How to research, prioritize, and test improvements for your growth marketing [review]

How to research, prioritize, and test improvements for your growth marketing [review], In this blog post, you’ll learn:  How to research what to improve in your growth marketing efforts How to prioritize the tons of ideas you have to improve your marketing How to test mini experiments like a pro



6 Step method to determine “what” to improve, XL way of prioritizing which ideas to test first, How to run a kick ass tests to experiment





Section 1: How do you decide “what” to do to grow your business in the first place?  First things first, let’s review the concept of growth marketing  Then let’s talk about 6 methods to brainstorm what to improve


Let’s review the concept of growth marketing real quick:

What is growth marketing and how is it different than traditional marketing?

Growth marketing is a series of mini experimentation that you do that will eventually lead you to achieve your business goals e.g customer acquisition, retention or referrals. In some ways, you could say it’s “growth hacking” or  “marketing science”.

Another way to look at growth marketing is applying “lean methodology” to marketing. Where we create a hypothesis, design an experiment, and repeat the process till we achieve a desired goal

There are two key differences between growth marketing and traditional marketing:

  • Focus
  • Philosophy

Traditional marketing focuses on the top of funnel process e.g. generating leads or awareness while growth marketing focuses on everything end to end throughout the funnel. Traditional marketing emphasizes the “get it right as close as possible” attitude while growth marketing emphasizes “get it experimented quickly” mentality to marketing

If you want to know more about growth marketing – read this blog post we created that share more about it

How do you know what to improve in your marketing?

A lot of people will probably be focused on doing one of the following:

  • Copy what a competitor does
  • Look up articles online to figure out what to do
  • Brainstorm ideas and come up with ideas you heard from your friends or family

If there is one thing you’ll remember, remember this:

Understand your current situation and build from the ground up solutions or marketing ideas that are unique to you. In other words, ditch a mold and create your mold.

So instead, what can you do?

Understand your current situation and build from the ground up solutions or marketing ideas that are unique to you. In other words, ditch a mold and create your mold.


How do you understand what things you should improve on that is unique to you?

 The brilliant CXL institute has shared a fantastic framework called ResearchXL that shows key areas you can dive into to understand the current state of your business and what are possible things you want to improve on.

There are the following 6 areas for you to get started:

  • Technical analysis
  • Heuristics analysis
  • Digital analytics
  • Mouse tracking analysis
  • Qualitative analysis
  • User testing

#1 Technical analysis:

No, before you freak out – we’re not asking you to do complex high-level derivative maths, but instead, we just want you to ensure that your website is loading fast enough, and there is not a reduction in bounce rate from specific browsers.

To gauge your site speed, use free analytics tools e.g. Page Speed Insights, where you can plug in your site address and it’ll give you a score and tips on how to improve your speed for both mobile and desktop

page speed insight for google

#2: Heuristics Analysis

Before you get scared by the fancy term “heuristics” understand that it simply means “ a short cut”.

In this case, it means you are sitting down with someone you trust and going through one by one the pages on your site to check on the following:

  • Relevancy – does the page meet expectations between the titles, photos, contents? Are they coherent? E.g., if you’re selling cups but instead have images of bananas – that’s simply not relevant
  • Clarity – is the message on this page clear? A quick way to do it is to flash the message or the page to someone for 5 seconds and ask them, “what’s the one thing they remember.” If it’s not what you want them to remember – then you’ve got one improvement area to work on
  • Value – is your page actually communicating value to your user? Is it increasing the motivation of the user to do something?
  • Friction – the most underestimated force in the world is the concept of friction. Is there anything on this page that is causing some doubts or looks “sketchy” – if so, add it your bucket of things to improve on.
  • Distraction – is there anything on this page that is distracting you from doing the one call to action that you want customers to do. E.g a giphy image that distracts instead of complements the message? If so – add it to your bucket.

As much as heuristics is fun, don’t forget doing it in groups in going to be a lot more comprehensive and fun instead of doing it yourself. Even better, rope in your friend who’ve been your customer to help out.



# 3: Web Analytics Analysis

Now is the fun part. If you don’t know what Google Analytics is or how it works – drop everything you’re doing and look it up. In simple terms, Google Analytics give you the ability to track and analyze users behaviors on your website. Think of it as an invisible spy that records every click, every interaction and when you leave the site. These information can be used to improve your goal

So before you get drowned on what to look for, here are some key questions you should have at the back of your mind answered:

  • What page has the best “time on page” ranking
  • What page has the worst “bounce rate.?”
  • Where are my users from?
  • Do they use mobile or desktop to view my contents?

If you find anything here, add it to your bucket of things to improve.


#4 : Mouse Tracking and Analytics

Image if someone could show you precisely what each user did on your site like a video feed telling you if they scrolled or not, or did they stop to read something or not. Well thanks to technology, this is now possible. Our personal favorite tool is “Lucky Orange” that shows you this.

So what can you do with it? For starters, you can optimize placements of your call to action. If the heatmap shows that many people actually spend staring a the corner of your site > then place your button that says “click here” right on that spot.

how to use heatmap and scrollmap to improve conversion

Also, having a video feed on how the users behave will tell you some key things like did they scroll fast or slow, did hover around the sidebars or not? It’s the perfect observation of how people behave without influencing them.

#5 Qualitative Surveys

That simply means going back to the good old days of asking people to fill up quick survey forms on why are they leaving the site, or maybe asking them to fill up key things that are important for them for you to focus on.

Some pro tip : there are two school of thought when it comes to surveys on whether “open ended” is better than “closed ended”. Personally, try both and see which sticks.

If you’re confused on what to ask in these surveys refer to customer profile to understand their fears, motivations, and hopes for you to be more relevant and persuasive.

#6 User testing

Big companies usually hire customers to come over or do certain tasks on their site before launching a website or improving it. For example, ask them to buy a green jeans and check out or ask them to fill up their customer details – collect feedback on the specific task you’ve asked them to do and spot if anything is glaring. If it is – then add it to your bucket of things to improve on.

Now for the fun part (almost fun part)

Now that you’ve exhausted multiple things for you to improve on, now it’s time to collect and consolidate all those ideas into a gigantic spreadsheet. Carefully label these “ideas” into 5 categories:

  • Test - If there is something you feel could improve conversion and you need to test it out, label it under this category
  • Instrument – this is just another way of saying you need more “data” on this issue. A solution to these kinds of improvements could be adding pixels or google tags to measure carefully the problem you’re trying to solve.
  • Hypothesize - This is where we’ve found something, where you know, is a problem but you simply can’t find a clear cut solution and require you to come up with guesses.
  • Just Do It – JFDI. This is a bucket for issues that are a “no brainer”. Solve them asap.
  • Investigate - You need to dive deeper on this problem and set aside time to go over it or craft a plan on how you want to tackle it.


 How to priortize growth marketing ideas

Now is where the reality check comes in. You’ll probably end up having hundreds of ideas in this list the question is, are you going to go and test them all out?

The most commonly known methods to prioritize ideas are the PIE and ICE method.

The PIE method

The pie method looks at three key variables:

  • Potential – How much improvement can be made on the pages?
  • Importance – How valuable is the traffic to the pages? (amount of traffic, etc.)
  • Ease – How complicated will the test be to implement on the page or template?
The ICE method
  • Impact – What will the impact be if this works?

  • Confidence – How confident am I that this will work?

  • Ease – What is the ease of implementation?

Priortizating gorwth marketing

The problem with these frameworks – it’s subjective

Yes, think about it. How do you assign a score for ease of implementation? Based on your gut feeling? What about impact? Who decides what impact? Customers point of view or your point of view? The truth is, both these frameworks provide an easy option to prioritize the tens of ideas that you have but on the flip side it is nothing more than taking your gut feeling and applying a number to it to make it more numerical to convince people.

So what’s an alternative?

The XL prioritization method by the CXL institute shows us a different way to remove “gut feelings” from the equation by using this:

how to priortize growth marketing method

  • Is the change above the fold? → Above the fold meaning areas within your website that are the first thing people see when entering your site are noticed by more people, thus increasing the likelihood of the test having an impact
  • Is the change noticeable in under 5 seconds? → Show a group of people control and then variation(s), can they tell the difference after seeing it for 5 seconds? If not, it’s likely to have less impact
  • Does it add or remove anything? → Bigger changes like removing distractions or adding key information tend to have more impact
  • Does the test run on high traffic pages? → Relative improvement on a high traffic page results in more absolute dollars

One key difference between this framework

Other frameworks you assign a “score” from a scale of 1 to 10 to gauge importance, this framework relies on a “yes” or “no” option denoted with 1 or 0 to gauge rank. This leaves behind any interpretation and gives us a clear cut way to prioritize.


how to do experiments for growth marketing

So now that you’ve brainstormed ideas, and prioritized them, time to test out possible solutions to see if you can actually boost your marketing growth.

Here are the top 5 mistakes people make when testing their solutions.


#1 people think tests are stupid

If you’re a seasoned entrepreneur, you might have a confidence bias where you think you have it all figured out. But remember our rule of thumb – every situation is unique and requires a tailored solution. A/b testing solutions is the only way to figure out what works and what doesn’t with certainty.


#2 Sample size is too low

How do you expect to come up with a realistic conclusion whether your idea worked or not based on 10 users – you can’t. The smaller the number the less stable the conclusion.

So how do you know what number to test? Well, thank god there are tons of online resources like this that tell you what the exact sample size you need to run your ab test is. Now, if you’re a small business owner, skip this. Because in reality is you won’t be able to spend hundreds of bucks finding traffic if your site is just beginning – aim for the maximum number of customers you can test on and go for it. It’s okay, don’t cry.


#3 Test don’t run long enough

As much as it hurts, we are guilty of this as well. On our quest for fast results, we tend to stop test midway if it doesn’t work. The reality is, user behavior is different for the most time. Weekend traffic might be different than weekdays. The only way to ensure consistency is normalizing your testing duration over a longer period for at least a week to cover both weekdays and weekend. If you’re a gigantic company, extend it further to a month to “smooth” our certain weeks in the month to take into consideration “pay week” vs “non pay week”


#4 Giving up after the first hypothesis fails

So you plan a test, test it out, and you tanked. You cry and call yourself a failure. No, you’re not, you need time and multiple attempts to get it right. Do you think billionaires got it right the first time around? Hell no, they all tried and tried and pivoted before finding the right solution to the problems they are solving. So don’t give up, keep trying.


#5 Not testing all the time

Testing is a habit you must have. If you don’t you’ll end up not improving over the long term. Here is something for you to think about:



Conclusion: now that you’re armed with ideas, prioritization, and way to test things out – how are you going to up your game? Tell us.



Older Post Newer Post

Leave a comment