Hire an AI repeat purchase improvement reviewer

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A repeat health score out of 100, with the three parts and the sums
• A verdict on last month's fix, either keep, drop or too early to tell
• The one stage to fix, with the cause and the evidence
• A plan for that fix, ready to approve, with its cost
• The one number to watch next month

How it does the job well

• Write for one named audience and one action. Each asset has a single job.
• Use the customer's own words from reviews, calls or surveys before inventing language.
• Every claim needs proof the business can show. Leave unproven claims out or mark them.
• Judge results against the business goal, such as profit or qualified leads, not clicks alone.

Mistakes it avoids

• Starting with email. The skill's own order is product, range, who you win, the buyer's experience, then messages.
• Changing several things in one month.
• Judging a fix on one month of one cohort.
• Quoting a vendor benchmark as a result you will get.
• Running this when most of what you sell is bought once, such as a sofa. Use it where a second order makes sense.
• Generic copy that could fit any brand.

What a good result looks like

An example, not your task: You sell things people could buy again and want to know where buyers drop off.

• Repeat health score. A score out of 100 with the three parts, the sums and the change from last month.
• Last month's fix. What you changed, the number it should move, what happened, and a call: keep, drop or too early.
• The stage to fix. The stage holding buyers back, the likely cause and the evidence for it.
• The one fix. A plan from the skill that fits the cause, with its cost, ready for your yes or no.
• The measure. The one number to watch next month, its value today and the group it covers.
• Not this month. The other causes it saw, in order, so next month starts from them.

How the work is checked

• The score shows its sums, and any part that was not scored is named.
• Only one fix was chosen, and the measure for it is named.
• Last month's fix was checked before a new one was chosen.
• No repeat rate was taken from a store tool's returning-customer percentage.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

A second opinion

Before it hands back, your AI reviews the draft the way Paul Graham and Clayton Christensen would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Read last month. Opens last month's score, the fix you chose and the number it was meant to move. On the first run, this month becomes the base.
• Score repeat health. Scores the repeat rate, the cohort trend and how many causes are clear, and names the stage that holds buyers back.
• Check last month's fix. Compares the number the fix was meant to move. It waits two months before calling a fix a failure, and says when the data is too thin.
• Choose one path. Picks the first-use check, the reorder plan, the delivery flow, the best-and-worst buyer profile or the top-customer touch plan, from the cause. If the cause is the product, it says so and runs nothing.
• Set the measure. Names the one number that will show next month whether the fix worked, with today's value and the cohort it covers.
• Hand back. Gives the score, the verdict, the plan and the causes left for later, then stops for your yes.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id monthly-repeat-buyer-keeper only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/monthly-repeat-buyer-keeper/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• An orders export with customer, date, value, discount code and refund flag
• Your products, and how long one unit lasts
• Returns, support tickets and delivery dates (optional)
• Reviews or survey comments (optional)

Helpful context (optional)

• Delivery dates against the promised date
• Return reasons and support ticket topics

Tools and access

• Claude Code: It can keep the repeat folder, read last month's files and run on a schedule.
• Your store or sales system: Export orders with customer id, date, value, discount code and refund flag.
• Shopify: Paid store platform. Its order export gives customer ids, dates, values, discounts and refunds.
• Klaviyo: Paid email tool with a small free tier. It gives flow and campaign revenue so you can see the email and SMS share.
• Any good chat AI: Any current chat AI that reads a spreadsheet can build the cohorts.
• Any good chat AI: Any current chat AI can sort messages onto a timeline. Paste the text, with dates.
• Shopify admin: Its order and returns export gives the order count and return reasons for the same window.
• Klaviyo: Click data on a how-to email shows who opened the setup guide.
• Klaviyo: Delivery, order-count and predicted next-order data can drive these triggers.
• Shopify: Order history decides who is recent but not loyal.
• Any good chat AI: Any current chat AI can run the skill from your product list.
• Any good chat AI: Any current chat AI can do the sums and write the messages. Paste real numbers.
• Klaviyo: Paid service with an account. It sends the update flow by email and SMS from your order events.
• AfterShip: Paid service with an account. It hosts a tracking page and sends delivery updates. Check the current price against your order volume.
• Gorgias: Paid service with an account. A helpdesk that can tag order-status tickets, so your WISMO count is real and not a guess.
• Any good chat AI: Any current chat AI that reads spreadsheets can run the ranking. Attach the export and check one group by hand.
• Shopify: A paid service that needs an account. Its order and customer exports carry discount codes, refunds and first-order source. A plain order export from any store also works.
• Klaviyo: A paid service that needs an account. It can send a post-purchase survey, which gives the customer's own words for each group. The skill works without it.
• Any good chat AI: Any current chat AI can rank a list and write notes. Attach your customer list.
• Handwrytten: A service that writes real notes with a pen machine, for when you pass the number you can write yourself. It is a paid service.
• Loom: Free plan with limits; paid plans cost more. Records a 60-second personal video you can send by link, at the cost of your time.

Functions and deliverables

• Repeat purchase improvement reviewer: Check where first-time buyers stop returning. Review the last fix and propose one change tied to the evidence.
  • A repeat health score out of 100, with the three parts and the sums
  • A verdict on last month's fix, either keep, drop or too early to tell
  • The one stage to fix, with the cause and the evidence
  • A plan for that fix, ready to approve, with its cost
  • The one number to watch next month
• Repeat order bottleneck finder: Share your order export. Get a repeat health score out of 100 and the one stage to fix first.
  • First-to-second order rate and 90, 180 and 365 day gross profit per customer, by cohort
  • The share of revenue and profit from your top 20 percent of customers
  • The single bottleneck stage, with evidence, and a ranked list of causes checked top down
  • A test list tied to the bottleneck, and a repeat health score out of 100
• New customer experience reviewer: Find where buyers stall before first use. Paste your returns, tickets and reviews to get ranked gaps, a fix for each and a score out of 100.
  • A first-use timeline from delivery to repeat use, with drop-off points marked
  • Ranked gaps, each with evidence quotes, the barrier type and one fix
  • A first-use score out of 100, with what it measures and the sum shown
  • Three measures to track in your own store, in place of "shipped" and "sold"
• Reorder and second-order planner: Plan reorder and second-order emails from how long each product lasts. You get refill timing, message goals and an offer cap for each message.
  • A flow plan with triggers, delays, a goal and one offer rule per message
  • A setup sequence for products that need teaching
  • Refill reminder timing per product, with a cap of two messages per cycle
  • A second-order nudge, a subscription prompt, and the timing for review and referral asks
• Delivery update flow builder: Share shipping data and order-status contacts. Get a date rule, four updates, a tracking-page plan and a reply for late parcels.
  • Your WISMO rate and delivery accuracy, each rated against a limit
  • A delivery date rule that pads your real times
  • A four-message update flow with timing and copy
  • A tracking page plan and a script for delayed parcels
• Customer profitability and ad source analyst: Share orders, refunds, costs and first order sources. Get top and bottom profit groups, offer comparisons and a score for new customer sources.
  • The top 20 percent and bottom 20 percent of customers by profit, with the traits of each
  • The angle and first-order offer behind each group, and repeat rate and 90-day profit for discounted versus full-price first orders
  • Angles to scale, angles and offers to stop, and page promises to fix
  • An acquisition fit score out of 100, with what it measures and the sum
• Top-customer outreach planner: Share customer spend, order history and team limits. Get personal touches for top customers and a reorder control group.
  • A ranked list of the top 50 to 100 customers to touch
  • A touch for each, with its cost
  • A note and video script for each tier, with no discount code in them
  • A control group split, a reorder check and a rule to expand or stop
• Brand profile builder: Share your website. Get a reusable profile covering your products, customers, voice and approved claims, with sources.
  • A folder of eight plain files, from a start file for your AI to a claims list and a change log
  • A tag on every fact, and every claim marked approved, pending or banned
  • A two-sentence diagnosis of your biggest gap, and a list of what's missing and who can answer it
  • A short slice for your next step, if you name one
• Paul Graham: Plan early-user work or sharpen an argument using Paul Graham’s recorded approach. Find a real problem, start narrowly and check the central claim.
  • Your problem worked on the way Paul Graham would: their questions, their method, their push-back
  • The answer or the work, in Paul Graham's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Clayton Christensen: Explore why customers buy using Clayton Christensen’s recorded approach. Start with a real situation, the struggle and what the buyer used before.
  • Your problem worked on the way Clayton Christensen would: their questions, their method, their push-back
  • The answer or the work, in Clayton Christensen's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step

Tasks you can use it for

• You sell things people could buy again and want to know where buyers drop off.
• You made a change last month and want an honest read on whether it worked.
• You are about to rebuild your emails and want to be sure that is the real problem.

Limits to check before starting

• You sell things people buy once, like a sofa: skip repeat tracking and look at how you win new buyers instead.
• You run a subscription business: cancels need their own check, so use your billing tool's cancel reasons or ask a person who knows churn.
• You have fewer than a few hundred orders or under six months of history: the numbers will be rough, so read them as directions only.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Start a chat with the agent. In Claude Code, add it as a skill or save it in .claude/agents, then ask it to run this month's check.
• Export your orders for at least six months and remove names and emails.
• Add your product list with how long one unit lasts, and any returns or ticket data.
• Read the plan and approve what you want. Make the changes yourself, or say yes to each one.
• Run it again next month. It compares with last month's files.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/monthly-repeat-buyer-keeper for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.