When it comes to engaging with your site’s users, conversions are the single most important aspects for you to focus on. Visitors are important to your site, but conversions are even better. Having lots of people visit your page won’t matter if they don’t read or view the content that you have, or worse, never come to you again.
In our previous blog post, we were discussion some basic metrics. We defined conversion rate as the total number of conversions (the ones you defined, could be sign ups to your newsletter, or people buying your ebook for example), by the number of visitors to your site.
#1 Setting up a baseline
The number of visitors to your site can be defined in two different ways, you can see them as Total Visitors or Total Unique Visitors. Whatever number you decide to choose, consistency is key. If you think that for your situation Total Unique Visitors gives a more accurate measure of your conversion rate than Total Visitors, be sure to use it consistently from there on. If you change this later on, of course, your trends will be off.
So what is the difference between the two? Think of it this way..
You own a brick-and-mortar shoe store and a person walks in. That person checks out a couple of shoes, your staff helps them pick a nice set of shoes.. Then that person sees a friend walking past the shop and goes outside to greet her, or maybe they forgot their wallet in the car, or maybe that person decides to visit some other shoe stores first to compare some of their offerings.
After some time, he or she comes back to the store, and each time he or she walks in counts as a single visit. If they drop in four times before making a final purchase, this off course will count as four visits ( four Total Visitors). This person however off course is the same unique visitor (one Total Unique Visitor) making four visits to your store.
The same thing happens online, visitors to your site look around a bit, they often get distracted (a friend sending a funny cat picture) and online even more than offline they check out the competition.
For this reason, a lot of people decide to go with the Total Unique Visitors as their basis for determining their Conversion Rate. Although this might sound really easy, there is, of course, the problem of identifying if a visitor is a unique visitor. You need to put some technology in place, like setting a cookie when someone visits your website, however, this is not always perfect. What happens for example if someone clears their cookies from their browser?
The last thing to take into account when setting your base metrics is the time period in which you track. Just like with determining if you want to go with Total Visitors or Total Unique Visitors, it doesn’t really matter what you choose as long as you keep it consistent. Dividing a day’s Total Unique Visitors by the number of people who converted that day and you’ve got a daily statistic on conversion on your website. You can also look at statistics per week, this makes sense for example when you run different campaigns which last a week. This way you can objectively see which campaign converts better.
#2 Finding problems in your Conversion Funnel
At its most basic level, conversion rate optimization (CRO) is meant to increase lead generation and sales by maximising each visit to your website. It’s the process of finding out why your visitors aren’t converting (doing the main goal you want them to do) and then fixing it.
How do you find the problems on your website? Where do you start?
User Scenarios: Roles and Tasks, NOT Personas
Start by putting yourself in your visitor’s shoes and looking closely at your site. You can get the conversion by identifying who are showing up on your site, and checking if you’re supporting them in a supporting way to accomplish their tasks. Think about the baggage the visitors are bringing in, their knowledge or lack thereof, and their apprehensions. Then, taking all of these into account, think of how you’re going to match their intent.
When coming up with user scenarios, use roles and tasks instead of personas. Think of the visitor’s relationship to your site and what they’re there to do. Roles and tasks are more durable than visualizing user segments summed up in a made up character persona. A person is represented by one persona, but really a person’s roles and tasks will vary depending on the context he/she is in.
A visitor may be an expert on the latest apple gadgets but totally clueless about shoes. He or she will navigate the online apple store differently when looking for a new iPad than he or she would, say, zappos.com while looking for a present for his niece
So think about the person’s relationship with your site, and what they’re there to do.
When organising your site, think about roles or the classes of people that interact with the converting-critical parts of your site.
Some of the areas to look at
Now that you defined the potential user scenarios on your site, it’s time to move on to your website itself, what are some of the areas you should take a look at?
- Is your Call-To-Action (CTA) easy to find?
- Are you communicating in a simple and clear way?
- Can visitors easily search your site for what they’re looking for? If you’re in e-commerce, is it easy to complete your checkout process? How many pages and clicks does it take to complete the key conversions you’re measuring? Is there a mobile version of your website? Keep your navigation, registration, contact, and payment uncluttered and easy to find and operate.
- Are you building trust and credibility? Are you offering your website on an SSL domain? Do you have user testimonials on your website?
- Is the underlying HTML of your website structured properly? Are you using correct titles, relevant keywords etc.? These things help not just for your Search Engine Optimization (SEO), but if people are coming in via a search engine on your website and the content is something completely different than what they were looking for it is not really helping your conversion 😉
These are just a couple of items to look into, as you can see this list has a striking resemblance with our previous post on “How To Make Great Landing Pages (With Crazy High Conversions)”. This list is off course only a start, what are the things you are looking at on your site? Leave them in the comments to make the list bigger and bigger!
Every website is different, each site has its own unique mission, challenges and strengths. So this means that what converts for one site, doesn’t have to convert for another website. Someone might report that changing their CTA buttons to the colour green creates amazing conversions, but it’s important to realise you need to test for your own situation.
Whenever you make changes to your CRO strategy, continually test them to determine whether they work. We have defined our baseline, we found potential problems on our website, the next step is to start creating and running some tests.
Form Some Testable Hypotheses
Using your analysis on potential problem areas on your site and your analytics tool you use for your website your find out that a particular page on your website is having an increasing bounce rate. You decide that maybe you want to implement a chat box, or an on-page survey on that particular page to ask your visitors what they were looking for and if they were able to find it. You could also run some tests to see what people doing while they are on that page. Additionally you could ask some people to look at your page and watch them try to navigate through it.
With all of this information you get from observing you can start forming a hypothesis that attempts to explain why no one is converting on that page. Ending up with some changes and alternative versions of that particular page. That brings us to the next step.
Design Your Tests
Now it’s time to take all your learnings and come up with a strategy about what to test. Start by making a list of your priorities. Which are the points, which are coming up again and again when you did your surveys earlier? By looking at that data, what do you think are the biggest issues on that page? So what are the items you are going to address first?
When coming up with a strategy there are a few points to always keep in mind:
- Start small, something the smallest little tweaks with the least amount of development can have great potential for getting better conversions
- When starting small, always think of two potential solution. This way you can split test the improvements and learn the fastest
- Change one item at a time, otherwise you won’t know if the different colour on the button or the copy on the button is the one leading to better conversion
- Get a second opinion, if you are the one always working on the site, it is easy to spot the problem. You are simply to involved, ask the opinion of an outsider
- Try looking if you can find some benchmarks from other companies in your industry. You might think you are having a bad conversion on your specific page, but when looking at other companies you might be having an amazing conversion!
- Double and triple check your tracking on that page, if you start doing a test but are not recording the results properly you will never know if your test worked
- Set a minimum success criteria and stick with it. If you think based on your visitors, you need a minimum of let’s say a 1000 visitors to make an objective call what is the winner of a test, don’t stop after 200, just because you think you’ve found a winner. Let the test run it’s course and then decide.
Run Your Tests
When you run your tests, always measure the outcome against the baseline we set up in step 1. The data you’ve got after running the test, when compared to this baseline, will tell you where to go from there.
If the test wasn’t a success. No worries. This only means that it is time to go back to reexamine your data and design a new test. Not all test can deliver a winner straight away and you can learn as much from a non-successful outcome as from a positive one.
Was the test a success. Great! Open the champagne bottles! 😉 Now you can either cross this page off your list and move on to the next problem or you can continue to come up with even more tests to keep on improving this particular page.
Congrats, you’ve made it all the way to your initial round of testing, however you should think of optimization not as an end goal but as an ongoing process. You will never reach the point where you’ve run “enough” tests. Apply these steps though, and you’ll soon see the results of your efforts. How has CRO helped your bottom line? Let us know in the comments below!