Not all traffic is created equal – a guide to ecommerce

While recently preparing a report on the annual traffic to a large corporate website, I had many opportunities to pause and reflect on the nature of Internet traffic and the difficult job businesses have in procuring this traffic, holding on to it, channelling it in the right ways, and turning browsing into the business that will help the company increase profits and improve its bottom line.

A successful online business transaction has so many variables — some technological, but most human, that make the process of conversion less than an exact science. Web metrics have gone some way towards providing answers for measurables like click-through rates, attrition rates, cost per acquisition and return-on-investment (ROI), but they tend to address the what of browsing behaviour, and not the why.

The understanding of real user behaviour on websites is often based on broad assumptions made from analytics and user feedback, or information from small usability testing groups that could never hope to replicate the nature and complexity of traffic coming to the site.

I often draw the analogy of trying to understand the myriad of ways that shoppers interact with a bricks and mortar shopping mall. Could one possibly track in which entrance they came, the paths they followed from store to store, where they window shopped, paused to interact with promotional material, looked at products they didn’t buy, the time they spent on each activity, money spent or frequency of return visits? It’s virtually impossible. The reality is that the number of permutations are endless, and this reality also applies to web shoppers in many instances.

We also need to understand that the more sophisticated and intangible your product and brand experience is, the more ways people will find to interact with it online. Fast moving consumer goods have simpler paths to conversion and higher conversion rates than products like training courses, retirement annuities or online legal advice. Your job as an online business architect is to understand the ways that your products, the interface and your audience work together most effectively, and optimise the entire experience to take advantage of that.

So what should you be thinking about when trying to understand your traffic and the way it converts online?

Measure with analytics

If you can’t measure it, you can’t manage it. Although many webmasters have analytics solutions in place, it’s often the reporting and interpretation of these analytics that don’t go far enough in adding any real business value. It is no longer enough to be looking at the number of visits, unique visitors,  page impressions or form submissions.

With a little learning, Google Analytic’s (free) Advanced Segmentation tools allow you to dig deeper into user behaviour by segmenting user types based on diverse variables ranging from geographic location to entrance page. It’s easy to build up multiple segments and test these against your conversion events to see which user groups are your most profitable. Thinking about these groups allows you to take programmatic steps to recognise users from those groups when they enter your site, and direct them more swiftly to conversion scenarios.

Analytics can also tell you about bounce rates, page depth, time on the site and entrance and exit pages, all of which provide important clues to how users are engaging, or not engaging, with your content and products.

Another useful tool in Google Analytics is the Goals and Funnels feature. This allows you to configure up to ten pages in a defined conversion path through the site, as well as to assign a monetary value to the completion of each goal. The interface provides clear representations of attrition rates through page sequences and can assist greatly in optimising pages and also your calculations of cost per acquisition or ROI.

If you are running paid display or search campaigns, event and campaign tracking will give valuable insights into the performance of particular ads or creative executions and how they convert. You should be making retention decisions about these ads as your campaigns progress, but even if your team is not agile enough to juggle creative material quickly during shorter campaigns, the data can be analysed afterwards to optimise your next one.

Using and understanding analytics can be time consuming and resource intensive, but there is no doubt about the value these metrics can bring if they are used consistently in providing ongoing feedback into design and execution.

Understanding types of traffic

The analytics model generally groups traffic into four categories – organic, cpc, referral and direct. Again, you can use Advanced Segmentation to look at these groups and see which is converting most efficiently. Generally it will be the cpc and the organic traffic because it is the most targeted. But you may also be surprised by the way that referral and direct traffic interacts.

People may arrive by word of mouth or come from partner sites and be just as willing to make a purchase. The important take-out here is to try understand whether it is more cost effective to channel traffic purchasing budgets into paid search, SEO, affiliate programs or offline initiatives. Doing your sums in this area could make a big difference to the bottom line in the longer term.

A number of other traffic models have made useful distinctions between user types to help us understand that not every browser is a potential customer. Some of these user types include:

The newbie shopper – they are still trying to grasp the concept of ecommerce and mainly use the internet for researching purchases. They are likely to make low value purchases in safe categories and prefer a simple interface and easy checkout process.

The cautious shopper – is nervous about security and privacy issues and probably needs online customer support to ease their fears. They also mainly research purchases online.

The bargain hunter – has little brand loyalty and knows the tools to help them find the best deal. These shoppers put the time in and will respond to special offers and deals if they feel they have found the lowest price.

The decisive shopper – these folk know exactly what they want before going online. Typically they will have a list of criteria to be matched and will seek out user reviews and real time customer service from knowledgeable sales consultants.

The recreational shopper – every business should want to capture these. They shop for fun, are adventurous and purchase frequently. What they need are interesting tools, creative presentation and customer feedback via reviews or forums.

The power shopper – shops out of necessity, doesn’t waste time looking around and develops sophisticated strategies for finding what they want. These shoppers want relevant information, expert opinions and support to help them make decisions quickly.

Think about these different shoppers and what their needs are, and try to formulate enticement and fulfilment mechanisms both in online advertising, and on the interface, to improve conversion on your site.

Interface, interface, interface

So much has been written on design, usability and web interfaces in the past few years, trying to find a way in through all the noise can be a daunting prospect especially when starting out. You may be lucky enough to employ specialists in these areas who will continually optimise your pages for conversion based on analytics feedback and other mined data. But if you are not, there are a few areas in which you should concentrate your energies.

One, understand shopping patterns and types and try to design for all of those on the page. FutureNow has identified four buying personas – spontaneous, competitive, humanistic, and methodical. These personas all shop in different ways and respond to different cues on the page.

Below is a great little video on this subject, that encapsulated all I couldn’t find after many hours of searching:

Two, check out the latest eye tracking studies from designers doing work in the area. Study articles like 23 Actionable Lessons From Eye-Tracking Studies . This is really not rocket science. People do spend more time looking at and clicking on certain places and elements on the page. Design around this to improve the chances of users seeing and clicking on your conversion areas.

Three, do some A/B roll testing. Yes, everyone talks about it but no one has met anyone who actually does it. It really does work, different designs will have higher clickthrough success rates and this could make a big difference to your bottom line. You need to find the best designs for your most important landing and conversion pages. Sites like Unbounce and SplitTesting allow you to get your feet wet and learn for a little investment of time or money.

Four, keep your product presentation simple and consistent. Use good quality images that are large enough to see necessary details, keep written descriptions concise and have prices on your summary and details pages for easy decision making. Display your view cart and checkout links prominently and in the same place on every page. Offer search if you have a large number of products.

Five, keep your checkout process simple and down to the minimum number of steps. The more complex the checkout, the more chance users have of picking up technical problems. Examine the number of details you are asking for and also the navigation – is it user-friendly, and is it easy to add and remove items?

Are you on form?

If you are capturing information for leads purposes on web forms, you’ll need to get up to speed on the latest thinking in this area. Form design on web pages has become a science. Nowhere is the pain of leads attrition felt more acutely than on this simple input interface we see and interact with every day. Although users are slowly accepting them as a fact of life, most still hate filling in forms. They are demanding, ambivalent and labour intensive, often with uncertain outcomes for the user.

Crucial in your approach to forms is understanding how much information is enough, how much is too much to ask for, and where the tipping point is. Also, try to design your validation intelligently and clearly and make sure it is working one hundred percent. If you have to ask for a lot of information, try splitting your form into several panels separated by tabs or a next button. Group information logically, starting with contact details, and capture to your database at each step along the way. Then if the form is abandoned, at least you can follow up with a phone call or email from the details you have captured. Be sure you capture each form step as an individual event or impression in Analytics so you can monitor where abandonment is happening and try to optimise those areas for better retention.

Some of the newer ideas around form design propose capturing information in the body of the page text using open spaces and more friendly narratives. Although there is no hard evidence around the success rate of this approach, the principle of experimentation is an important one to follow. Small changes in design can make big differences to form completion rates. There have also been some very useful eye-tracking studies done on forms. Spend some time on the article Webform design guidelines: an eyetracking study for insights and discussion. You’ll be amazed at what you never knew.

If, after all your efforts, completion rates are still depressingly low, one must simply blame the user. With attention spans what they are these days, it’s quite possible that your form was abandoned for many reasons other than poor usability. Your potential customer may have been distracted by any number of environmental or web-based factors.

Lets face it, at the time Lady Gaga’s latest YouTube video or another skinny latte may have been infinitely more appealing than your cutlery set, funeral plan or training video, no matter how well priced or enticingly presented. Users may not have the necessary information at hand, get stuck with a slow connection or suddenly need to run an errand. They may or may not come back later. That’s retail for you.

Still feeling abandoned?

And if getting all this right is not enough of a challenge for you as an ecommerce business architect, online shoppers present many other challenges. Studies on cart abandonment have shown that the devil may often lie in the detail. Apart from obvious abandonment reasons like the price being too high, the shopper running out of time, or unsuitable payment mechanisms, other top reasons for abandoning a purchase may include:

• Not being able to see shipping charges before registering
• Too much personal detail required
• Perceptions of an unstable or unreliable interface
• Lack of visual progress indicators
• Poor form field labelling
• Inadequate product filtering options e.g. colour or size
• Inefficient customer support

So you need to throw these considerations into the mix as well when planning your processes and workflows.

Does more traffic always mean more sales?

This question has always been the source of much debate in ecommerce circles. There is the argument  that conversion is just a numbers game as it was in the days of direct mail, or  still is in the world of telesales. You start with a certain number of prospects and through predictable attrition at each step of the way, you will land up with a certain percentage of sales, and, that the conversion rates will be range bound no matter which way you do it.

The truth rather seems to be that simply increasing traffic volume will not necessarily increase sales proportionately. It’s a much more complex equation than that.

Ultimately it’s the quality of the traffic that counts and it’s not always easy to understand the difference between someone who reached your site from an online ad, a blog link, from word of mouth or an organic search, and what type of shopper they are. But the more you think, test, measure and refine, the closer you will come to understanding your traffic and turning it into profitable business.