Introduction to Conversion Optimization
The digital marketing stream that deserves more attention than it receives
As a digital professional, you never cease to be a student; though this is true in the case of any area of expertise, it is even more so for the digital space due to its constantly evolving nature. Though I’ve dabbled in almost all areas of digital marketing, I invariably gravitate towards the awesome trio that I consider is my sweet spot — Search Engine Marketing (paid or organic), Web Analytics and Conversion Optimization. While I think that the first two have tons of content and certifications, there is relatively less when it comes to CRO, at least from my experience. While I have been consuming bits and pieces of CRO information, my search for a structured course that is based on a proven framework and digs deep into the specifics of each subject area came to a happy end with CXL.
So over the next 12 weeks, I will be sharing my knowledge, the learning experience, and the insights I gained from their Conversion Optimization course. And I believe there is no better way of assimilating or reinforcing the knowledge gained than by documenting it and sharing your thoughts on it.
What is Conversion Optimization (CRO)?
While the subject sounds very geeky, it is much more than the hard numbers and statistics that you tend to associate with CRO. It is an essential skill that will enable the spectrum of roles associated with digital marketing, be it business owners, entrepreneurs, or professionals in the trenches to create a smooth and happy experience for the audience that they come in contact with.
So why is CRO important?
There are any number of changes and updates that we do on websites — redesigns, launching campaigns and funnels, microsites, and not all of them have the desired outcome and have the potential to cost your business millions of dollars in revenue. This is where the CRO comes in and in many ways, it is sort of a safety net that ensures all the changes that your website goes through, affect the business in a positive manner.
A classic case of this was the massive website redesign that the sports retailer finishline.com underwent. In spite of the improvements in the look and feel, rich imagery, minimalized category pages, and high-profile endorsements, the site tanked in its core objective of generating revenue for the business. That was back in 2012 and the site has got a brand new design now, hopefully with all the CRO fixes that cost them badly.
What CRO does to business or website owners is to prevent them from being caught in subjective judgment or the various cognitive biases that humans tend to fall for and instead facilitate decision making based on objective data, analytics, and insights. And those subjective calls and biases cause a lot of stress — be it loss of time which means you lag behind and end up playing catchup with your competition, have your management get more intrusive since you have a failure against your name which all eventually lead to a loss in leads, sales, and money. Equally important is that it cultivates a culture of experimentation and data-driven decision making which empowers the business in the long run.
Ideas & Their Validation
When you cut to the core basics of CRO, it is about the ideas and they can be either good or bad. It becomes bad when those are not backed by data or reality that is evolved from a fair sample size of information. And conversely, if an idea to be deemed good needs to pass the validation through tests, how far can we go to test these ideas? And the key here is to be able to read the data and be able to tell what is useful from what is useless and what needs to be tested further. Fortunately, humans innately possess the ability to read data and its context to a certain degree. That is where factors like sample size become important as it gives a certain context to the data being presented.
When you cut to the core basics of CRO, it is about the ideas that we would like to try and they can be either good or bad. It becomes bad when those are not backed by data or reality that is evolved from a fair sample size of information. And conversely, if an idea to be deemed good needs to pass the validation of tests, how far can we go to test these ideas? And the key here is to be able to read the data and be able to tell what is useful from what is useless and what needs to be tested further. Fortunately, humans innately possess the ability to read data and its context to a certain degree. That is where factors like sample size become important as it gives a certain context to the data being presented.
So as a CRO practitioner you should be able to ask good questions which should be presented in the form of a hypothesis that needs to be proved one way or the other through sheer facts and figures. The good thing about creating a hypothesis is that the idea becomes a really specific task with a clear and measurable objective; secondly, this list of hypotheses will evolve into a useful source of reference while evaluating ideas in the future or tracing back in history. Once these hypotheses are listed, they can be ranked based on the Impact on the ROI on the business, the Confidence the test provides, and the Effort it takes to carry out; in short, the ICE framework.
Sources of Insights for the Ideas
So you have the ideas all listed and ranked as clear hypotheses with clear measurable objectives and are ready to draw insights from. What sources do we use for that? There are plenty of them as listed below.
Google: There isn’t a better source on planet earth to search anything and everything, you’d agree. Chances are there will be people who would’ve done these validations and would’ve shared that info and so why bother to reinvent the wheel. Word of caution here is that the info may not apply directly to your business and those are best used for reference.
Analytics: The wealth of insights that your website analytics provide can be truly priceless as it is first-party data and the feedback you receive are from your own customers or potential customers. You should go beyond the banal and vanity metrics and look for answers to the hypotheses that you had created to make it valuable. And the analytics can range from raw quantitative data to more qualitative stuff like heatmaps, scroll maps, eye tracking, and screen recordings.
User Testing: This can be as simple as an internal focus group of colleagues or friends or a bigger set of people who could be remote as well. The idea is to watch them perform the actions that you need your website visitors to do and record the experience and take their feedback on the same. The “5-second test” is an interesting method where you show the site or page to an audience for 5 seconds and let them recollect what they saw to ascertain if the message that you want to convey hits the mark in that time period.
Customer Feedback: Lastly you could ask the customers as soon as they buy as to what caused them from almost making the purchase. This is a more useful audience than people who dropped off the buying journey since you know that the customer has successfully completed the purchase cycle and his reservations could be better handled once you know it from them first hand.
A/B Testing
Though you would get a lot of insights from the above sources, there would still be a few that need testing and there can’t be a better set of data you can collect than from A/B testing. In essence, you pit two options regarding any element of the website against each other and see what works and the winning option is then expanded across the full audience for optimum results.
A few things to take care of while doing these tests are,
- The sample size should be high enough
- The task should be spread across a few weeks
- The tests should be quantitative and not subjective
- The test audience should be your customer and prospects
It is important that you expect not to get decisive feedback, but you need to persist and eventually, you will get better. You should keep in mind that the process is meant to eliminate the bias and so you should run with it and let the data do the talking. The process is worth its time and more as it lets you understand the data and messages that come along with it.
To sum it all up, going through all this process will for sure make you a better marketer, business owner, or professional. You get more authoritative and confident of your ideas since they are thoroughly defined and tested. You can see a clear shift in the company culture where decisions will be data-driven and less arbitrary or whimsical. To top it off, you can do this with all the cool funky tools yourself without having to involve the IT which makes speed to implementation that much quicker. It is a creative as well as a scientific effort that will eventually result in more leads, sales, and growth. You cannot ask for more, can you?
A brief note on the instructor, Brian Massey. The way he knits and weaves the concepts together and intersperse them around quick stories and anecdotes make for very engaging listening. And it makes assimilation of the concepts simple and smooth. Looks like I’ll have a lot of fun learning in the coming days.
