How to find your Shopify customer personas
Build customer personas from your own Shopify orders, segments and reviews: who buys, why they buy, what nearly stopped them, and the words they use.
What this is for
A persona is a short description of one group of buyers: why they came, what they worried about, what made them order and the words they use for it. Built from guesses, it produces ads that sound like everyone else's. Built from your orders and reviews, it tells you which angle to write first.
By the end you will have two to four persona cards, each tied to a segment you can count in Shopify and to sentences real customers wrote. If the term is new, the customer persona glossary entry covers what one is and is not.
Time needed and what you need
Time needed: about 3 hours. Building the segments takes half an hour; most of the time goes on reading reviews and support emails properly.
- Staff access to your Shopify admin with permission to view customers and create segments
- Your product reviews exported to a spreadsheet, from whichever review app you use
- The last few months of customer service emails or chat transcripts
- A blank document for the persona cards
The steps
Work from numbers to words: the segments tell you which groups are big enough to matter, the reviews tell you what those groups want.
1. Decide what the personas will be used for
Write one line at the top of your document: these personas are for ad angles, or for the welcome emails, or for rewriting product pages. The use decides the detail. Ad angles need the trigger and the doubt; a product page needs the questions people ask before buying.
2. Split buyers into one-time and repeat in Shopify segments
Shopify's customer segments run on a small query language, so you can count groups without an app. Start with the simplest split: people with one order and people with two or more. The attribute behind it counts orders over the customer's whole history. source, checked 28 September 2026
Save both as segments and note the size of each. A group of repeat buyers is usually where one persona lives, because they have told you with their money that the product did its job.
- number_of_orders = 1
- number_of_orders >= 2
3. Count who buys each hero product
Build one segment per best-selling product. Shopify's reference shows the function, including a count so you can find people who bought the same item more than once. source, checked 28 September 2026
If two products have very different buyers, for example a starter kit bought once and a refill bought every month, you are probably looking at two personas rather than one.
4. Check Shopify's own groupings before inventing yours
Shopify already sorts customers by recency, frequency and spend. Open a segment on that attribute and look at which products your top group bought first: that first product is often the persona's way in. source, checked 28 September 2026
5. Read reviews for motive, doubt and result
Go through the review export and tag each review with three things: why they bought, what they were unsure about, and what happened. Copy the exact phrases into your document. "Bought it for my sister who is impossible to shop for" is worth more than any demographic guess.
- Motive: gift, replacement, upgrade, problem to solve, curiosity
- Doubt: price, size, quality, delivery time, whether it suits them
- Result: the sentence they use to describe it working
6. Add the questions from support
Customer service messages show what reviews leave out: the questions asked before an order. Tally the repeats. If the same pre-sale question comes up every week, it belongs on the persona card as that group's main doubt.
7. Group the tags into two to four personas
Sort your tagged reviews by motive and look for clusters that line up with a segment you counted. Name each persona after the motive, not an invented person: "the gift buyer" tells a copywriter more than "Sarah, 34". Drop any cluster smaller than a handful of reviews; it is an anecdote, not a group.
8. Write one card per persona
Keep each card to one page so people actually read it. Every line should trace back to a segment count or a quoted review.
- Who: the motive, and the segment and its size
- Trigger: what made them start looking
- Doubt: the main hesitation, in their words
- Proof: what settled it (a review, a guarantee, a photo)
- Words: three phrases lifted from reviews
9. Write to the persona without naming it
A persona tells you what to say, not what to call the reader. On Meta, an ad cannot claim to know a viewer's health, age, finances or similar traits, but it can still speak to them directly. Write about the product and the situation, and read the guide on the personal attributes policy before the first ad goes out. source, checked 28 September 2026
A worked example
For a handmade ceramic mug store with 2,300 customers
Worked example (an invented store with invented numbers)
The store's segments show 1,610 customers with one order and 690 with two or more, which adds up to all 2,300.
The owner reads 120 reviews. 46 mention buying the mug as a present, 38 talk about it surviving the dishwasher or daily use, 22 name a specific glaze they wanted to collect, and 14 fit none of these.
The gift reviews come almost entirely from one-time buyers, and the glaze reviews from the repeat segment. That gives three cards: the gift buyer (doubt: will it arrive in time and wrapped), the everyday user (doubt: will it chip), and the collector (trigger: a new glaze). The first ad set is written for the gift buyer, the largest group.
Common mistakes
Starting from demographics. Age and location describe who someone is, not why they bought. Two buyers of the same age can want the product for opposite reasons. Start from motive and add demographics only where they change the message.
Writing ten personas. Nobody can write for ten audiences, and the small ones are rarely big enough to target. Merge until each card has a segment you can count.
Paraphrasing the reviews. The value is in the customer's own words. Rewrite them into marketing language and you lose the phrases that make an ad sound like it came from someone who has used the product.
Never checking them again. Buyers change when you add products or channels. Re-run the segments and re-read the newest reviews every six months, and retire a card whose segment has shrunk.
Doing this with Tilly
Tilly's free audit reads your storefront, reviews and marketing and builds customer personas from them. It needs only your store URL; nothing has to be connected to run it.
The personas go on one brand profile that all 8 of Tilly's agents write from, so the email, ads and SEO work start from the same buyers. Treat them as a first draft: check each one against the segment counts from steps 2 to 4 and edit what does not match.
Questions people ask
How many customers do I need before personas are worth building?
There is no fixed number. What matters is whether each group you describe has enough reviews and orders behind it to be more than one person's story. With a young store, start with one or two cards and add more as orders come in.
My store is new and has no reviews. Where do I start?
Read reviews of similar products on marketplaces and competitors' sites for motive and doubt, and talk to the first twenty buyers. Label the cards as provisional and rebuild them from your own reviews once you have them. The free persona generator can give you a starting outline.
Can I use Shopify's predicted spend tiers for personas?
Use them to size and prioritise groups, not to describe them. A spend tier tells you who is worth more, not why they buy. Pair it with the review tags to get the why. source, checked 28 September 2026
See this on your own store
Paste your store URL. The first pass takes about a minute and needs no account. Save a card (nothing is charged) and Tilly reads the whole store and works out who buys from you.