already talked about is making checkout for returning visitors easier by remembering their details using auto-login.

The other one that few e-commerce sites are using well is return-purchase promotions. The whole issue of promotions during the checkout is a controversial one and it is something that I advise caution with.  

When a customer gets to checkout they just want to buy and leave. Distracting them with temping offers at this stage can be a recipe for increased abandonment. 

As they finish the checkout, however, can be a great time to tempt them back for a quick repeat purchase. If your e-commerce system is smart enough you can tailor the promotion to suit their browsing, add-to-basket and purchased products. The promotion can be a time-limited discount, free delivery or simply a ‘you might also be interested in …’

This should then be followed up consistently in the confirmation email, the dispatch notification email and possibly a follow-up ‘review your purchase’ email.

But when purchase-recency is one of the most powerful determinants of customer value, it is madness not to tempt customers back for a repeat purchase as soon as they have finished the current purchase.

I’ve noticed a trend towards one-page checkouts – what are the advantages and disadvantages of this approach?

The trade-off is simple in principle but hard to weigh up in practice.

Is it better to have one page, but risk the customer thinking this page looks busy or complicated or hard to fill in? Or is it better to have more pages, each of which looks simpler and easy to fill in?  In the old days of e-commerce, the time taken to load new pages tended to shift the balance in favour of one-page checkouts.

Now this isn’t an issue.  Dynamic forms, with page elements hidden and revealed using javascript, on the other hand, can make one-page checkouts look much less complicated. So much so, that it sometimes takes a bit of effort to work out if a checkout is actually using a one-page format or not.

In which case, I feel that the whole one-page / multi-page checkout debate is a bit of a red herring in 2010. The rules for good customer experience are the same:

  • Keep the whole checkout process as simple as possible, only asking for information you need and never asking the customer to enter the same information twice;
  • Don’t make any form look complicated and hard to fill in;
  • When one bit of checkout is complete make it clear that progress has been made and the end is getting close;
  • …and make sure the checkout doesn’t break when the browser back button is used.

What are the most common errors that etailers are making in checkout design?

It is details, details, details! Get the most awkward, fastidious, picky, detail-obsessed person in the office to go through your checkout and write down EVERYTHING they find wrong with it.

Then give them the checkout report and get them to do it again with the 70 point checklist! Every small detail that annoys your customer pushes them one step closer to abandonment.

Depending how good the rest of your site is, you may not have that much goodwill left in your customer by the time they even start the checkout.