skip to main content

How to Find the Blind Spots in Your Customer Journey

Nov 21, 2026

Ecommerce Optimization

Nathan Pitchan

Quick Summary

How to Find the Blind Spots in Your Customer Journey

This article draws on Episode 3 of the DTC Revenue Rocket Podcast, recorded live at Shoptalk, where Klaviyo’s Gina Potenza answered the question “Where do I start?” with a spicy take: even if you could get anything you want out of Shopify, what would you ask for? Her point is that you cannot know what to ask until you know what you are trying to find out. Going through customer data without a plan is like grocery shopping without a list, and the question nobody thought to ask is the one that catches you halfway through cooking.

The post walks through mapping the customer journey as the questions a customer is trying to answer at each stage, sorting what a team has confirmed from what it only believes or cannot yet explain, deciding which blind spots are worth investigating, and finding the evidence to answer them. It closes with a one-page template. The main takeaway is that the goal is not more data. It is enough evidence to make one decision well.

  • Map the journey as customer questions, not channels A channel map shows what the brand sends. A map written as what the customer is wondering at each step, such as “What does this candle smell like when I can’t smell it through a screen?”, shows what the brand still needs to understand about them. Start with one customer group tied to a decision coming up.
  • Beliefs feel like knowledge, which is what makes them risky Every answer falls into confirmed, believed, or can’t explain yet. The believed group deserves the closest look because it shapes decisions without anyone checking it. A brand that believes holiday buyers purchased for themselves will send a “time for another one” discount to people who may have bought gifts.
  • Only chase the blind spots that would change what you do For each open question, ask whether knowing the answer would lead to a different plan. If three likely answers point to three different actions, it is worth investigating. If none would change what gets sent, it can wait. Rank the survivors by the value of the decision and how hard they are to answer.

Walk into a grocery store without a list, and the trouble starts in the first aisle. There are forty kinds of pasta sauce. You stand there longer than you’d like to admit, pick one, and do the same thing in the next aisle, and the one after that. By checkout, your cart is full of things you never planned to buy: crackers that were on sale, a snack you saw someone else grab, a second jar of peanut butter just in case.

Then you get home, start on dinner, and realize the recipe needs an ingredient you never thought to look for. It wasn’t on your mind in the store, so it never made it into the cart. Now you have two options: accept that dinner won’t be quite as you planned, or begrudgingly go back to the store.

And then it hits you, punctuated by a facepalm: I really should’ve made a list. The game plan for next time? First, you decide what you’re making. Next, check what’s already in the kitchen. Whatever’s left is the list: only what dinner needs and you don’t have yet. You buy less, skip whole aisles, and nothing catches you off guard halfway through cooking.

Customer data can feel a lot like that trip without a list. Shopify shows your orders and repeat purchases. Klaviyo shows who opens, clicks, and buys. Your ad platforms and analytics add traffic, conversion rates, and a dozen more dashboards. Go through all of it without a plan, and it’s easy to get stuck comparing numbers, chase the ones that look interesting, and never notice the question nobody thought to ask.

On Episode 3 of the DTC Revenue Rocket Podcast, recorded live at Shoptalk, our CEO, Zak, hosted a panel with leaders from Klaviyo, Shopify, PostPilot, and Clear.io on how ecommerce brands are using AI to grow. Toward the end, they took on a question PostPilot’s Michael Epstein said people tend to get intimidated by: with so many tools and so much data, “Where do I start?” Clear.io’s Sebastian Brown Glad suggested connecting all your tools and putting that data in one place. Klaviyo’s Gina Potenza followed with what she called a “spicy take”: even with everything in one place, “if I could get anything I want out of Shopify, what would I even ask for?” 

Her point was that you can’t know what to ask for until you know what you’re trying to find out. So before you go looking in Shopify, she said to “take the time to map out your customer’s journey”: how you want them to find you, why they buy, and what brings them back. Then, in her words, “take it one level deeper.”

Writing out that journey is the “decide what you’re making” step. Once it’s written down, you can check what you already know about each part of it, and what’s left over becomes your list: the questions you can’t answer yet, and what to go look for in Shopify. You’ll find plenty of gaps. The ones worth chasing are the questions whose answers would change what you do next.

How to Map Your Customer Journey 

Customer journey mapping means laying out the steps a customer takes with your brand, from the first time they hear about you to the moment they decide whether to buy again. Most maps list those steps by channel: an ad, an email, a product page, a text. That shows what your brand sends. It doesn’t show what the customer is trying to figure out at each point, and that’s where the useful questions come from.

Start with one customer group. Mapping every type of buyer at once spreads the map too thin to tell you much. An easy way to choose is to start from a decision you have coming up. Say a candle brand is planning what to send first-time buyers after the holidays. The group to map is people buying from them for the first time this season.

On the panel, Gina started with the customer journey map almost every brand would sketch first: a customer finds you, buys from you, and stays a customer forever. “Fine, that’s not interesting, right?” she said. It’s true of every brand, so it can’t tell you anything about yours. Going “one level deeper” means breaking each step into what your customer is wondering at the time.

For the candle brand’s first-time buyers, that looks like this:

  • Finding you: Is this for me, or a gift for someone else?
  • Deciding: What does this candle smell like, when I can’t smell it through a screen?
  • Buying: Will it arrive in time?
  • Using it: How long will it last?
  • Coming back: Do I want another one, and which?

Not every customer moves through these stages in order. Some buy on their first visit, while others browse for weeks before deciding. The map doesn’t need to predict every path. It needs to capture what your customer has to answer before they move to the next step.

This is also a different exercise from a performance review like our Mid-Summer Ecommerce Checkup. That checkup asks where your marketing is underperforming. A map built this way asks what your customer needs to know at each step, before you decide what to change.

A map written as your customer’s questions shows you what you’ll need to understand about them.

How to Separate What You Know From What You Don’t 

On paper, this part should sound easy, right? And before anyone gets defensive, this isn’t a knock on how well you know your customers. Nobody knows what they don’t know, right? That’s exactly what makes it hard to find. You see the orders every day, read the reviews, and answer the emails, so it feels like you know your customers well. But go stage by stage and ask how you know each answer, and the line between what you’ve confirmed and what you’ve always believed gets blurry fast.

So go through each question on your map and ask how your team would answer it today.

  • Confirmed: You have evidence. If the candle brand’s Shopify orders show most first-time buyers chose standard shipping, it knows most of them weren’t in a rush.
  • Believed: Your team has an explanation it’s fairly sure of, but nobody has checked it. “Most of our holiday buyers are shopping for gifts” is a belief until something backs it up.
  • Can’t explain yet: You can see something happening, but not why. Most of last season’s first-time buyers never ordered again, and nobody knows the reason. Data is good at showing what customers did and much less reliable at showing why, which we covered in more detail in our post on behavioral data.

Every brand has answers in all three groups. That’s normal. Data piles up faster than anyone can ask questions of it, and plenty of what a team believes started as a sensible explanation that never got tested.

The second group deserves the closest look, because a belief feels like knowledge. It shapes decisions without anyone stopping to check it. If the candle brand believes its holiday buyers bought candles for themselves, it will plan a “time for another one” discount for January. If most of those orders were gifts, that email goes to people who never lit the candle.

If your team has trouble seeing past its own beliefs, Gina had a suggestion on the panel: use an AI assistant to “interview you… to figure out your blind spots about the customer journey.” Describe your map and ask it to question each answer: How do you know that? What would change your mind? Treat its questions as prompts, not evidence, since it knows nothing about your real customers. And keep customer data out of any tool your company hasn’t approved.

Anything in the last two groups is definitely a blind spot.

Which Blind Spots Are Worth Investigating? 

Once you’ve sorted your map, you’ll likely have more open questions than time to answer them. That’s fine. Most of them don’t need answering right now.

To find the ones that do, ask one question of each: If we knew the answer, would we do something different?

Take the candle brand’s biggest open question: why didn’t last season’s first-time buyers order again? There are at least three likely answers, and each one points somewhere different:

  • They bought it as a gift. A “time for another one” discount won’t reach anyone who used the candle. The better move is reminding the gift-giver before the next holiday.
  • They’re still burning it. They haven’t gone anywhere. They just aren’t ready yet. A reminder timed to how long the candle lasts makes more sense than a discount in January.
  • They didn’t like the scent. Then the problem started before the purchase, and clearer scent descriptions or a sampler set would do more than any follow-up email.

Three answers lead to three different plans. That makes it a blind spot worth investigating, because without the answer, the brand would have to pick one plan and hope it’s the right one.

Compare that with another open question: which day of the week most first-time orders came in. The answer might be interesting, but none of the likely answers would change what the brand sends in January. It can wait.

When several blind spots pass the test, rank them by two things:

  • How much the decision is worth. A question tied to your biggest upcoming campaign or largest customer group comes first.
  • How hard it is to answer. If two questions matter about equally, start with the one you can answer from information you already have.

The blind spots worth chasing are the ones standing between you and a decision.

How to Find the Evidence You Need 

By now you have a short list of questions worth answering. The next step is working out where those answers might be. Start with what you already have, and only collect something new when you have to.

When Your Existing Data Can Answer It

Start with the question, not the dashboard. For each possible answer, ask what you’d expect to see if it were true, and then go look for exactly that.

Here’s how that works for the candle brand’s question about first-time buyers:

  • If they were gifts, you’d expect orders where the shipping address doesn’t match the billing address, gift notes if your checkout offers them, and orders bunched in the weeks before the holiday. All of that is in your Shopify orders.
  • If they’re still burning the candle, you’d expect repeat customers to take longer to reorder than the time that’s passed so far. Shopify’s Customer cohort analysis report groups customers by the month of their first order and shows how many came back over time. Klaviyo also shows each customer’s average time between orders once your account has enough order history. If repeat buyers usually come back after about 60 days, a December buyer who hasn’t reordered by mid-January isn’t late yet.
  • If they didn’t like the scent, you’d expect it to show up in reviews, support conversations, and return reasons. Support conversations are often the most direct source, which we covered in our post on support tickets.

Each source shows part of the picture, and none of them settles the question alone. A mismatched address suggests a gift, but it doesn’t prove one. When several sources point the same way, you can move that answer from “believed” to “confirmed.”

If you know what you’re looking for, the data you already have can usually answer more than you expect.

When You Need to Ask Customers

Some answers aren’t in any system. Your orders can show that a candle shipped to a different address, but only the customer can tell you it was a gift for their mom. Your returns can show who sent a candle back, but not who disliked the scent and kept it anyway.

When that happens, ask directly and keep it short:

  • A one-question survey after purchase. “Who was this for?” with two options: me, or a gift.
  • A short email to first-time buyers asking how they liked the scent.
  • A few conversations with real customers, if the question is big enough to justify the time.

Ask only what your question needs, and check whether an email or form you already use can ask it before adding a new tool. Until the answers come in, keep the belief labeled as a belief on your map. A handful of responses can point you in the right direction, but they might not speak for every customer.

The goal isn’t more data. It’s enough evidence to make one decision well.

A Customer Journey Mapping Template for Ecommerce Brands 

Put it all on one page, and the map becomes something your team can use. Here’s what the candle brand’s map might look like by the end:

Journey stage What we don’t know What we believe Evidence needed Decision it informs
Finding you Who are holiday buyers shopping for? Mostly gifts Shipping vs. billing addresses, gift notes, a one-question survey Whether January emails speak to the buyer or the person who received it
Deciding Can shoppers tell what a candle smells like from the page? The scent descriptions are clear enough Reviews and support questions about scent Whether to rewrite scent descriptions or offer a sampler set
Buying Why do abandoned checkouts rise in mid-December? Shipping cost When abandonment rises compared with shipping cutoff dates, a short exit survey Whether to promote a shipping deadline or a free-shipping threshold
Coming back Why haven’t first-time buyers ordered again? They need a reminder Cohort report, time between orders, scent reviews, a survey A gift-giver reminder, a reminder timed to burn time, or better scent guidance

To use it, copy the column headings, pick one customer group, and fill in one row per stage. Once you have an answer, make the decision it was meant to inform, then look back after a few weeks to see whether it worked. If it didn’t, the question goes back on the map.

How ECD Helps Brands Turn Customer Questions Into Marketing Decisions

A list makes the trip easier, but it doesn’t cook dinner. Once you know why customers do what they do, that answer still has to show up in what they see: the product page, the ad, and the email that lands in their inbox in January.

That’s where we come in. Our team works across Shopify development, paid media, email, and SMS for ecommerce brands, so when a customer question gets answered, we can act on it in whichever channel it points to. If the candle brand learns most of its holiday buyers were shopping for gifts, that changes its January emails. If it learns shoppers can’t tell what a candle smells like, that’s a job for the product page. Because we work across all of those channels, the answer doesn’t stall with whichever team happened to find it.

ECD is a Klaviyo Master Elite Partner, one of about 35 agencies worldwide out of more than 22,000 to hold that status, and Klaviyo named us its Agency Partner of the Year for the Americas in 2026.

Before you decide what to change in your marketing, figure out what you still need to learn about your customers. When you’re ready to turn those answers into a plan, we can help.

Get Your Free Revenue Forecast 

Frequently Asked Questions

How should an ecommerce brand map its customer journey?

Start with one customer group rather than every buyer at once, since mapping everyone spreads the map too thin to be useful. An easy way to choose is to start from a decision coming up, such as what to send first-time buyers after the holidays. Then, instead of listing steps by channel, write what that customer is trying to figure out at each stage. For a candle brand’s first-time buyers, that might be “Is this for me or a gift?” while finding the brand, “What does it smell like?” while deciding, “Will it arrive in time?” while buying, “How long will it last?” while using it, and “Do I want another one, and which?” while coming back. The map does not need to predict every path. It needs to capture what the customer has to answer before moving to the next step.

How do you separate what you know about customers from what you only believe?

Go through each question on the map and ask how the team would answer it today, sorting answers into three groups. Confirmed means there is evidence, such as Shopify orders showing most first-time buyers chose standard shipping. Believed means the team has an explanation it is fairly sure of that nobody has checked, like “most of our holiday buyers are shopping for gifts.” Can’t explain yet means something is visible in the data but not the reason, such as most first-time buyers never reordering. The believed group deserves the closest look because a belief feels like knowledge and shapes decisions unchecked. Gina Potenza also suggested using an AI assistant to interview the team about its blind spots, treating its questions as prompts rather than evidence and keeping customer data out of tools the company has not approved.

Which blind spots are worth investigating first?

Ask one question of each: if we knew the answer, would we do something different? In the candle example, the question of why first-time buyers did not reorder has at least three likely answers, each pointing to a different plan: they bought gifts, so remind the gift-giver before the next holiday; they are still burning it, so time a reminder to how long the candle lasts; or they disliked the scent, so clarify scent descriptions or offer a sampler set. That makes it worth investigating. A question like which day of the week most orders came in would not change what gets sent, so it can wait. When several questions pass the test, rank them by how much the decision is worth and how hard they are to answer, starting with the one that can be answered from information already on hand if two matter equally.

Where can a brand find the evidence to answer its customer questions?

Start with existing data by working backward from the question: for each possible answer, decide what you would expect to see if it were true, then look for exactly that. For the gift question, that means orders where shipping and billing addresses differ, gift notes, and orders bunched before a holiday. For the still-burning-it question, Shopify’s customer cohort analysis report and Klaviyo’s average time between orders show whether a customer is actually late to reorder. For the scent question, reviews, support conversations, and return reasons are the place to look. When no system holds the answer, ask customers directly and keep it short: a one-question post-purchase survey, a brief email asking how they liked the scent, or a few conversations with real customers. Until answers come in, keep the belief labeled as a belief on the map.

What does a customer journey mapping template for ecommerce look like?

The template is a single page with a row for each journey stage and five columns: what we don’t know, what we believe, the evidence needed, and the decision it informs, alongside the stage itself. For example, the “Coming back” row for a candle brand might read: why haven’t first-time buyers ordered again, believed to need a reminder, evidence needed from the cohort report, time between orders, scent reviews, and a survey, informing a choice between a gift-giver reminder, a reminder timed to burn time, or better scent guidance. To use it, copy the column headings, pick one customer group, and fill in one row per stage. Once an answer comes in, make the decision it was meant to inform, then look back after a few weeks to see whether it worked. If it did not, the question goes back on the map.

Written by: Nathan Pitchan

Full-time daydreamer and professional persuader, Nathan is a fearless word alchemist crafting copy that connects, converts, and feels undeniably magnetic. E-commerce. Food & beverage. Horticulture. Destination tourism. And beyond. In a world overrun by AI-generated fluff, his secret sauce is unmistakable: engaging, conversational, and deeply human storytelling. Why? Robot copy gets read. Human copy gets remembered.