Previously, on Added Bytes: Ten Tips for Improving Your Website Conversion Rate.
Once you've built your website, the temptation is to think of it as finished. Maybe build another one. Certainly aim to drag more visitors, kicking and screaming if necessary, to your shop. After all, if you're making 10 sales for every 1000 people to visit, then getting 2000 people to visit will mean 20 sales! Right?
Not necessarily. Were that true, however, how much will it cost to double your visitor count? For more sites, plenty. What if you, instead of doubling your visitor count, worked to double your sales from your existing customers? How about bumping it to 5%? 10%? Higher?
Think a 5% conversion rate is aiming high? The top online converter in December 2008 was turning 31.1% of visits to sales. Amazon was at 23.7% for the same period. Yes, it was Christmas, but looking at the same report for July 2008 you can see top sellers with conversion rates of 15% to 25%.
You are not Amazon. I repeat, for those of you in the cheap seats: You Are Not Amazon. Amazon is a global household brand with different stock to you. And a different market to you. Something that works for Amazon might work for you. But just because something works for Amazon does not guarantee that it will work for you.
Be aware of what your audience wants. The odds are they are telling you what they want. Do you have a customer support line? Listen to it. If all your calls start with "I tried to order online, but ..." then you have a problem with your site that your customers are actually taking the time to tell you about. How many potential customers who had the same problem simply went to a competitor instead of phoning?
Even if you don't have website problems (and if your conversion rate is 1% to 2%, I'd be shocked if you didn't), that phone line is a window into the requirements of your customers. If they are phoning to find out what it costs to have your blue widget delivered, or to tell you that your site is too slow, or to ask whether the green trousers come with a belt, they are telling you what they want your site to do for them.
Most people are terrified of failure, or more specifically of being seen to have failed. Often, even when there is a great potential reward, people will turn down a reasonable idea - just because they are afraid of how people's perceptions of them will change if they are wrong.
I have known website owners who have refused to experiment with ideas because they didn't want to risk reducing their conversion rate (of 1%). I've known others who've refused to experiment with one part of a site because last time it was changed the conversion rate improved (therefore, presumably, changing that same bit in any way could only result in a decrease in conversions).
In the same vein, the sunk cost fallacy is a common cause of low conversion rates. Once someone has made the decision to do something with a site, you will find that person resistive to change. This is especially true where they have paid for the implementation of the change. It is even true when it can be demonstrated that the change in question is directly responsible for a drop in the site conversion rate.
Most people are remarkably blinkered to these sorts of issues - you may even be suffering from one of these blind spots yourself.
Learn to experiment, and learn to let things go - you learn from your mistakes, not your successes. And don't be afraid to try the same reasonable ideas that once failed again at a later date. You will rarely know precisely why an idea failed to improve a conversion rate, but an idea that doesn't work one year may be spectacularly successful the next.
It is extremely difficult to predict accurately what changes will have a significant positive effect. So many factors work together, and there are so many differences between users of one site compared to another, that experience can often work against you when you're trying to improve a conversion rate.
Let me take you on a journey through Site A (an online shop I had the pleasure of advising). The owner of Site A had changed his product listing. He used to have ten items per page, and he increased it to 20. The extra products listed on each page gave the users more choice, but his conversion rate didn't change. Next, he added larger product imagery to the listing - unfortunately this slowed the site down considerably, and his conversion rate went down 10%. Finally, he split his categories up, as they were getting a bit crowded and unwieldy. After this last change, his conversion rate increased back to its original level.
After three rounds of changes, the owner of Site A had the same conversion rate for his site, and viewed the only positive change as the last one, change 3. He attributed it to better organisation of products. And here we have the problem - he was wrong. The positive change was actually change 2.
I'm going to run through this slowly, as this is counter-intuitive - after all, how can the change with the measured negative effect actually be positive? The reason the change was actually positive is that these things work together - they are not independent.
The first change to Site A was neutral. When the second change was made, 20 extra images per page were being loaded. These images slowed the site to a crawl. When the third change was made, several categories contained fewer products than before - around half of his categories had 10 products or less - and these new, smaller category pages loaded faster. Much faster. These smaller category pages were much faster to load, and the extra imagery did a good job of helping to sell to the users. The increase in sales came from these pages, after the third change, came from these smaller categories.
After change 3, Site A still has the same conversion rate as it did before starting. But once the changes were correctly identified as positive (change 2), negative (change 1) and neutral (change 3), it is a simple matter of reversing the negative. So, the new categories were kept, the images were kept, and the number of items per page was reduced back down to 10. The final result of all of this was a positive change to the conversion rate.
And the moral of the story - it is very easy, in hindsight, to attribute positive and negative effects to changes, but it is extremely difficult to do so accurately and correctly. Examine the effects of changes and, if you can, isolate and test individual effects.
Many people make the mistake of only tracking small pieces of information. Or tracking everything but only looking at a small part of the data. Data is your friend, and you need to understand what the different metrics available to you are actually measures of if you want to have any clue what effect your work is actually having.
Most important, of course, when looking at the conversion rate, is the volume of sales. Some people only look at that number. But other numbers can tell you about how useful your site is throughout the shopping process.
The percentage of repeat visitors tells you something, about whether you are engaging visitors early in the purchasing process and bringing them back for the sale. A high bounce rate indicates pages that are failing to deliver on their promises. Add to cart rates, cart abandonment rates, login vs registration vs abandonment rates, product removal (from cart) rates - all of these will identify areas of your checkout and purchase process that are underperforming, or improving.
There's no better way to ensure an unchanging conversion rate than doing nothing. Don't wait - start today. The more you experiment, the more you will learn and the more opportunities you give yourself to improve your numbers. Get in the habit of testing and tracking constantly. There's always room for improvement.
How many sites have a nice piece of prime real estate at the top of the page simply saying "your cart is empty"? Too many. The user usually knows their cart is empty. They've not added anything to it. Why occupy a prime sales spot with worthless information? You could hide the cart, or link to a "getting started" article - an introduction to your products, or recommendations.
A cart is designed to catch the eye - but before something is added to the cart, this is an unwelcome distraction from other, more relevant, calls to action. You want your users clicking through to products and then adding those products to their baskets - and you don't want to distract them in the meantime with an empty cart space or a useless "checkout" button.
When I worked with Juno Records in London, one of my most startling discoveries was that far from being obsessed with speed, some of the users of the site would spend a day - or more - building a cart of products.
There was a good reason for this - the shopping page of the site at the time was several megabytes - but users would happily load the page, and browse the mammoth list of products. They could listen to snippets of tracks, and they would browse the entire list - top to bottom - listening to snippets and adding things to their basket.
One of the first versions of the new Juno site had a cart on a separate page to the products (as is common for many ecommerce sites). When you added a new item to your cart, it took you away from the page you were on. Given that users had spent time at this point browsing through the product list, and by moving them to a new page you lost their place on that product list, this was not a usability improvement.
Likewise, some users would leave their cart half-built and return to it the next day. Or several days later. The default session timeout on one early version "broke" the site for those users.
One trick to avoiding problems like this is first to know your users. Their habits may not be what you expect.
But more importantly than that, think "unobtrusive convenience". Make the user's life easier, but don't make them work unnecessarily for that convenience.
For example, save the user's cart for them, but don't force them to create an account to do so - cookies are your friend. The user should be able to leave, and then come back later and continue shopping. Check that moving users to a new page when they add an item to their cart doesn't disrupt their shopping. Don't force users to create accounts when they want to buy - offer them the option but don't let it get in their way if they're not interested.
Discount codes and coupons can be a great way to drive sales. They can also backfire, making customers who would otherwise buy feel like they're not getting a good deal, or missing out on a saving. This isn't good for business.
There are several solutions to this, but ideally you should look to downplay discount code redemption in the checkout process. One good trick is to ensure everyone has a discount, even if it's not a great one. Another is to move code redemption into the account system instead of the checkout system. Yet another is to identify users with discounts (if you can) when they log in and then offer to apply the code if they have one available.
Pricing is tricky for almost every industry. Price yourself too high and you'll lose out to cheaper competitors. Price yourself too low and you'll end up appearing cheap. Price some products high and others low and you'll look like you're pushing some brands over others, or worse that you're expensive for some products (which tends to make people think you're expensive for all products, even if you're not).
It's important to be clear with your pricing - if you're aiming to be the cheapest then be the cheapest. Offer price comparisons or price matching. Tout your prices above all else. If you're going for higher prices and selling quality of service, don't shout about your expensive prices.
This also applies to shipping. Delivery costs are almost always an additional expense incurred during the checkout process. If you add delivery costs during the checkout process, then exactly when you don't want people reconsidering their purchase, you're giving them a reason to. Make delivery free or flat rate and add it from the start of the shopping process. There's no harm in reducing delivery cost - and telling the customer you've done so - during the checkout process.
One excellent (and practical) way to increase your website conversion rate is to add consumer reviews to your store. They are a proven way to increase sales, and they have an excellent positive effect on your search engine optimisation work. A service like FeedbackFair will give your reviews extra credibility.