A full conversion rate optimization plan isn’t something you can do on your own. Here’s what you CAN do instead.
Special Note: This article was inspired by an episode of our podcast, Drive & Convert. We go into much more detail here than we did on the show, but you might enjoy listening to the original conversation and subscribing to receive future episodes.
I recently tried a large do-it-yourself (DIY) renovation. It involved stripping and refinishing an 8-person dining room table, painting and sealing chairs, and, oh yes, a bit of reupholstery. The Pinterest posts said it would take “just a few steps” and I could “even do it during naptime.”
One month, $250, a pile of second-hand tools later, and I was sitting on the garage floor feeling way out of my league. I was covered in dust from the sander, splotched from days of stripping away old stain, and speckled with bits of polyurethane.
And I wasn’t even close to done with the PREP work, let alone the refinishing.
Painting walls and switching out cabinet hardware were home improvements I could manage. But this? This, I decided, was the kind of project better left to the trained carpenters of the world.
For ecommerce store owners, true conversion rate optimization is a bit like that project — you can try to do it yourself, but it’s going to take longer, be more expensive, and deliver a less impressive outcome than hiring a professional.
Yet at the same time, this doesn’t mean you can’t make other important upgrades that improve life around the house. In fact, as a store owner, there are plenty of conversion rate improvements you can make to improve your customer experience on your site and see higher conversions in return.
This playbook outlines how to identify and take advantage of these improvement opportunities with your existing resources. I won’t promise you can do it all during nap times, but I will guarantee you won’t wind up covered in sawdust or sitting discouraged on the garage floor.
Here’s what’s ahead:
- Why CRO is worth your attention but hard to DIY
- The difference between CRI and CRO
- 3 steps any ecommerce store can take to identify issues
- How to skip straight to a list of prioritized, high-impact improvements
Why conversion rate optimization is worth your attention (but hard to DIY)
With so many competing business needs — Accounting, Product Development, Marketing, Logistics, Customer Support — it’s worth asking why conversion rate optimization (CRO) is worth your limited time and energy to begin with.
Some quick math will help. Let’s say you’re getting 30,000 site visitors a month, you average order value is $50, and your conversion rate is 2%. If you bump that up that conversion rate just one percent, you’re going to add $15,000 a month in revenue.
If you increase your conversion rate from 1% to 3%, you’re going to double your monthly revenue from $30k to $60k.
Pro Tip: This screenshot is from a tool that we built to help our clients understand the value of CRO. You can enter your own details and play out what life might look like when you implement the recommendations in this article. Onward!
Increased revenue is at the heart of better CRO, but that’s not all it impacts. In our 11 years of experience, we’ve seen taking steps to improve your conversion rates also has attractive secondary benefits like:
- Improved customer acquisition and retention
- Increased brand exposure
- Reduced customer service workload
- Reduced customer churn
Which is why, for modern ecommerce brands, having a CRO program is no longer a competitive advantage — it’s table stakes. If you’re not investing in it, you’re going to get left behind.
What all goes into conversion rate optimization?
Before we get much deeper into CRO, let’s clear up what it is and isn’t. Plenty of tools claim they help with it and there’s no shortage of consultants who claim they’re good at it. You’ve likely heard the term before.
However, they’re not always talking about the same thing we are.
Conversion rate optimization isn’t:
- Plugging in a few tools and hoping they work
- Using your gut to tweak parts of the website
- Fiddling with 50 shades of blue for buttons
- A passing marketing fad
- Only focused on cart and checkout
- Optional for brands who want to succeed
- A one-person show
True conversion rate optimization is:
- Data-driven (changes aren’t made on hunches)
- Reliant on a scientific process
- Concerned with high-impact areas
- About making your site work better for customers
- Improving a variety of conversion points, including sales
- Table stakes for ecommerce companies
- Dependent on a team of experts
Meaning, true CRO is a scientific process, not a checklist. It’s more like the Scientific Method you learned in High School biology than the tried-and-true recipe from a Chrissy Tiegen cookbook. It includes audits, assessments, hypotheses, tests, and analysis. It’s getting 1% better every day, week, and month, and seeing compound growth. When done right, it always builds exponentially and generates a high ROI.
However, it’s also intensive; one person can’t tackle it on their own.
Why you need more than Iron Man to take on CRO
Brands often come to us and say, “Hey, I have one staff member who’s a conversion optimization specialist, but the needle just isn’t moving the way we want it to.”
That’s because not even an Iron Man type genius can single-handedly craft a full CRO program; one person can’t be an expert in every area that’s important for true CRO. Rather, these brands need The Avengers — an entire team of experts with unique super-skill sets all working together to rid your site of sub-optimal customer experiences.
This is why Amazon’s conversion rate optimization team is over 100 people. They have the equivalent of a tiny town focused on nothing but optimizing the Amazon experience and looking at every data point. They have:
- data scientists to spot trends
- conversion strategists to come up with ideas
- test developers to build out variants
- experts in user testing to see how customers respond
…all working full-time to get customers where they need to go as quickly and easily as possible. Any one of those individuals wouldn’t get Amazon very far on their own. But as a team, they’re able to do some pretty amazing things.
Now, if you’re running a small ecommerce store, your reaction at this point might be a growing sense of overwhelm — “if Amazon has 100 people, what on earth could I do in a few hours a week?” True, there’s a lot that goes into a full-fledged CRO gameplan (it’s why we’ve been in business for 11+ years!) and you can’t do it all yourself.
But this doesn’t mean your brand is sequestered to the shadow of Amazon and other retail giants.
There’s another path you can take. When you’re starting out, you can work on conversion rate improvement. That’s something any ecommerce company, of any size and maturity, can focus on right away.
Improvement vs. Optimization — and what it means for smaller brands
Full-fledged conversion rate optimization, stretching from the first impression to the post-purchase journey, isn’t something one person can DIY; conversion rate improvement, on the other hand, is.
Sound like splitting hairs?
It may look like a linguistic trick, but there’s a conceptual difference between conversion rate improvement and conversion rate optimization:
- Conversion rate improvement is using a variety of tools, best practices, and a bit of the scientific method to improve conversions and traffic. It’s suited for smaller ecommerce brands and sites with under 50,000 unique monthly visitors because it doesn’t rely on statistical significance.
- True conversion rate optimization comes after improvement. It relies on the scientific process and a team of experts. It’s best suited for ecommerce brands that have 50,000 unique visitors a month and product-market fit. That’s because you need a lot of traffic to see statistical significance (high confidence test results aren’t luck or chance) in your experiments.
Put another way, Improvement is the street smaller brands take as they travel toward optimization.
With improvement, you can start collecting the data you’ll need to do true CRO when you’re ready for it. In the meantime, you’ll also:
- Gain a better understanding of your customer
- Move closer toward product-market fit
- Start seeing a sustainable return on ad spend
- Work toward statistically significant traffic (so you can take on optimization)
Below is how you get started with conversion rate improvement. You begin by gathering better data (just below) and then using that data to run smarter tests on your site (further down).
Let’s start with gathering better data.
3 ways you can gather better website data this week
When you have more time than money, you want to start with tools that let you collect rich data about your visitors and customers 24/7/365.
You’re inside your business day in and out. This makes it nearly impossible for you to know what it’s like to experience your site as a visitor. You’re “inside the jar,” so to speak, and can’t read the label from where you sit. You need quantitative and qualitative data to help you do that.
There are three ways to gather that data:
- Track your site data with a suite of Google Analytics reports — don’t worry, we’ll show you which ones.
- Implement user engagement tools to figure out where visitors are getting stuck or confused.
- Talk with your customers first-hand to see things from their perspective (remember, you may use your product, but you are not your customer).
You’ll want to do all three of these, so let’s unpack each one.
1. Track your site data with 5 Google Analytics Reports and features
Many ecommerce store owners have Google Analytics (GA) set up, but they rarely touch it once it’s running. We get why.
If your day job is founder and not analyst, the Google Analytics dashboard can be about as overwhelming as climbing behind the wheel of a Boeing 747; the screens and buttons clearly do something, but it’s less obvious what and why it matters.
Luckily, there are five features and/or reports you can use to find signals in all that noise:
- Enhanced Ecommerce
- Goal Flow report
- Users Flow report
- Channel Groupings
- Custom alerts
These will help you see what’s happening on your site in real-time, as well as create a timeline of data you can look back on in the future (high-five to your future self!). Below is an overview of each report, plus links to more details and setup guides.
This is the mother lode of ecommerce analytics information. Once you activate this, you’ll be able to dive into:
- Overview and product performance: revenue data and conversion rates, average order value, and refunds.
- Shopping analysis: how often and how long each page is viewed, what products customers add or remove, abandoned cart data, and completed transaction info.
- Product list performance: information about which pages and features drive the conversions you want to see.
- Advanced perks like tracking internal promotions, affiliate codes, and coupon codes
Plus, once you have this up and running, you’ll be able to answer questions like:
- What is my average order value?
- Which products are becoming more or less popular?
- Which product pages lead to the most conversions?
For more guidance on Enhanced Ecommerce, check out this help doc from Google and our in-depth guide:
Goal Flow report
The next thing you’ll want to look into is Goal Flow reports. In Google Analytics, a goal is an activity the visitor completes, such as making a purchase, submitting a survey, or signing up for a mailing list. The Goal Flow report shows you whether your visitors are achieving those goals.
Looking at this report will help you answer:
- Where are visitors starting to work toward a goal?
- Where are they getting stuck once they’re on their way? (If customers are stuck, so is your revenue.)
- Where are they abandoning a path or circling back for more info?
For example, a Goal Flow report might show you lots of people are adding items to their cart and starting the checkout process, but most of them drop off before completing their purchase.
From there, you might explore ideas like:
- Is my payment form broken?
- Does my payment form have too many unnecessary fields?
- Am I offering enough payment options?
- Should I consider offering monthly payments instead of 100% up front?
These ideas can form hypotheses and determine what tests you run to see if certain improvements get more people to complete their checkout going forward. (More on that whole process later.)
For help setting up goals and navigating the Goal Flow report, check out these help docs:
Users Flow report
While the Goal Flow report looks at how visitors march towards a specific goal you’ve set up, the Users Flow report looks at how visitors move through your site at large. This makes it a good complement to the Goal Flow report.
Most founders operate as if their visitors go from the homepage, to a category page, to a product page, to their shopping cart, to checkout, and to a thank you page.
In reality, visitors might:
- click a facebook ad and land on a product page
- then click on a related product and explore that
- then check out a blog post and then click on a product reference in that blog pots
- then click to explore your bundles or a subscription offering
- and ultimately leave without purchasing.
There are an infinite number of possible interaction paths your visitors can have with your site, but the User Flow Report will show you the most common (and probably unexpected) ones.
Think of it this way: if you’re looking at a real-time map of an amusement park, the Goal Flow report traces the path a family takes from a roller coaster to the gift store, while the User Flow report looks at how everyone mills about the park in general.
With the User Flow report, you can answer questions like:
- How many of my visitors are coming from each traffic source (social media vs. blog vs. affiliate sites)?
- How much traffic is coming from my email campaigns?
- How many page views are individual pages getting? Which are most and least active?
For more help navigating the Users Flow report, check out these pages:
- Google’s guide to the Users Flow report
- Our guide to Users Flow report
- The poetry of User Flow: The Straight Path to ROI
Not every visitor who lands on your site is the same. Some are curious window shoppers who clicked over from the Instagram ad, some came from your email list and know exactly what they want, and some are researching a specific type of product and found you through Google.
Where these visitors come from shapes how they shop and engage with you. Wouldn’t it be nice to know what that looks like?
Channel Groupings can pull back the curtain. In GA, a default channel groups all the traffic coming from a specific medium.
For example, “Organic Search” is a default channel, and it groups all the traffic you earned from search engines like Google, Yahoo, and Bing. Likewise, “Social” is a channel, and it groups the traffic you earned from Facebook, Instagram, and other social media platforms.
Custom channels allow you to create unique groupings so you can drill down into those lumped channels even further. For example, that Facebook ad you want to run next week.
This will help you answer questions such as:
- How much traffic is my Facebook ad driving to my site?
- How many visitors are coming from organic searches?
- What percentage of my visitors are referred by a specific partner?
For guidance on custom Channel Groupings, check out this help doc:
Chances are, there are a few metrics that really matter to you right now. Custom alerts let you track those very-important metrics and any activities that relate to them.
Not only does this save you time digging through the various (and seemingly endless) screens in GA, it can alert you to troubling scenarios (e.g. a Facebook ad stops working) so you can fix it ASAP instead of days later.
Plus, with custom alerts, you can answer questions like:
- Has my traffic significantly spiked or dropped this week? What about this month?
- Have any of my conversion points jumped today?
- Have customer returns increased or decreased substantially this week?
To set it up custom alerts for your important metrics, check out:
Those are the five places we recommend you start. Now, If you’ve poked around GA much, you know there are many more reports and features stashed all over the place. But remember, the goal is to highlight meaningful signals and tune out the noise. Start with these 5 reports and features to do exactly that.
2. Track user engagement with heatmaps and related tools
Google Analytics is going to help you see what is happening on your site. GA tells you about the numbers — how many visitors, how many sessions with at least one product added to cart, which products are gaining in popularity, etc.
To understand the people behind the numbers you’re going to need a few more tools.
That’s where user engagement tools come in. These tools help you see what visitors notice, click, and get hung up on when they visit your site. Meaning, they help you color in what you notice in GA and identify high-priority improvements with more confidence.
If you’re new to user engagement tools, one of the best places to start with is a heatmap.
Heatmaps use colors to show you where people are clicking, hovering, scrolling, and/or focusing their attention on specific pages. They’re useful for understanding what customers are noticing and where they’re getting stuck.
Tools that offer heatmaps
- Hotjar: One of the most accessible and powerful tools out there, Hotjar helps you see what visitors see (via heatmaps), record real visitor behavior on your site, and survey visitors based on what they do. Hotjar has some usage limitations on its free plan, but they are generous enough to get you started. If you like it and want to collect more data, you can upgrade for just $39/month.
Other options for heatmaps include:
- CrazyEgg: Access heatmaps, website recordings, and click reports. CrazyEgg offers a free 30-day trial and then pricing starts at $24/mo.
- Clickheat by Dugwood: A slim tool that gives you a heatmap with hot and cold click zones. It’s not robust, but it’s free thanks to open-source software.
- Inspectlet: With this tool, you can record up to 1,000 sessions per month (see every mouse movement, scroll, and click) and receive heatmaps…all on the free tier. That’s hard to beat!
- Feng-GUI: This AI tool simulates the first 5 seconds of someone interacting with a page or design. Starting at $67/month, it’s the most expensive heatmap option we’ve listed. A big difference with this tool is it’s predictive, meaning it simulates how visitors will engage instead of collecting data from actual visitors.
If you’re already using a heatmap tool and looking for other ways to gather engagement information, check out these additional options:
Layering insights: other tools that de-mystify visitors’ experiences
- Fullstory: Fullstory collects “an ocean of interaction points” and then makes sense of them for you. They emphasize helping you see the entire customer experience and exactly where you need to improve. This is one of the most powerful tools on the list — and also one of the most expensive. There’s a free 14-day trial but then you’ll need to contact their team for business pricing.
- Surveys: Some of the tools above, like Hotjar, let you run surveys. When you’re not sure why customers are getting stuck or why they’re there in the first place, surveys can help you figure that out in the customer’s own words. (Note: We have even more survey tools listed over on our top 80 usability testing tools.)
Pick one or two of these powerful tools, then spend one hour each week reviewing the data alongside trends you’re seeing in GA. You’ll quickly:
- Identify a list of improvements you can make for higher conversion rates
- Gain a better understanding of what visitors want and are trying to do (which will help you with product-market fit)
- Figure out how your traffic sources behave and which ones are high-value (so you can drive more of that kind)
- Build a culture of understanding customers and prioritizing meaningful data
3. Interact with customers first-hand — they’ll fill in any remaining gaps
In addition to GA and user engagement tools, there’s one more way to find out how your visitors experience your site and what you need to fix — talk with them.
Well, it’s true you can record sessions in tools like Hotjar. These recordings let you see every scroll, pause, navigation choice, and click.
But there is nothing more frustrating than watching someone hit your site, explore around, add a product to the cart, fill out all of the form fields, get to the last step, and then back out without buying.
You’re sitting there screaming “WHYYYYYYY?!” at the screen. The session recording doesnt tell you — it’s just a recording.
Now, imagine if you were still right next to that person and you could say, “Hold on a second. May I ask why you’re considering walking away after all of the steps you’ve taken? What’s holding you back?”
That’s what customer interviews (and user testing) do for you.
In a recent Drive and Convert podcast episode, our CEO Jon recommended a simple way to do the customer interview piece:
- Go to your local coffee shop with your laptop.
- Find someone waiting and say, “Hey, can I buy you a coffee if you give me five minutes of your time?”
- Ask the person to complete a task on your website while they wait for their coffee. (Don’t sell them on your brand or product — that’s off-putting. Just ask them to do a task on the site.)
- Watch what they do and ask them non-leading questions about the steps they take.
- Thank them for their time and hand them their coffee.
If you repeat these steps for an hour or two and get roughly five people to do this, you’ll walk away with a laundry list of improvements.
The idea is you’re seeing how someone who’s never been to your site before navigates it. You’re not trying to get them to purchase; you’re simply watching what they do and seeing where they get stuck.
A great guide for running this type of thing is a book called The Mom Test by Rob Fitzpatrick. It outlines what questions to ask so you get helpful feedback instead of the mom-ish response, “Oh yes honey, it’s lovely. This is the most usable website I’ve been on all day.” (Flattering but useless.)
Bonus: If you have a customer service team, do this whole coffee shop thing and talk with your service crew. They’re interacting with your customers all day and can likely rattle off all kinds of persistent complaints and issues they hear. So, set up a reminder to check in with support each month. Or, if you’re still running support yourself, tag feedback in your software as “website improvements” so you can quickly find those when you’re ready.
So you have all this data…what next?
Between GA, user engagement tools, and facetime with site visitors, you’ll quickly amass a laundry list of items you could to improve. That’s great — you’re making progress!
But it can also feel overwhelming because now you’ll need to:
- Identify which changes are high-impact
- Prioritize the changes that could make the biggest difference
- Figure out how to test those changes
- Keep doing that each week
This where the scientific method comes in.
How to turn a laundry list of issues into more conversions
To turn the problem areas you identified above into improvements that boost conversions, you’ll want to use the scientific method (yes, that thing you first heard about in 5th-grade science) to hypothesize, test, analyze, and improve.
This is a much more structured and effective approach than the default spaghetti-on-the-wall method (aka making gut changes willy-nilly and seeing what sticks).
In most cases, a scientific approach will look like:
- Define high-impact issues that are hurting your conversions
- Examine the quantitative and qualitative data you’ve collected around those issues
- Create a testable hypothesis (one you can disprove with results)
- Form a structured testing plan (define what, who, how, when) that tests the hypothesis
- Analyze the results (Did you disprove your hypothesis? Gather support for it? Hard to say?)
- Learn from your test and identify where to go next
The one exception is when you have an issue that’s obviously, detrimentally broken. For example, if the “add to cart button” isn’t working, don’t formulate a structured testing plan — fix the darn thing asap!
But for other scenarios where you think you have an issue and suppose you know a way to solve it, use the scientific method.
Here’s an example of how this could play out for an ecommerce store:
Define a high-impact issue
Let’s say you’re looking at your Goal Flow report. (Remember, this shows how a visitor progresses toward a specific goal, or completed action, you want them to accomplish.) The goal is “checkout complete” and the expected way customers complete this goal is: Product Page → Shopping Cart (Shipping) → Shopping Cart (Payment) → Confirmation Page (Thank You Page).
You’re reviewing that flow, and you notice there’s this huge cascading red waterfall off the checkout page. Meaning lots of customers are adding products to their cart, but few of them are actually completing their checkout.
Examine the data to get a clearer picture of the issue
The quantitative GA data tells you there’s a dropoff, but you’re not sure why. So you turn to qualitative data in your user engagement tools to see if you can get a clearer picture. The heatmaps show customers are spending a lot of time hovering over shipping costs. On top of that, some quick research tells you something like 50% of customers abandon cart because extra costs are too high. You may have a culprit.
Create a testable hypothesis
Based on the data, you come up with the following hypothesis:
- Customers who are ready to purchase are leaving the cart page because shipping costs are much higher than they expect. If your brand lists shipping costs earlier in the flow, cart dropoff rates will improve over the next 2 weeks because the shipping costs are no longer a surprise.
Form a testing plan
To test your hypothesis, you could switch where you mention shipping costs and keep an eye on those dropoff rates. But remember, you won’t have enough data to see statistically significant results. Any improvement could be due to luck or chance. (A big reason why A/B tests aren’t a good option here.)
One alternative you may want to consider is running user tests to see how users navigate and your current path to checkout. For example, you could sit down with a few coffee shop visitors and ask them to purchase an item from your store and then pause them at the very end of the checkout process and say ‘At this point, would you be comfortable completing your purchase or not? If not, what’s holding you back?’
Keep in mind, you only need roughly 5 users to uncover 80% of usability issues and we’ve found that at just 3 users, we start to see patterns between sessions. For a deeper dive into running this kind of test, check out our 3 part series on how to conduct high-impact user testing:
- Part 1: Thoughtful panel selection
- Part 2: Formulating tasks and questions
- Part 3: Analyzing the results
Another alternative to A/B testing is adding surveys or a chatbot to your cart page, either in a subtle bottom right corner bubble (“hey, can I ask you a question?”) or in a pop up that appears when the customer moves toward the exit. What customers tell you in these surveys/conversations could be useful data.
Hojar is a great option for adding these types of surveys:
Choose a good time to test
When you pick a time period to run your test, keep your other business activities in mind — they could skew or influence results. For example, if you’re testing a checkout modification and you’re also running a Facebook ad campaign with a “Free Shipping on all orders” offer, that will skew the results.
Ideally, you want to run most of your tests during “calm” periods where things like big ad campaigns, website redesigns, product launches, etc. aren’t a concern.
After two weeks of user testing, surveys, and/or watching GA, you pull up everything you’ve gathered. At this point, you’ll either:
- Have evidence that supports your hypothesis. E.g. users in user testing all cited scary shipping costs and drop off rates improved quite a bit when you explicitly listed shipping options earlier on.
- Have evidence that disproves your hypothesis. E.g. users in the user testing all cited issues other than shipping costs — like an incredibly confusing interface, or the page takes f o r e v e r to load — and dropoff rates haven’t improved with upfront shipping costs.
- Have no conclusive evidence one way or the other. Not ideal, but it happens.
Figure out where to go next
Where you go from here largely depends on the type of results you saw:
- If your results support your hypothesis, you could keep the change and move on to your next high-priority improvement.
- If your evidence didn’t support your hypothesis, you can form a new one based on the results you did see or revert back to the original version
- And if you didn’t gather conclusive evidence, you could repeat the experiment, run it longer, or try a variation.
However your experiment turns out, remember you’re not looking for silver bullet solutions; you’re looking to get 1% better every week, month, and quarter. The goal isn’t running through a one-time checklist.
The goal is continuous improvement so you earn enough traffic to step into a full, expert-led CRO program and see even bigger gains down the road.
Wish you could skip straight to the ‘Action Plan’ and start making improvements?
While it’s a good idea to go ahead and have those GA reports and user engagement tools set up, there’s actually a way to skip digging through all the data and sussing out potential improvements for yourself.
If your site is getting under 40,000 visitors per month, we offer a Conversion Growth Assessment™ and it takes all the guesswork out of where you need to improve. Meaning you get to skip the whole, “is this broken and worth my time?” and jump straight to, “here’s exactly what to change and why it matters.”
When you sign up for an Assessment, within a week you’ll receive a:
- dedicated strategist
- detailed report identifying why your visitors aren’t converting
- prioritized checklist of items you’ll want to address
- 60-minute strategy session to talk through it all
We’ll also follow up 30, 60, and 90 days after completion to check on results and see if you have any trouble implementing changes.
It’s an approachable way to work with our experts who’ve helped big brands realize 10x returns — on a small business budget.