Web metrics are everything these days. With free analytics tools like Google Analytics or services like Kissmetrics and Mixpanel that tie user activities to real identities, there’s no reason you shouldn’t be capturing data from your website.
When it comes to online marketing, the ubiquitous availability of useful data has been a game changer. No longer does marketing have to be a guesswork. Feeling as though you are dumping piles of money into a dark bottomless pit can be a thing of the past. You now have the opportunity to be very deliberate with your marketing efforts, using your dollars only where they actually benefit your bottom line.
But all the raw numbers and data visualizations are nothing unless you know how to act on them. This is where conversion rate optimization (CRO) comes into play.
What is Conversion Rate Optimization?
Simply put, conversion rate optimization attempts to make a website better at turning visitors into customers. Increasing the total number of visitors to your site is an invaluable part of a holistic online strategy. Conversion rate optimization, however, generally focuses on activities that gain more conversions from existing traffic numbers.
This is nothing extraordinary by itself. It has been the goal of countless redesigns and platform shifts since it was discovered the web could be a useful marketing platform. The real magic happens when these efforts are combined with proper data analysis to form an extremely efficient and powerful method for validating almost any idea you can think of.
What does this mean in practical terms? Four key elements come to mind when considering what really makes conversion rate optimization stand out.
- Define measurable goals.
- Know your customers.
- Test everything.
- Evolve your site. Don’t overhaul it.
Define Measurable Goals
It’s obvious that a website should have a goal. Anyone who sets out creating a website has a purpose for doing so. Often time, however, these goals are vague and unmeasurable. If your website goals cannot be objectively measured, you have nothing to compare any “improvements” against.
If you haven’t identified goals for your website or the ones you have defined cannot be easily measured, spend some time bringing some clarity to the subject.
Start by answering this question. What action do you want your visitors to take before they leave your website? Your answer should be any action that you can tie back to your bottom line. Examples include buying something, creating an account, completing a contact form, scheduling an appointment, sharing a piece of content, and so on. Regardless, clearly define your goals and be sure they can be objectively measured.
Know Your Customers
It is crucial to understand your customers, their needs, and their online behavior patterns so you know what information to provide, at what time, and in the correct format. This is a huge topic that extends well beyond that of website conversion rate optimization. I plan on going into more detail on customer research tools and methods in future posts. To start with, however, here is a list of questions you should consider when attempting to optimize your website for conversions.
- How many different customer segments do you need to speak to?
- What questions would they be asking at each stage of the buying cycle?
- Are you providing useful content at each stage for each customer segment?
- Is the content properly formatted at each stage for each customer segment?
- Is the content in the optimal order according to the behavior patterns of your visitors?
- Have you included the necessary call-to-actions using appropriate trigger words for your audience?
Customer feedback helps you make informed decisions in all of these areas. Many useful tools exist to make engaging with your customers easier than it ever has been. Again, there is no reason why you cannot or should not be engaging with your customers.
If you think you know about your customers without actually talking to them, you are building a house of cards and setting yourself up for failure. Any successes you have had up until this point are most certainly the result of lucky guess work. Repeat the phrase “you are not your customers” until you understand what it means. Then, repeat it some more until you start acting on it.
If you’re looking for a new podcast to listen to, I would like to suggest the Spoolcast from usability export Jared Spool. It’s always entertaining and you’ll almost certainly learn something new each episode.
A while back, I was listening to one episode in particular. Jared interviewed Jeff Gothelf, author of Lean UX. As usual, the conversation was insightful and entertaining. However, Jeff’s view that “designs are hypotheses” resonated with me a bit more than usual.
Every scenario is different. What works in one situation may not work in another. We must remember that the web is still a very young medium. Things can change very quickly. Just when the dust begins to settle and everyone begins writing their “top 10 best practices” posts, some startup appears out of nowhere, introducing a new technology that turns everything on it’s head.
It’s impossible for anyone to keep up. The field of conversion rate optimization is not one in which you can gradually build up a body of knowledge overtime, eventually hitting a threshold that finally qualifies you as an expert. As conversion rate optimization specialists we are not experts at what works. We are experts at figuring out what works. The only best practice is to test everything you do.
The testing method I’m mainly advocating here is A/B (or multivariate) split testing. This method allows you to compare new design concepts against the current version to see, live and in real-time, which does a better job encouraging visitors to complete a desired action.
I’ll be posting a detailed post on split testing, so I won’t say much more about it here. Just know that until a design has been validated, it is nothing more than an educated guess. The more time and money you spend before you validate your work, the more you have to lose when it turns out you were wrong.
Evolve your site gradually. Don’t overhaul it.
The last sentence in the previous section is extremely important. Small incremental changes can be validated quickly. Fewer variables will make it easier to understand the precise cause of any increase or decrease in conversion rates.
Think of it this way. If you wait six month to validate your work, you’re not just wasting the time and money it took to do it. You are also wasting any potential gain you could have had if the change was implemented sooner. More importantly, if 6 months of work is released in a single moment, it will be impossible to identify the exact cause of any increase or decrease is conversion rates.
With small, incremental changes, you can clearly see the effect of each change. All future work can build atop earlier insights. Five smaller tests, each with a 5% increase in conversion rates, will net a better result than a single increase of 25%.
The bottom line is “Test early. Test often.”
I haven’t even begun to scratch the surface on the topic of conversion rate optimization. I’ve said nothing of conversion funnels, or content strategy as it applies to conversion rates. I only briefly mentioned web metrics and split testing. With this post, I wanted mainly to set the overall tone and to lay a foundation I can build from with future posts and content.
Though I always enjoy discussing theory and high level strategic concepts, I hope to focus more on the practical moving forward. I’ll be satisfied if I’m frequently and consistently providing useful information, advice, tips, or tools to help you optimize your online business efforts.
Help Me Improve
This being my first post and my first serious attempt at writing about my work, I’m extremely interested in hearing your thoughts. I’ll always hope for and encourage active discussion (and healthy debate), but I would really appreciate your initial feedback on my writing style and structure.
Let me know if you’d like me to expand on anything from this post, or if you have suggestions for future topics. Also, writing is not my expertise, so fixing spelling errors and gramatical errors will be a community effort.