How does Peep Laja perform CRO?
2 Easy lessons below
If you’re a part of the digital marketing field, you would have come across CXL or Peep Laja at least once in your professional journey.
In case you haven’t, Peep Laja describes himself in the following way on his website:
Hi, I’m Peep Laja. I’m a former champion of optimization and experimentation turned business builder.
I’ve been following his work for a few years now and I agree with that definition entirely.
I was recently watching his videos about how to start a CRO process and the various steps involved in the same. The various points mentioned are actionable and also practical.
Let’s dive in:
Lesson 1 — New Business
If you’re a brand new business i.e., you have no data, no customers, no website visitors — then do not, I repeat do not go for any conversion optimisation.
The most important thing you need to do is talk to customers. Talk to as many as you can that fall in your TG.
You can get valuable insights that can help optimise the experience on your website.
For example, Food on the Table shadowed a mom for three weeks and monitored her cooking habits, how she picked up groceries at the local supermarket, etc. They implemented all the learnings into their website and they clocked over 1 million visitors in a matter of months.
Lesson 2 — The Research XL model
After years of working with leading companies on CRO, Peep Laja designed the Research XL model.
It details the various steps and instructions you need to follow to build and implement a successful CRO operation.
The most important takeaway was that you need the necessary data for CRO, not every data available out there.
For example, if you’re crossing the road and all data is given to you — angle of the sun, color of the buildings, number of doors in the opp. building — but all this is useless. If the info given is if any cars are coming or not, you can take necessary action about crossing the road.
Step 1 — Heuristic Analysis
This refers to the website experience as a normal user. You should go through the website page by page and make notes.
Make separate columns for each device (phone, desktop, tablet etc.)
Check all pages for the following:
a) Clarity
Which info is crystal clear and which is not
b) Friction
Which parts of the website make it difficult for the user to take action
c) Anxiety
Which parts of the website create anxiety. Example — oh, so much data to fill
d) Distraction
Are there any portions of the website that cause distraction to the user
Step 2 — Technical Analysis
At times, even if the content is perfect, if the site doesn’t load properly or has other issues, CRO will still be affected.
In technical analysis, test the following:
- Cross Browser Testing
- Cross Device Testing
- Speed Analysis
The page should load fast (10 seconds is too long), it should work on different browsers (you don’t know which one your user will use), and it should also work across devices.
Step 3 — Digital Analytics
Once you have completed the technical analysis, you need to look at the digital data. Some of the points to note:
- Where is the flow stuck?
- Where is there a lot of friction?
- Where are people dropping off?
- What type of behavior correlate with more purchases?
A person who purchases, searches for x,y,z item, they use a,b,c filter, they research d,e,f before making the purchase
Ensure everything the user does on your website is being measured
Digital analytics can give you a lot more information and details about the happenings on the website than you would be able to measure on your own.
Step 4 — Qualitative Research
- Online surveys — survey people who are on the website
Ask them if they’re willing to buy or if there is anything stopping them from buying
On a t-shirt page the concerns can be will it fit me, can I return it etc.
These questions give credible insights that can be used for UX
- Follow-up survey
Ask people a week after they made the purchase
What doubts they had, what difficulties they faced, what helped push them over, why did they end up buying, etc.
This can also help improve the experience
Step 5 — User Testing
User testing is one of the key ways you can understand how your audience navigates through your website
Get someone in your TG or at least someone you know (friends, family etc.) Ask them 2 things:
- It’s your birthday, buy something on this website
- Find a pair of jeans, 34 size, dark blue color
These two tests are very different and will lead to different experiences.
Always remember:
What users say vs what they do is very very different.
So observe what they do and make notes!
Step 6 — Mouse Tracking Analysis
There are multiple tools (hotjar for example) these days that you can use to measure heat maps, mouse clicks etc.
In most cases, the mouse tracking is not of much use because people don’t usually scroll where they are looking. You don’t read an article and move the cursor as you read.
Check session replays, you might see something that’s a friction and not allowing your users to complete an action
What to do next?
Okay, so now you have collected the data, and understood the various issues. What’s next?
First step — Categorise
Rank each issue in the order of priority — 1 to 5
★★★★★
This rating is for a critical usability, conversion or persuasion issue that will be encountered by many visitors to the site or has high impact. Implementing fixes or testing is likely to drive significant change in conversion and revenue.
★★★★
This rating is for a critical issue that will be encountered by many visitors to the site or has a high impact.
★★★
This rating is for a major usability or conversion issue that may not be viewed by all visitors or has a lesser impact.
★★
This rating is for a lesser usability or conversion issue that may not be viewed by all visitors or has a lesser impact.
★
This rating is for a minor usability or conversion issue and although is low for potential revenue or conversion value, it is still worth fixing at lower priority.
After that, categorise them into different buckets. This can be:
- Instrumentation — all metrics not mapped
- Known problem but not sure how to solve
For example, A pool parts company. The biggest issue was that customers didn’t know whether the pool part would fit their pool or not.
The solution to this requires multiple teams (product, developer, design, marketing etc.) so it will be in the hypothesize category
3. Investigate, we know there might be an issue here but we need to dig deeper and investigate
4. No brainers — These are the issues which need to be resolved right away. For example, the issue is that the grey text of 8 pixels on a black background is not readable. This doesn’t require an A/B test, it just needs to be actioned.
Make a spreadsheet and make a note of all the problems
Measuring the Effectiveness of a Testing Program
There are 3 metrics to follow when measuring the effectiveness of a testing program
- Testing velocity
How many tests are you running in a month, year etc.
If you can be running 5 tests in a month but test only 1, it’s bad
2. Percentage of tests that provide a win
Most tests end up in either failure or an insignificant win. This is because most of the tests out there are pointless. And people run wrong kinds of tests
3. Impact of a successful win
This depends if the test gave a 5% increase or a 20% increase
If the site is optimised poorly in the beginning, the avg. impact goes down. As the site improves over time, the % impact will start reducing.
This topic is explained in detail in the Researching and Testing lesson of CXL Institute’s Growth Marketing Minidegree. The lessons in the minidegree are crafted nicely with tons of additional reading material that enabled me to understand the concepts in depth.
