Speak To The Dog, About What Matters To The Dog, In The Language Of The Dog
This article is a transcript of my answers to questions posed by Chris Lake of e-consultancy.com from December 2005. Hopefully you will find the grilling I got useful!
What are the most important factors that influence conversion rates?
A joint of roast beef is sizzling over an open fire on a glorious summer day. The aroma fills the air as you cut the juicy meat into generous slices and stack them on a plate to pass around your friends. Your pet dog, driven crazy by the smell starts begging, whining and running around excitedly, hoping for a piece of the delicious steak.
"Speak to the dog, about what matters to the dog in the language of the dog"
Jeff Eisenberg - Call to action
What Jeff doesn't mean is start barking at people.
What he means is write for your target audience, about what it is they want, in a simple and attractive way. I'm sure that you're now thinking about roast beef. That's what you need to do when writing for your website, try to put a picture in your prospects head. The most important factors influencing conversion rates are your copy and content, your copy attracts, persuades and provides momentum, while your content answers all your prospects questions. Design, usability, measurement and testing should all be taken into account but they come after you figure out what it is you want to say and to whom you want to say it.
Where do you normally start when working with a new client? Is there a set process you follow?
Yes. We start by asking lots of questions. Initially we get the basic business information so we can ascertain how deeply the client has thought about their website as a business. You'd be surprised how many companies don't know what their cost per visit is or their profit per website visitor. So we develop a business objective so that we can say, 'this is where you are now'.
Next we'll do an analysis of how far we think we can take them, based on their own limitations and budget. Then we can say what service they might need to take them to the next level. Not all businesses need the same things, some need tweaks to marketing strategies, some need better copy and content so it's hard to answer. But the process we use is the same to find out what the client really needs.
How important is it to use web analytics tools to find on-site problems?
You can't prove whether anything works on a website without web analytics tools. I advise every client I've ever worked with to get web analytics. I won't work with an enterprise level company if they haven't got web analytics. The tools allow you to do 2 things very well, pinpoint problems and measure your tests. Without this capability you're pretty much guessing.
Can you suggest some generic KPIs that e-commerce teams should be monitoring?
There are lots and you should decide upon metrics which you can act upon. However here are some you might want to think about adopting;
· Site-wide conversion rate
· Product conversion rates
· Percentage of new and returning visitors
· Sales per visitor
· Average order value
· Average number of items purchased
· Shopping cart abandonment rate (step conversions)
· Revenue and Profit per product
· Repeat order rate, to help calculate long term value
· Cost per visit
· Profit per website visit
Also don't forget to set trip wires like page views per visit. I've explained it better about in this article; How Web Analytics Found A Million Dollar Hole
Is Google Analytics going to be good or bad for the industry?
Basically yes I believe it is going to be good for the industry. I'm delighted about it. Rather than write another 1000 words on the subject though; Read this article
Too many paid-search and affiliate programmes do not create campaign-specific landing pages. What are the key elements that should be found on a landing page?
Compelling copy and content. Complimentary design, a graphic should help people to see the text or reinforce the message not just be there for the sake of it. One link from the landing page - the call to action, IE the buy now button. A landing page should already attract the audience who want what you have so you needn't persuade them to go elsewhere by giving them any other links or options. Nick Usborne wrote a nice article about this;
How To Write A Landing Page
Do you have any metrics to share about landing pages, eg before / after conversion rates?
On average e-commerce landing pages get about 2-2.5% conversions. After we've worked on changes we tend to improve things by 40-50%. So usually 2.8-3.3% convert after the changes to a single call to action landing page. Of course it depends on the type of business and the traffic source. Banner ads are always lower than PPC ads for instance.
You're a big fan of A/B testing. How can web teams implement these tests and what should they be testing?
The basic principle is that you write two test pages, direct equal traffic to each and see which works better. It's possible to write your own traffic splitter code if you are only testing a few things at a time and you have in house programmers. Or you could outsource the testing to something like Offermatica which handles the test pages and the traffic splitting.
Things you might want to test using an A/B split are headlines, copy blocks, graphical images, banner ads, PPC ads, button colours, in fact anything where you have the potential to improve the response and be able to measure the improvements.
Do you have an idea of conversion metrics by sector? Which sectors are leading the way in terms of conversion rates? Which ones are underperforming?
We've compiled a list of averages by sector from a variety of different sources (including e-consultancy.com) which you can see here; What Is An Average Conversion rate?
No-one is really sure how accurate these figures are. They seem reasonable based on the top e-tailers who hit global conversion rates of between 11-16% (we're talking about online retailers like Amazon, Ebay and QVC)
The sectors which have always done the best are the catalogue companies and the average figure suggests a 6.1% conversion rate. In our experience this seems about right. I have worked up conversion to those levels with a number of catalogue companies that were hitting lower than 6% before and some have come to us with slightly higher conversion rates than 6% initially.
The underperformers seem to me to be sport/outdoor retailers or fashion and apparel. They really should be hitting much higher levels than the average level 1.4-2.2%.
Creating web pages is a balancing act between the needs of the business and the needs of the user. Factor in the needs of the search engines and that's quite a challenge. Where should your priorities lie?
The needs of the user are the most important. The business need in most cases is simply to make or save money. Businesses should achieve both their own business objective while catering for the user so that she can accomplish her goals. The more that the she accomplishes the better it should be for your business. Search engine marketing is important but you should never sacrifice copy or content simply to rank highly on Google. The search optimization should and can be implemented, but it should still persuade the person reading the copy that your solution is for them. Yes it is a balancing act but if you prioritize with the users in all cases then you're on the right track.
How can accessibility and usability play a part in increasing conversion rates?
Usability is necessary. You should follow all the best practices regards navigation, colours and layout. Accessibility is becoming increasingly important. We've seen our own visitors ask us to change font sizes (make them bigger) for instance. We now have a function on our own site which allows you to adjust the text through the browser to any size you like.
Which online retailers do you most admire for their shopping cart processes? And which ones are a 'dropout from basket' waiting to happen?
Amazon and Cafepress.com have very friendly and simple interfaces. The interesting thing about these two sites is that while they do take your details and record them so it's easier for you the next time you visit, they do not ask you to "open an account" or "create account" because the wording itself puts many people off. How many times have you bought something from Amazon because they "suggested" something based on your interests? Amazon really understands about building a site around what the user wants (or might want based on their preferences).
The sites which are a drop-out waiting to happen are the ones which have common problems, like requiring registration before a purchase, long checkout processes (I've seen ten steps to purchase before), no shipping information displayed, no privacy policies, no security (SSL encryption), lack of guarantees and return policies, the list is endless.
Copywriting is an increasingly essential art form for many e-commerce teams. But the challenges for large websites can be immense, if authoring occurs at a local level. How can you manage authors so that they stick to the conversion-focused rules?
Firstly we figure out what their limitations are. The content management systems or technical backend may mean that we need to adapt our methods to their systems. Then we hold them by the hand for a number of their most popular products showing them how we've come to the copy and content we've developed. Then we design a custom template with rules to follow so that the content managers can work to a specification. The content managers are then responsible for their own testing to improve the site copy and this is done in conjunction with A/B split testing and or web analytics. We either handle that ourselves or train the managers to do it themselves. Ownership of the site is key, by getting the content managers to own and develop their content based on what works, gradually the conversion improves.
Do you have any insight into how the demographics and conversion rates differ between Google, Yahoo and MSN users?
The demographics are important and can mean a whole change of strategy for some businesses. This observation of the Nielsen//Netratings figures by Danny Sullivan is particularly good and current;
As for conversion rates from different engines, we have yet to do a study of that and I don't know anyone who has. We tend to work on a client by client basis and work out the best sources of traffic for each, based on their own particular product ranges.
(Obligatory end of year question alert.) What do you think will be the top trends for webmasters in 2006?
I'd like to think that the focus will shift from traffic acquisition via search engines to traffic conversion but I think it's still a little bit too early for that in Europe. It's beginning now in the US and I think it will take another year before the UK and later Europe really start to focus on it. I do feel that this year because of Google Analytics more focus will shift toward web analytics, which in my view is as important for the industry as when PPC was introduced. It means businesses will be able to use a quality tool without paying a fortune and see for themselves how a good tool can affect the bottom line. This can only be a good thing as the trend will shift towards measurement rather than best guessing which has been going on for far too long.
About the Author:
Steve Jackson is the Editor of the Conversion Chronicles, a website conversion rate marketing newsletter dedicated to improving website conversion rates. He is also the CEO of Aboavista a web conversion and web analytics consultancy based in Finland and the USA. You can get a free copy of his e-book sent to you upon subscription to the Conversion Chronicles web site.