Customer Persona Generator for Shopify

A persona is only as good as the evidence behind it. Asking a model to invent "Sarah, 34, loves yoga" produces something that looks finished and tells you nothing you did not type in. So personas here come out of the Tilly store audit: it reads your public pages and the reviews on them, groups what customers say into distinct reasons for buying and distinct worries, and names each group.

It looks for four to six groups. When the reviews only show one or two kinds of buyer, it splits by situation instead (who the product is bought for, the occasion, the problem it solves) so each persona is a different reason to buy. A persona with thin evidence behind it is marked low confidence rather than padded out.

• A plain name that describes their situation, not a made-up person: "The cautious gift buyer", not "Emma".
• Their pain points, ranked by how often and how strongly they come up, with the quotes behind them.
• What makes them buy, the objections they raise, worded the way a customer would ask them.
• The language they use, and language to avoid with them.
• Message angles for each pain point, ready for an email, an ad or a product page.

Take the fictional candle shop above. Several reviews mention buying for someone else and being nervous about the scent, and a question on the product page asks whether it will arrive before a birthday. Those signals become "The cautious gift buyer": pain, worry that it arrives late or smells wrong; cares about, delivery dates, scent notes in plain words and easy returns.

One message angle for that pain point: "Order by Thursday, and it arrives by the weekend. If they do not love the scent, returns are free." It turns into a line under the add-to-cart button, a subject line for the abandoned checkout email and a headline for a gifting ad, all answering the same worry. The quotes and the shop are made up; the method is the real one.

Write each email, product page or ad for one persona at a time. The cautious gift buyer needs delivery dates and an easy return; the repeat buyer needs something new or a reason to restock. When one message tries to speak to both, it usually speaks to neither.

In Tilly, confirmed personas feed every agent: the email marketing agent writes flows for them and the email calendar plans sends around them. More on the method on the customer personas page.

• Demographics instead of motives. Age and city rarely change what you should write; the reason they buy and what worries them does.
• Personas nobody checks. Read the evidence behind each one and delete any that do not match what you hear from customers.
• Personas you cannot act on. If you would write the same email for two of them, merge them.
• Making them once. New products bring new buyers. Run the audit again when your range changes.

From public pages alone it cannot see order history, so it cannot tell you which persona spends the most. A store with very few reviews gives it little to work with, and the personas will say so through their confidence rather than fill the gap with guesses. Connecting Shopify adds order patterns to the evidence.

Personas are built by the full audit, which reads a few hundred pages and runs the analysis. It needs an account and a saved card; the card is not charged. The first pass of the audit, with no account at all, shows what to fix on your pages and products.