Conversion Optimisation – the untapped potential for Affiliates?
iGB Berlin Affiliate Conference
1-4 November 2017
What I want to talk to you a little bit about today is, is what you can do in that space with CRO and what’s the most effective things that you can be doing on your site to to increase their sales. So, first of all, a little bit of background to me. So Browser to Buyer, we are an expert, data driven conversion optimisation agency and we work with brands around the world to basically to understand what it is that stops their users from converting.
And then we run AB tests that significantly increase their conversion rates. And this is this is something that can make a big difference to a business. Most of our clients get 10 to 30 percent increase in their conversion rate within the first few months of working together. But I’ve worked with all sorts of different businesses, but actually some of the most interesting and some of the ones where there are the biggest potential, our affiliates. And that’s because there is a huge amount of potential in this industry and there’s a huge amount of opportunity to change the way that people act now.
If I just start off and talk about conversion optimisation in simplest form and why it’s so effective, if you have a funnel or a Web page and you can make a change to it, that we can measure as an increase of, say, 10 percent in the conversion rates on the home page, it’s then going to filter down. It’s going to go right down to the bottom there. And you end up with 10 percent more people reaching the end, 10 percent more sales, 10 percent more revenue.
And that’s all coming from the same traffic. So you can see why in e-commerce and retail and lead generation, this is a big deal. And this is what a lot of businesses are doing. And when it fundamentally comes down to how you achieve those sorts of results, it’s really down to two things. First of all, you want to understand your visitors. So that’s about understanding what their problems are. It’s as simple as that. If you can understand why someone doesn’t convert that, it’s quite easy to fix that problem.
And then when you know what the problems are, you then start influencing them. And that’s all about using things like psychology, persuasion, great copywriting, changing your headlines, anything you can do to sort of change the way that the user will act on your site. But where in the world of affiliate marketing and we don’t often have a funnel to work with, we don’t have all that huge amount of information on what products people are viewed or where they’ve gone on a website.
To us, the world typically looks a bit like this. You’re bringing in your traffic over here on the left, which you guys here are all really good at, even though that’s an extremely challenging thing to do. They’re coming onto the site. They may only be viewing one page and then they’re clicking out and they’re clicking out to a provider where basically they’re going out into this black box over here. You’ve got absolutely no control over what the service is doing.
It’s someone else’s website. You can’t see what they are doing. You don’t know where they’re dropping out of what they’re doing. And of course, the majority of them do drop out at the end of the day, even if you’re sending out hundreds and hundreds and thousands of people to an operator site, most of them are going to drop out and end up going somewhere else or not signing up for the lucky few. You get some registrations out of it.
They turn into players and eventually they might turn into revenue for you. So the question is, if we can’t look in this box and we can’t influence the people in there, how are we actually going to use conversion optimisation? And if you say, does it work for affiliates? Well, fundamentally, yes, I’ve seen some fantastic results. So, for example, typical results with affiliate sites. I’ve seen 10, 20, sometimes 30 percent increases in the number of people clicking out from just one test.
And in fact, one affiliate I worked with when they added up the value of conversion optimisation, it was somewhere in the region of 20000 dollars per week. So there’s a lot of opportunity here. And what I want to do is talk a bit about how you can basically do these two steps on an affiliate site to generate a significant increase in the amount of sales you do. So first of all, it’s about how you understand your visitors and as I say, this is about understanding what their problems are.
Now, the simplest and easiest way to do this is to use three tools and services on your site. I’ll start off with Google Analytics or another analytics package, if that’s your your preference. Now, it’s not just about installing these packages on your site. It’s about using them properly and using them in a way that tells you something about those visitors. So what I mean is don’t just work out where people are entering and where they’re clicking around, but make sure all your affiliate links attacks that you know where they’re clicking out, not just where are they clicking out, but which operator they’re going to if you’ve got multiple links on the page, which one of them are they following?
You know, is it the one up at the top of a review of a site or is it the one where you you mention a specific slot that they offer? It’s all about getting that really rich, detailed information about your visitors. The next tool I’d recommend is called Hotjar. Some of you may have come across this. It’s a qualitative and quantitative research tool. It’s a line of code to add to your site, basically, and it’s really cheap.
Most sites, it’ll be less than about 90 euros a month. So you’re talking about really effective from a cost perspective. And what Hotjar will do is give you so much more detailed information about what the people are doing on your site and some of the useful things you can get. Heat maps. So Heat Map will show you, for example, where are people moving their mouse on the site. And this is useful because basically we move the mouse to where we look so you can start to see things like it’s this content up the top left that people are looking at.
People may be spending quite a lot of time over here and knowing what content that is. You can then start to find out what’s interesting to your users, you know, is the part of a review on your site that they read the bit that explains about which payment methods the casino uses, or is it that when someone’s browsing down a comparison table, they’re always looking at the number of free spins they get? So you can start using these. As I say, this is on mouse movements.
You get them based on clicks and also scrolls, which is really important to see which information people are spending the time reading, and especially on mobile devices. Are people even reading the right information on your page? Another great tool is session recordings. This is actually taken from a travel site I work with, but what you’re seeing here is actually a recording of a real user. This is their mouse movements. This is the clicks. This is what they’re typing on the site.
And you can get this for all your visitors. And it’s hugely valuable because you can start to see where the problems are, where people are struggling again, what sort of information they’re looking at, where are they clicking around on the site? It’s this sort of insight that can start to tell you a lot about what your visitors are actually doing. And a really nice tool is that comes as part of Hotjar is polls. So if you have traffic that spends a little bit of time on your site, actually interacts with you, start asking little questions like this down the bottom so you can pop up something, you know, what do you look for in a casino?
What’s your favourite slot? Why why do you use this particular site over that one? You can start to ask these simple questions and then you’re getting in your users own words what it is that’s important to them. When you start to analyse this, you basically then know more about your users and what you need to offer. And the third tool is called What Users Do, and this is more a service than at all. So what users do is a user testing service.
Now, when you hear user testing, you probably think of usability. And again, big e-commerce sites running sort of these sort of user tests and finding out where the bugs are on their site. But what users do offer a panel of users, they have a few hundred thousand users around the world. And for a small amount of money, you can pay to have one of these users who match certain criteria. So you could say someone who currently plays slots or someone who is doing spread betting and they’ll come until they’re come up to you and you can set them a set of tasks and a scenario and you’ll get back to them about a 20 minute video which shows that you are carrying out those activities.
So it’s not just about what they do on your site, but you can ask them anything. You know, you could get them to say find yourself a new casino and you can hear them talking about why they’ve gone and chosen that one. You can see the process they go through. It’s even more valuable because it gives us an insight into that black box. So it means that once someone even if someone goes off our site because we’re actually recording a particular session, we can see what they do.
We can see when they click out to that casino and we can see what the problems are when they’re signing up. So basically, these give you a lot of opportunity to gather a lot of data about your users. And really what you’re trying to do is find out two things. First of all, is what we call the barriers. So a barrier is something that will stop someone or make them less likely to convert on your site. So it could be, for example, that when they click out, they see a horrendous sign up form and it stops them from converting.
It could be that, you know, there’s a complex process when they go out onto one of those operator sites where they, you know, they’ve got to download and install. They get confused. They don’t know what they’re doing. They never end up signing up. It could be that they’re not clicking out because they don’t trust the Kseniya linking to maybe they’ve never heard of it. Maybe they think it’s in a strange country and they don’t trust it.
So it’s all about identifying what the I would typically look for sort of the three to seven top barriers are for your users. What are the things that would stop them from clicking out and what are the things that stop them from signing up and playing? And then you need to look on the flip side of that, what are the motivators? And these are things that will make someone more likely to take an action, things that will make them more likely to click out or things that will make them more likely to sign up or play.
And again, these these are very unique to your site, but they can be any sort of things. You know, it could be that William Hill are offering an enormous bonus that’s much better than you’d normally get anywhere else. It could be that they particularly like a specific slot. So by telling them this slot is available at this casino, they’re more likely to click out. It could be that you’ve negotiated an exclusive bonus. So it means that I’m more likely to convert through your site with 888 than with any other site.
And it’s all about identifying, again, what are these things that make people more likely to convert? And once you’ve got those sort of three to five barriers and those three to five motivators, it’s then all about going in there and actually starting to influence people. And this is about, if you can, running split tests. You can make these changes to your site anyway. But I strongly recommend you install a testing tool and that the split test, these changes so you can measure the real impact that they have.
So what I want to talk a little about in the sort of second half of this talk is some of the things you can do, some of the general techniques you can use, and some of the tips I’ve got for how you can actually influence users once you know more about them. So I’m actually going to start off with them. Some things you shouldn’t be doing, some some things that are basically a waste of your time and effort. Now, you’ve got a limited amount of traffic on your site and you want the majority of them to convert.
So don’t waste your time doing things that aren’t going to make a difference. And often these are the small tweaks. These are the things that give conversion optimisation a bad name, things like changing button colours. If you’re a big retailer, if you are a John Lewis or something of this world with millions of transactions going through your site, getting one or two percent out of little changes like this do make a difference to your business. If you’re not like most of us, then you’re going to have to go for the big things, the things that are 10 or 20 or 30 percent, the things that really make a difference.
So don’t worry about little things like button colours and little tweaks and moving things around. What you really want to be doing is influencing the way the user thinks and changing the way that they interact on your site. The second thing I’d encourage you not to do is to follow best practise so you can search for conversion optimisation tips on the web. You get hundreds and hundreds of pages of. Examples 544 conversion optimisation tips would keep you going for years. The problem is that fundamentally there’s two problems with with best practise.
Number one is that this isn’t specific to your site. So there will be things on there that people suggest that generally work. If you try them across enough sites, they will probably work. But they’re not what your users are struggling with. You know, it might be that there are examples around sort of form validation and things like that where your users aren’t filling in many forms, for example. So it might not be a problem. But the really big thing is, especially in this industry, most of this best practise just doesn’t apply.
You know, a big one that you’ll see quite a lot is things like embrace new design trends or remove the clutter, which is fine if you’re running some sort of high end fashion site. But quite often, if you’re a gaming affiliate, you need that authenticity. You need to be coming across as someone who’s believable and that people are going to look at. And actually these sorts of modern designs are possibly going to turn users off. So if you’re not following the best practise and you’re not following those little tweaks, what should you be doing?
Well, the first simple thing is give the users what they want. And I’ve already talked a little bit about this, about how you discover what people want by using things like heat maps and polls and surveys and all sorts of things. But once you’ve done that, it’s then about giving it to them and making that the most important thing on your site. So here’s an example, actually, from a Web hosting affiliate that I worked with for a number of years.
And what they did is they offer reviews on my postings of Web hosting services. And what they found is if somebody is going and looking for somewhere for their WordPress blog, the single most important thing to those users was that the host had been recommended by WordPress themselves. So once we knew this through research and testing, we could start putting that all over the site. So you can see, hey, you know, we had a key feature recommended by WordPress and not just that, but then we started putting badges across the site showing where WordPress, which hosts were recommended by WordPress.
You start to make this more important to people. And because it’s the thing that they value, it makes them more likely to convert. Here’s an example. Actually, this isn’t a site that I work with, but you can see exactly the same sort of thing going on here. We’ve got an affiliate page for 888 Pocho. And clearly what they believe now, whether this is through research or not, I don’t know, is that easy withdrawals are the important thing to their users.
So what we’ve got here is things that say withdraw in just one day. We’ve actually got a review here that explains easy withdrawals. So because that’s the thing that’s important for our users, that’s what you need to make sure it’s the key critical information that you’re sharing on the site. And as you start to learn this, of course, you can start taking out the things that don’t matter to people and start making sure that they’re getting the really, really important things.
The second thing that I’d strongly recommend doing is actually test the offers and programmes that that are available to you. I mean, even if you went round this whole today, you could probably find yourself 20 or 30 new deals, new casinos and new operators to be promoting on your site. So there’s a major question, how do you do something like this and get as many on there as possible? You know, someone’s going to find something they like there, so they’ll probably click out or do you go the other way like these affiliates do, and just recommend one key offer?
And, you know, this works both ways. And this is why I would encourage you to test it, because if you go this way, you can spend a lot of time selling that offer. You can benefit from things like the volume from sending most of your traffic to one casino. But of course, if they don’t like casino land, you’ve probably lost them as a sale. So the first thing I’d recommend, they start to look at how many different programmes you want to offer to your users.
And then again, through split testing, find out which ones are actually best for your site. So I’ll show you an example here. If we have two offers, one from casino and one from casino, B, a three hundred dollar bonus and a thousand dollar bonus, what would actually happen if you split test them against each other? What if that’s the one big offer that you’re showing on your site? So if you put, say, 50000 visitors through each one, you might find that unsurprisingly, if my if my clicker works, there we go.
Unsurprisingly, the casino B, for example, generates more clicks there. Quite often that’s where people stop because that’s working better for what you might find. This casino has a much better sign up process. Perhaps it’s much more user friendly. Perhaps Casino B makes things very difficult for people. And what you find is from that traffic, you’re actually generating more. Sign ups and fundamentally a lot more revenue from those customers. So what I’d strongly encourage you to do is where possible track all the way through to the conversions and there’s various ways to do this and then start split testing and find out using data which are the best programmes to be promoting, how many of them you should be doing and which ones are best for your unique set of users.
Another nice example here is this is actually one that this is taking off an operator site, but I’ve done it very effectively with with affiliate sites as well. It’s about gamification. I mean, at the end of the day, we’re working in an industry where what people want to do is play and win. So why not offer them that opportunity? We spin to win. And amazingly, they I’ve won a 1000 pound sign up bonus, which is fantastic.
But this brings in actually some very interesting pieces of psychology. If you do this on the air on your own affiliate site, you say, you know, you’ve played you’ve won a deal with one of your partner casinos. Suddenly people have they have that aspect that this is something they’ve won and this is something they own. And it brings in a psychological principle around things like loss aversion. Once you feel like it’s my bonus and I’ve won it, I don’t really want to give it up.
So I’m far more likely to click out and I’m far more likely to convert. So not only are you creating a sort of innovative and interesting user experience on your site, but you’re also making them more likely to convert. A third technique you can use is upselling users. So, again, this is something that that works very well on any type of affiliate site. In this case, it was again in Web hosting. But but I’ll show you an example of how you can do it in in the gaming industry as well.
So when you think about upselling, you start to think of people like GoDaddy. You go on to buy your domain name and you end up getting Web hosting and an SSL certificate and everything else they offer at the same time. But even if you’re an affiliate and we’re just talking about clicks, you can start looking at, can you, Opsahl? People from low paying affiliates, from low paying operators to higher paying operators. Let’s take, for example, here, we’ve got a comparison and let’s take Luckie 24/7.
Apologies if they’re here. I’m not judging them in any way, but let’s say that they don’t actually pay out very much when when we send the user over to them. So when someone clicks this get bonus button, we could pop up a modal window and we can say you could double your your bonus by going to Petteway instead. And this can be hugely effective. And again, when I was working with the the Web hosting affiliate, what they found was because they did a lot of reviews, they were getting a lot of people on the pages that effectively weren’t making no revenue for them.
And by putting this sort of model, when somebody clicked out on a page that paid them no revenue, suddenly they were directing them to the higher paying programmes and they were turning traffic that was worth very little or nothing to them into a significant source of revenue. So it’s starting to think about these things, think about them in a different way. And it’s not about simple things, about just can I get someone to click? But how can I make that click as valuable as possible to my site and to my business?
Another thing you can do is to try and prepare users for what’s coming up. Now, I mentioned right back at the beginning, when someone clicks off your site, they go into the black box. That is the casino. They go through some sort of sign up process. And fundamentally, you have very little control over what’s happening there. But if you know what the users are going to see and of course, you can go through the process yourself, you can start to prepare them for things and prepare them for the problems they’re going to come across on the way.
So this is Sportingbet sign up form and has stayed the same way for four or five years at least. But this is the sort of thing that can put off users. You know, there’s a lot of complicated things and they they might not sign up. But if you can do things on your site to prepare them for this, you can overcome it. Similarly with things like, as I mentioned before, if somebody has got to download and install a piece of software that starts to bring in difficulties for them or even if they they go to sign up and they don’t get told that they’re actually getting a bonus.
So let’s say they’ve clicked out to this casino expecting to get a five hundred dollars sign up bonus. There’s no mention of it here on the landing page. There’s no mention of it in the sign up form. There’s no bonus code in here. The user doesn’t know if they’re even going to get it, so they may well drop out. So what you can start doing is this sort of thing. Explain to people what they’re going to experience when they get out.
You know, explain to them to claim the bonus click play. Now, install the casino, go to the cashier, and here’s the coupon you need to put in there. And whatever that process is, you know, if, for example, there’s a. Really complex sign up form, and they have to put in some information that they wouldn’t need to just warn people of that. So here are some of the things you’re going to need to sign up.
Here are some of the things you might get confused about. This is where you go to claim this specific offer. So it’s about what can you do to overcome those problems even when you don’t actually control that process yourself? So fundamentally, there are two things you need to be doing when when you’re working with users on your site, first of all, it’s all about understanding your visitors. And as I said before, use things like Google Analytics, Hotjar, and what users do to really understand what those barriers are, what the things that stop someone from clicking, what are the things that stop them signing up when they actually get out to that partner?
And then what are the things that will stop them maybe betting later on how do you get a more valuable customer and then identify motivators? What are the things that they want? What do these people actually want from your site? What do they want from the site? The deals that clicking out to and then you need to start influencing them. And I’ve only scratched the surface today, unfortunately, because we’re we’re quite tight on time. But what I’d encourage you to do is don’t talk about the small tweaks and don’t look at the best practises.
What you really want to be doing is making sure you’re identifying what users want and giving it to them, doing things like testing your offers and programmes, start game offering, start upselling the users and prepare the users for what’s coming next. If they’re going to have some problem when they click out, make sure you’ve prepared them for it. Now, as I say, this is really just scratching the surface. There are hundreds of different things you can do on affiliate sites.
It’s definitely a huge amount of potential. I’d encourage everyone here, if you do nothing else, get something like Hotjar on your site and ideally start running tests. And if anyone is interested in finding out a bit more about some of the things they can do or you want to talk to me about about the specific issues you’re facing on your site or what you can do to improve it, then I’m going to be around through through the whole break just down the front here.
So please come and talk to me. I’m happy to talk about anything to do with conversion optimisation. But now I’ll hand over if there’s any questions anyone would like to ask to have some questions or do I find it? I find it. Actually, all your summary was excellent, by the way, but I find it very difficult to take that second step, which is understanding the user, because I use the heat maps and I hire user testing dotcom and whatever else.
And I feel like when I’m hiring these services, I’m not getting results. They’re actually gamblers. They’re just people that they’ve paid to come and play with my site. I get a lot of stupid feedback and my heat maps don’t seem to be super useful because they’re all the same. So where do you find the real data? How do you decipher what people have said that have been paid to go and review your site and turn that into something that’s useful?
So there’s probably two things there. And yeah, it can be very challenging. And yeah, you look at heat maps look pretty and they look great on screen. Sometimes they won’t be very useful to you. And it really depends on the type of site. Personally, I find session recordings are a lot more valuable because you’re watching the real users going through there. But really it’s about trying a whole variety of techniques and working out what works best for your site.
You know, you may have the sort of traffic that you can ask them a question and pop up a poll and that answers your question for you. It might be that heat maps work very well for you when it comes to things like doing user testing. It’s about crafting the right sorts of questions. And yet people always try and game the system and they’ll say, yes, I came online and they don’t see you can start getting better, better questions when people come up.
So I think I think user testing have stopped people from doing this now. But what users do still allow it where you can actually ask a set of specific questions to people and and weed them out. If they’re not the right sort of users, you can ask them things like, have you played online in the last month? For which casino did you play with? Or something like that to try and weed out people who aren’t valid user testers. The other thing you can do and actually Hotjar gives you the facility for this is recruit people on your site so you can actually offer to your users fifty dollar Amazon voucher, things like that.
It just pops up during their session. Would you like to help us out? And then you book a time with them and you can actually send them to someone like what users do and run them through that process. So there are different ways you can do it. And it’s really about finding what works for your site. Once you start doing it, once you start doing some of these things, you’ll find out what works and what doesn’t and you can start to build on that.
Any other questions for Dave?
Could you please go back to the slide where you compared to casinos, one with 300000 and one was one thousand dollars?
Yes, let me actually I’ll do it on this probably a bit quicker. Yes. This one here. Yes.
And in the end, you compared the revenues of casino, which is higher than Casino B. And so you said casino is better. Yeah. But for us as affiliates, most casinos work this revenue share. And there is often it’s one player out of 100 who is a high roller and then completely screws the statistics. So I’m not sure if it’s the best approach to look at the revenue or maybe we should go one step back and look at sign ups or real money, please.
Absolutely. I think I think it’s really working out which metric is right for you. I suppose the message here is don’t just look at clicks. And ideally, if you’re working with casinos where you can set campaign parameters and things, when you send people out, what you’re able to do is for every test you can start to measure that. And it might be exactly if you get one or two people skew your your revenue or your revenue share because they’re particularly great players or particularly bad players, then, you know, you might not find that necessarily the revenue figure is what you go by.
But certainly at the end of the day, what you’re trying to do is generate the most play as possible and the most valuable players possible. So it’s finding, I guess, finding the point on this scale where you’ve got enough volume to start seeing that on a basis that isn’t just I’ve got three players here and five players here and this one’s great and this one’s not, you know, it’s if there’s X thousand or whatever. So, yes, certainly the sign ups, if the revenue figure isn’t isn’t telling you necessarily what what you are, what you’re looking for because people are skewing that figure.
Any last question? Yes, hi, first of all, many compliments for the presentation you talked about just, yeah.
After how you decide when it’s satisfactory to you, for example, after 100 users, a thousand users after a couple of weeks and.
Yes, it’s always a good question. And, well, the simple answer would be using in your testing to statistical significance, which, you know, it’s it’s a it’s quite an under you sorry and overuse figure. You know, you will look in the testing tool for 95 percent. Confidence is probably the first thing to look for. But fundamentally, there are a number of things you have to be careful of. So when you’re running a test, my rule of thumb, would you you’d be wanting to look for about three to five hundred actions for each variation.
So if you’re measuring two clicks, three to 500 clicks, three to 500 sign ups, whatever that is, you’ll be looking for 95 percent confidence in the testing tool. And they will all report that and and we’ll give you those sorts of numbers. But the other thing is it’s very important to look at things like business cycles. So if you’re looking at things like you will get different people in evenings and during the day, you’ll get different people at the weekends and at different times.
So work out what sort of your minimum business cycle is, which is usually a week, and then make sure you’re running not just one or two of them, but ideally a whole number of them. So, you know, if you’re running a test, don’t start it on a Monday and stop it on a Thursday. You want to do a full number of weeks because otherwise you could be skewing it because traffic at the weekend is is affected more by the test than others.
So fundamentally, you’re looking for a minimum of three to five hundred conversions on each. You’re looking for ninety five percent confidence and at a minimum, running it for a week, if not two weeks or more, if you have to. But those would sort of be the numbers that the sort of starting point from from my point of view, there’s a judgement, a round of applause for Dave.