Conversion Research & Testing - A CXL Institute Review

Conversion.

We all want it! We want web visitors to turn into leads, to turn into demos to turn into sales.

Everything we do at the core of business has to do with conversions.

Converting an SEO ranking to a click (from the right person).

Converting an above the fold visitor to go on and visit the sit more.

A blog reader, to clink internal links to dig into your knowledge hub.

To download a gated piece of content.

To interact with your brand.

To talk to a sales rep.

To convert from interest to purchase.

Everything has to do with conversions, conversion rates and the experience we offer our prospects in order for them to convert.

Here's the thing.

The way we get there normally tends to be wrong.

I'll never forget my very first Director of Sales role. We were having a down month and we couldn't figure out why.

People were downloading the content, they were interacting with us on social, having conversations with our reps, those conversations were going well, but when we got to the purchase portion of the funnel it came to a screeching halt.

Why?

We had no idea. Well... that's a lie.

We had TONS of ideas. So we tested ALL of them.

The first didn't work.

No stress. There's plenty more where that came from.

The second... nope.

Third?... No luck.

You tracking yet? We weren't getting anywhere.

Website copy changes, script changes, style changes.

All either doing zero for us or actually negatively impacting our conversion from one stage to the next.

Fast forward 30 days later.

We had our best month ever.

It was baffling. I had no idea why. And just so you're aware, to this day I still have no idea why.

We changed so much that we couldn't pin point the thing that moved the needle, and because we didn't have a process to the madness, the win got lost in the shuffle.

I tell you this story to let you know that there's a better way to optimize your conversions, whether it be on your site or even with your sales team.

I learned it this week, and I really look forward to sharing it with you today. It's called ResearchXL by CXL.

Instead of testing via anecdote and praying to the conversion gods, you can put scientific analysis to work and stop banging your head against the wall.

Without further ado..

ResearchXL

....Oops... Quick Caveat!

In most cases, the tests that we do as a business are what we would call an A/B test, a method of comparing two versions of a webpage, app, script etc against each other to determine which one performs better.

Here's the deal, if you're just getting started and have no customers, what you need is customer development – figuring out which product to build and for whom.

Most businesses fail because they build something that no one wants. Customer development is what helps you minimize risk and improve chances that this new business will make it.

Conversion research and testing is great when you have customers and you've proven product/market fit, but until then...

Talk to People.

One of the first things I did when I got to HALOS was speak to 15 of our ICP. A few were clients, but most weren't.

For the past 3 weeks I've been shadowing the COO of a major MSO in our market.

Listening in on operations calls, watching him do what he does, discussing possible issues with his current day to day and the things he wishes he could fix.

The last thing I'm focused on right now is conversions on our website (we're doing a full redesign of that anyway).

So before you continue reading. If you don't have a customer base, or a strong web presence, it's important that you go and test product market fit and you do that by talking to your ideal clients.

Anywhoser.... Back to...

ResearchXL

So ResearchXL actually has 6 steps that help you optimize your conversions. No more guessing or tossing stuff against the wall hoping that it sticks.

Those steps are:

1. Heuristic Analysis
2. Technical Analysis
3. Digital Analysis
4. Qualitative Analysis
5. User Testing
6. Mouse Tracking Analysis

If I lost any of you non CRO pro's, don't leave just yet baby birds, my goal is to feed you too, and for all of my sales followers that decided to read up to this point, don't worry, I'm going to tie it into your world as well.

So let's roll.

Step 1. Heuristic Analysis

Essentially heuristic analysis is an experience based assessment of your website (or for you sales folks, your process).

So the idea here is to walk through your site or your process piece by piece as a normal user would. And assess each single piece for a certain set of heuristics.

1. Clarity - Is everything as clear as it possibly could be?
2. Friction - What is causing doubts and hesitation in the person? What's holding them back from moving forward?
3. Anxiety - Are you asking too many personal questions? Do they not believe your free trial is free? Do they not trust you?
4. Distraction - In this process every piece should have a single goal. Are you getting too convoluted with your information?

So page by page on your site, or piece by piece in your process you'll identify all of these elements and move onto step 2.

Step 2. Technical Analysis

Next is a technical analysis. It doesn't matter how persuasive your website is, if it doesn't work on the device or browser that someone is searching they'll never get to be persuaded.

So in this stage, you'll actually be testing your site on certain browsers and devices specifically and testing the conversion rate on each.

Does Safari convert 20% less than Chrome? How about your mobile users? Are they converting at a higher rate than your desktop?

This is simply making sure that you don't have any kinks in the armor.

If my sales peeps are still alive and well, and reading, this is where you would take into account, your scripting and how it's being portrayed.

Tone of voice, speed of speaking, so actually going beyond the actual words your team are saying and digging into the technical aspects of what a successful call or meeting sounds like.

Once you have this data you'll move onto step 3 which is Digital Analytics.

Step 3. Digital Analytics

This is where we really find where the problem lies in the process. So we go to the analytics.

Right here we want to know where the bottleneck in your process is. Where do people exit the website? Where is your sales process stalling? Where are they dropping off?

This is where we pin point the friction point.

In this same stage, we'll start to segment who is dropping off. What persona is dropping off more than others, is there a stage that they drop off at a higher rate? And what behaviors correlate with more purchases?

So if someone is on your site downloads a certain e-book, or they come back to the site a certain number of times after having a demo? What behaviors correlate to becoming a customer?

This does mean that everything you want to find out, from finding the bottleneck in the process to behaviors that lead to buying need to absolutely be measured in some sort of way, whether that be through an analytics platform like Google Analytics, or call recording, email click through tracking etc.

Once you've got this solidified it's time to move onto Step 4.

Step 4. Qualitative Research

When we enter this phase we're going to be surveying our customers and also doing user testing.

The first survey is an on-site survey that may simply ask "Is there anything holding you back from downloading our e-book?"

On the site, we do want to do this on every page. Each page is going to have different concerns and we want to track what those concerns are.

If they're on the "Download this E-Book" page their concern is going to differ from the homepage or the pricing page. But we want to track all of it and understand all of it.

The next survey is the "Follow-Up Survey" to people that have just become customers.

***Sales Peeps*** this would be a Jobs To Be Done Survey!

In this we really want to understand their experience on the site, with our reps, and what made them decide to move forward.

Step 5. User Testing

This next step is actually having people interact directly with your site.

So have your clients go through your site and give them tasks and watch what happens.

Give them a task to book a demo and see what they do, actually watch their moves and how they interact with your site.

Do they read certain parts of the page, go to a blog, go to pricing? This will guide your UX design.

This is extremely important for B2B SaaS companies. See how people are interacting with your page. What page do they finally book a demo after. How did they get there? If it's deep within the site, would it benefit to move that to the homepage?

As for sales teams, this could simply be speaking to your prospects and asking them what they liked and disliked about the process. What was confusing, what was great. You'd be surprised how open your prospects are, especially if they've already blown you off.

Remember, in everything we do, we want to take away as much friction as possible, and that's what this step is all about.

Step 6. Mouse Tracking

Finally we've got mouse tracking or basically seeing what people are doing on your site.

Mouse tracking can provide valuable insights into viewing and information processing patterns.

The site tracking that I've used and think is really useful in finding out what is working and what isn't is called HotJar.

They have a scroll heat map, so you can see if the bottom of your pages are actually even being seen, and the session replays.

These are massively important so you can see what people are actually doing on your site. Where are they dropping off, what aren't they doing that you'd expect etc.

For my sales peeps. This is listening to calls. It's that simple. Dang I'm good at this stuff ;)

Now What?

So now you've got a ton of data. There are some issues you've got to take care of and you don't want to end up like Ned (that's me) a few years back not knowing what to do.

Your job here is to rank these issues in order of priority.

ResearchXL recommends a 5 point ranking scale.

5 is a severe issue. Something that is simply hemorrhaging money and it's affecting a massive amount of people that interact with your brand.

1 would be an issue that should be fixed, but it can go on the back burner for now.

There are 2 criteria that are more important than others when giving a score:

1. Ease of implementation (time/complexity/risk). Sometimes the data tells you to build a feature, but it takes months to do it. So it’s not something you’d start with.
2. Opportunity score (subjective opinion on how big of a lift you might get). This depends on how many users are exposed to the issue, and how close to the money the issue is. Let’s say you see that the completion rate on the checkout page is 65%. That’s a clear indicator that there’s lots of room for growth, and because this is a money page (payments taken here), any relative growth in percentages will be a lot of absolute dollars.

Essentially: Follow the money. You want to start with things that will make a positive impact on your bottom line right away.

Once you've prioritized your tests it's time to actually get testing!

Go and update what needs updating, testing what needs testing and modifying the things you saw that needed tweaking.

What you'll find over time is that this process gets easier and easier. It may seem choppy today as you're reading this magnificently written blog article, but after 2 or 3 tests, you know how it's done, what to do and where to go.

The one thing I'll leave you with today is this.

Don't test just to test.

Too often people say that A/B tests don't work, or don't move the needle for their business. I'd beg to differ.

When I dig into the material of that statement, I usually find that they're testing things that their customers could care less about.

The key is to really dig into the issues your customers have by actually speaking to them, don't just guess and use anecdotal evidence to run tests you think are going to move the needle.

If you do that, you'll be like Ned (still me) from 2017 banging his head against a wall not knowing what was going on.

Don't be like 2017 Ned.

Be Like 2020 Ned.

Listen to Peep and the crew at CXL and run ResearchXL.

Until Next Time.