Returning Customer Recognition

Returning customer recognition means knowing that a visitor is someone who has already given your brand their details, so the site can treat them accordingly. It works from your own first-party information, and it leaves strangers as strangers.

How does recognition work?

Stores generally have three routes to it, each with its own strength.

Route	How sure you can be	Main limit
Store login	High	Most shoppers don't log in
Tracked email click, such as Klaviyo's click identifier	Fairly high, though emails can be forwarded	Only works for clicks from that email
Identifier stored in the browser	Reasonable on a personal device	Browsers remove them, and it stays on one device

Each route ends by matching the visit to a profile you already hold. If no profile exists, recognition has nothing to match against. That's what separates it from website visitor identification, which tries to name people who haven't given you their details.

What gets in the way?

• Browser limits. Safari ITP deletes script-set cookies after 7 days of browser use without interaction with the site. A customer who reorders every six weeks may look new. Server-side cookies help on the same device, with exceptions.
• New devices. A shopper who starts on a laptop and opens your email on a phone has no stored identifier there. Device recognition can't help, and the email click or a login has to do the work.
• Consent. If a shopper declines analytics on your banner, what you can store changes. Recognition shouldn't try to get around that.

What does recognition make possible?

Knowing a visitor is a past customer changes what a useful page looks like. A new visitor needs reassurance and an introduction. A repeat buyer often wants the same thing as last time, with less searching. Good uses include:

• Showing their usual products or what's due on any page
• Restoring a saved cart after an email click
• Holding back messages meant for first-time buyers

Match what you show to how sure you are. A shared family tablet should be a reason for caution, so keep order history for person-level signals such as a login or an email click, not device recognition alone.

How does reOtter approach it?

reOtter recognizes people who've already given your brand their details, without a third-party network, in a record that doesn't depend on a browser cookie surviving. Strangers aren't identified. Its on-site widget, a paid feature, shows recognized customers their usuals and due items on any page with no login, capped in frequency and with a holdout to compare against. On the cart recovery page, a held cart comes back when reOtter knows who clicked. When the click matches a customer, the page also greets them by name and shows Buy it again, with their past purchases and anything due. They don't need to log in.

Key takeaways

• Recognition matches a visit to a profile you already hold, so it doesn't name strangers.
• Logins, email clicks and browser identifiers each help, and each has limits you should plan around.
• What a site shows should match how confident it is about who's there.

See how reOtter restores the cart → Abandoned cart recovery