SponsoredSynergy45Stronger looking, Thicker looking Hair SerumVisit Synergy45

Your business here

This example comes from the tool maker’s own case study, not an independent source.

An Apparel Brand Sends 40% Fewer Emails Using AI Segmentation, With Claude Building Segments

Criquet Shirts used Orita's machine-learning scores through Klaviyo to reach seasonal buyers, cutting send volume ~40% as revenue rose 30% in three months. Claude now builds location segments in 10-15 minutes and drafts emails.

Who this is about: Apparel brand (Criquet Shirts, Austin, Texas; one- to two-person retention team)

Get the prompt

What it takes

Cost
Free planClaude: Free plan. Pro is $20 a month billed monthly ($17 a month billed yearly), and Max starts at $100 a month. Team seats are $25 a month billed monthly.You also need Klaviyo and, for the scoring, Orita; the article gives no costs.Price checked Oct 24, 2026 on claude.com. Prices change; check before you buy.
Setup time
A few daysOur estimate from the source, for one business.
Skill needed
Some tech skillsComfortable installing an app, pasting an API key or running a command.

Explain it simply

A clothing brand's one-person email team stopped blasting everyone and used an AI scoring tool to pick who should get each message, including people who only buy in certain seasons. It sent fewer emails and made more money. The team also uses Claude to build mailing lists by location and to write first drafts.

New to AI? See how AI works

What was done

Retention lead Zach Scheimer first ran bucket (A/B) tests across email and text, grouping subscribers by engagement, to show leadership that fewer, targeted sends could work.

His 15 hand-built segments, based on 180-day engagement, missed seasonal buyers, so the brand used Klaviyo's integration with Orita, a machine-learning tool that scores each customer individually.

For a Father's Day push, he had Claude build segments of subscribers within 15 miles of the 215 stores that carry the brand.

The team is training a Claude copywriting skill on six years of email and text briefs in Google Docs, with access to Shopify product descriptions, to produce first drafts he edits.

How it went

Klaviyo reports send volume down about 40% and revenue up 30% in the first three months of the scored approach. The first location segment took 30 minutes while he learned to prompt Claude; a later one took 10-15. First drafts save him 10-20 minutes of digging through past emails.

  • This is Klaviyo's own article, so it's promotional: Klaviyo sells the platform and the Orita integration.
  • The 40% and 30% figures come from the Orita scoring and testing, not from Claude; the AI summary at the top of the article blurs the two.
  • He says AI isn't great at creative work and only gets you to a first draft.

Prompt

PromptWritten by our editors from the source
I need a segment for [campaign]: subscribers within [15] miles of any of the store locations in [file of addresses].

Use the location fields in our [email platform] profiles. Exclude anyone who [unsubscribed from SMS / bought in the last 14 days]. Tell me how many profiles match, list any stores where the match count looks wrong, and give me the segment definition to paste in. Don't send anything.

Setup notes

  1. Test before you pitch: a bucket test on engagement groups gave leadership evidence that sending less could work.
  2. Give the copy skill your real past briefs and product descriptions, and keep a person writing the final version.
  3. Add a post-purchase question (theirs: is this a gift?) to find customers who want fewer messages.

Business fit

Who this helps, and how:

  • Apparel and seasonal brands: send to the customers most likely to buy now instead of the whole list.

    Shopify & ecommerce
  • Brands sold through retail stores: build segments near stockists for last-minute in-store pushes.

    Shopify & ecommerce

Related

Similar tools

Other tools for this job, and how they differ. Chosen by our editors, never paid for.

“Vendor or secondhand” (grade C) means: A tool maker's case study, a promotional demo, or someone else's write-up. How we grade. Spotted a mistake? Suggest a correction.