My journey started off like pure design and then moved into front-end development. For some time, I stayed in front end development and then went to user experience, user research and similar roles. I fell into CRO through sheer boredom because one thing that I found is that CRO kind of appealed to me. With CRO, I was able to see the full cycle of actually seeing the impacts of the work we as a team were doing. That was the learning cycle for me. It’s interesting that you build something and test it out and you get the outputs and then that kind of keeps users engaged and it keeps you learning too.
That's the kind of thing that really gave me a kick being able to see that full cycle. I was head of optimization at Train Line and later set up a company. My present company is a SaaS company.
It's different for every website because sometimes you'd have a lot of data already approaching a brand new website. So, approaching a brand new website from scratch, I'd look into the data. User research and user feedback are really good sources of information. Analytics data that currently exists like Google Analytics, funnel reports, checking bookings, heat maps— all these are sorts of different segments to get some idea of how users are behaving on the site.
If you have access to client's data, you can come up with all of these experiment ideas and then after getting access to the data, work out an insights document for them, which is kind of like doing a deep dive into seeing where the traffic is coming from. It may also include getting the hold of some user feedback, user research, talking to the customer service department etc.
I would always approach it from a data first perspective, just kind of like creating an insight document. So, just go through step by step checking every single aspect of the site that tells you how various users are using the site.
It's different for different companies and different pages of that company as well. For instance, a company I was consulting with was a university. They were creating a website for teaching cybersecurity. So that was definitely information architecture. They explained to me how they had to teach new users, where to find stuff like how to use a website, what the products are all about etc.
Users always need some level of support. For them, it was information architecture while for others that can be different kinds of problems but the problems would be pretty much different.
So , I would say it kind of varies but information architecture is probably a main thing and so is the clarity of value proposition like getting rid of distractions. All that kind of stuff is pretty key to start off with.
My focus has mostly been CRO processes. We can put the process into place to come up with ideas and to increase conversion rates. It’s all about improving the product once we get the traffic. I've got sites where I’ve been trying to drive traffic and so I can tell you some of the learning.
It's quite easy to drive traffic from social media campaigns like Facebook ads but it’s also important to make them stick and make sure that the traffic is relevant. Focusing on content marketing drives better quality traffic.
First of all it’s about figuring out the goals of the website. Once that’s done, it’s important to track those goals and to see how those goals are performing. Not just in terms of the end metric, like for example, a conversion rate, but all sorts of metrics in between.
If you've got a funnel, then what's the conversion, what's the shape of the conversion funnel ( for example ) what’s the kind of user behaviors and where’s the traffic coming from?
So where is traffic coming from is important because then you can get a sense of what their expectation is when they land on the site and how you can achieve relevance for them.
What I actually do is basically I've got a set of conversion leavers that I've used in the past to kind of focus the thinking.
There's a conversion value proposition, clarity, relevance, social proofing, authority, confidence — a set of conversion leavers that I've known modifying or focusing on those specific areas from those specific looking at the site from those specific lenses can help increase the, the conversion rate, or at least come up with hypothesis as to what may or may not be working.
You also need to come up with ideas about experiments. Then once you've got some hypothesis then it's all about figuring out what the goal of the company is, what hypothesis fits within the goal of the path that the company wants to take? And how do you go about prioritizing all of these findings? That's the way I would go is to do some, it could just collect data, analyze it and then ideate, and then run some tests validate from those validations, go through the whole cycle again.
So you basically ideate and develop the tests and test out in the world, get the results and then let that reshape your thinking. If at all it needs to kind of come up with new hypotheses. We just go again.
The best thing is identifying what is a distraction and that’s a straightforward task, because something you might consider a distraction may not be one. So you need to validate all the hypothesis.
The next thing I would do is try and gather some clues to figure out can we gather some evidence to see what could be a distraction, like a clear way a process that has worked for me is to think about the product or the business as a whole and what the goals are for a particular page or the entire website. Then look at the data to figure out if that is true.
For example, if the business wants users to click through and go through a particular funnel, but uses all sorts of other different things. In this case, you’ll have to try to consolidate and focus on the information.
It all boils down to gathering evidence as we go and moving stuff, honing and making the content clearer and more specific and creating more clarity etc.
Getting test ideas is a messy exercise. One can go write down these ideas as and when they get. However, I do follow a process. We used to follow a process to get teams to ideate together. A version of design thinking can be useful.
1. Firstly it’s about understanding what a user desires from the website and what’s his behavior on it. After this, you can break the work into a list of tasks.
It varies from team to team. It also depends on OKRs that helps in ascertaining the team’s focus on various features of the product. Regarding prioritization, I recommend David’s post. There are so many different prioritization frameworks. The bottom is understanding the prioritization metrics like PIE. There are certain criteria for which you want to score your test ideas against.
Chris Gaut came with the concept. I always use it as a starting point and modify it for specific company.
Potential- the page with the highest drop off. Rank all your pages in terms of most and least performing.
Importance- how much useful traffic are you getting. Rank from highest to lowest.
Ease- How easy it is to develop each test.
Give an overall score.
So like the above example, companies have developed their own prioritization metrics for experimenting with ideas. In the end, it’s all about working within those constraints. Read more about prioritization in this article.
There’s CRO and the guys at Spiro. Apart from this, there are a whole lot of people following LinkedIn where you get a lot of high value content.
On Fridays, there's a conversion world round table, which I attend. The CRO experts from all sorts of fields come together and we discuss topics and talk about our experiences.
Additionally, I follow Craig Sullivan. These are the sources that I believe are good for learning about CRO.
If there's one piece of advice that I wish I had got at the beginning of my career in marketing is to try my best to get a mentor who can help in my learning.
I would say try and get a decent mentor if at all possible. Also, one thing about CRO is that we don't ever really know what's going to work. However, there are sort of best practices as a baseline that one can always learn about.
For example, when we are dealing with products, it's this, the union between the user and the product. And so learning about user centered design is a really useful thing to research.
So, for me it’s learning about the users and getting a mentor are keys to a great career in marketing and CRO.