Hire an AI offer profitability test reviewer

Everything it can hand back

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

• Each offer's profit per session, with the sums, this quarter and last
• A keep, switch or no-call verdict, with what is missing if there is no call
• What is weak, from the sale, threshold, rival and checklist checks that ran
• One change to test, with the test design and the read-out rule
• A log that grows each quarter

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

• Calling an offer the winner because revenue went up.
• Changing the offer and the creative in the same test.
• Proposing several changes in one quarter.
• Using a deadline or limit that is not real.
• Generic copy that could fit any brand.
• Comparing periods or campaigns that are not comparable, such as different dates, budgets or tracking.

What a good result looks like

An example, not your task: You run more than one offer, or the same offer for months, and want to know what is earning.

• Offers table. Each offer with profit per session, margin and safety score, this quarter and last, with the sums shown.
• The call. Keep, switch to a named offer, or no call, with what is missing if there is no call.
• What is weak. Findings from the order size, sale, rival and checklist checks that ran.
• One change to test. The change, why, the test design, the read-out rule and how many weeks it needs.
• Sale check. The ROAS a planned sale needs, a verdict and any urgency lines that are true.
• Look again. The date to check back and the numbers to bring.

How the work is checked

• Every offer is compared on profit per session, with the sums shown.
• Cost lines include shipping, handling, fees, gifts and discounts, rounded up.
• No winner is called when the safety score is under 70.
• Only one change is proposed.
• 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 Alex Hormozi and Gary Bencivenga would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Read the log. Finds last quarter's lead offer, its numbers and the change that was planned.
• Set the menu. The first time, it picks the offer types worth testing for your margins and order mix.
• Compare on profit. Works out profit per session for each offer with full costs, and scores how safe the comparison is.
• Choose the path. If the safety score is low, no winner is called. If one offer clearly leads, it proposes a switch. Otherwise it holds.
• Check what moved. Runs the threshold, discount and rival checks only when order sizes shifted, a sale is planned or profit is slipping.
• Review a weak lead offer. Checks the offer against a tested list and takes the three changes most worth making.
• Pick one change and plan the test. Chooses one change, designs the test and writes the quarter into the log.

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 offer-refresh-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/offer-refresh-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

• Your live offers with price, tiers, gifts, discounts and guarantee
• Product cost, shipping, handling and payment fees
• Sessions and orders for each offer for a closed date window, and order sizes
• Reviews and rivals, and any sale or deadline you plan (optional)

Helpful context (optional)

• Your sale and campaign dates for the next quarter
• Reviews that mention price, value or rivals

Tools and access

• Claude Code: It can keep the offers folder and read the log each quarter.
• Shopify: Export orders and order sizes by offer, discount code and date range.
• Google Analytics: Export sessions by landing page for each offer's window.
• Shopify: Its reports export order values, units per order and repeat rate.
• Google Sheets: Good for a quick order-size histogram before you paste the numbers in.
• Any good chat AI: It builds the cost tables and sums from the numbers you paste.
• Google Sheets: A sheet keeps the cost lines in one place so each new test only needs sessions and orders.
• Shopify admin reports: Sessions, orders and conversion rate by landing page for each variant.
• Shopify: Its order export gives order totals and units per order, and its settings let you set free-shipping rates by order value.
• Google Sheets: Use a histogram chart to see the peaks before and after the skill reads them.
• Shopify: Its reports give you AOV, units per order and discount codes used, which are the inputs this check needs.
• Google Sheets: Keep your cost table and past sales in a sheet, then paste them in.
• Any good chat AI: Reviewing terms and ranking changes is text work any strong chat AI does well with this skill.
• Any good chat AI: This is text and number work that any strong chat AI does well with the skill.
• Any good chat AI: It plans the tests and builds the table from the numbers you paste.
• Klaviyo: Sign-up forms there can store a quiz answer as a profile property and split-test forms.
• Shopify admin reports: Sessions, orders and conversion by landing page feed the table.

Functions and deliverables

• Offer profitability test reviewer: Compare offers using profit per visit. Get cost gaps, weak terms and one proposed change to test.
  • Each offer's profit per session, with the sums, this quarter and last
  • A keep, switch or no-call verdict, with what is missing if there is no call
  • What is weak, from the sale, threshold, rival and checklist checks that ran
  • One change to test, with the test design and the read-out rule
  • A log that grows each quarter
• Store offer picker: Pick which store offer to test first. Give it your margin, order sizes and repeat rate. You get two or three offers with the maths for each.
  • The one margin trade your store most needs, with the reason
  • A shortlist of two or three offers, ranked by margin impact, each with its mechanic
  • The maths per offer: AOV, gross margin, break-even ROAS and target CAC
  • A test plan on separate pages, and what not to do for your store
• Offer profit-per-session checker: Share offer variants, full costs and traffic results. Compare contribution per session, break-even figures and a custom evidence score.
  • A table of net profit per order and per session for every variant
  • Break-even ROAS and CPA for each, and a flag for offers that raise AOV but lower profit per session
  • A winner and a recommended next test
  • A trust score out of 100 with the sum shown, and a list of checks it could not run
• Order threshold planner: Set a free shipping threshold from your real order sizes, not the average. Get offers just above each peak, plus a multipack ladder.
  • A read of your order spread: one peak or two, and where each sits
  • A threshold for each peak, set as a small stretch and not a jump
  • A multipack ladder with the default option marked
  • A choice of dollar off or percent off for each price range, and the profit change per order for each tier
• Discount margin and ad-cost checker: Model a discount using price, costs and order size. Get the ad-return target for a profit goal and a rule-based review.
  • The full-price numbers and the same numbers under the discount
  • A supported positive-CPA ad target, or a not-viable/zero-CPA result with its reason
  • A custom planning score only for valid complete inputs; points do not prove success
  • The discount depth with the most profit per order, when your history allows it
  • The same promo run as a percent off, a bundle or buy-one-get-one, and a gift, side by side
• Offer checklist reviewer: Check your offer, price and guarantee against a short checklist. Get a score out of 100 and the three offer changes to split-test first.
  • The offer in one line
  • A checklist review with a score out of 100 and how it was worked out
  • A revised offer
  • The three changes to test first, each with a test and a measure
• Offer price-comparison reviewer: Check if buyers see your offer as a commodity. Paste your offer and rivals, get a verdict, a price-war sum and the first row to change.
  • A side-by-side of you and each alternative, row by row
  • A count of differences a buyer can see quickly, and a verdict
  • The price-war sum for your numbers
  • The first row to change and what it would cost
• Offer limits and deadline planner: Share your offer, confirmed capacity and real dates. Plan clear limits, truthful deadline messages and a waiting-list approach for review.
  • One scarcity rule and one urgency rule that match how you work
  • The numbers and dates, with the sum behind each
  • The line for each channel, a sold-out message and a waiting-list message
  • An honesty check you can run before each launch
• Offer test planner: Plan offer tests for your online store. Give it your offer and weekly visits. Get two to four separate tests, a tag to track and a spend rule.
  • Two to four offer or buyer-type tests, each on its own page
  • The pop-up or survey question to ask and the tag to store the answer
  • A table to fill in per test, plus a rule for moving ad spend
  • A survey to send 90 to 120 days later, and a list of checks it missed
• 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
• Alex Hormozi: Improve an offer, price or lead plan using Alex Hormozi’s recorded frameworks. Check buyer needs, costs and honest promises before adding more leads.
  • Your problem worked on the way Alex Hormozi would: their questions, their method, their push-back
  • The answer or the work, in Alex Hormozi's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• Gary Bencivenga: Review a sales promise and its proof using Gary Bencivenga’s recorded approach. Match the problem, believable claim and clear offer before planning a test.
  • Your problem worked on the way Gary Bencivenga would: their questions, their method, their push-back
  • The answer or the work, in Gary Bencivenga'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 run more than one offer, or the same offer for months, and want to know what is earning.
• A sale is planned and you want to know if the maths hold before you announce it.
• Buyers compare you with cheaper rivals and you want a plan that is not a price cut.

Limits to check before starting

• You have one offer and few orders: run the offer checklist and one test by hand instead, as the sums need more sessions.
• You want a new offer idea: use the New offer opportunity finder, then bring the offer back here.
• You want the agent to change prices or codes for you: it will not. It writes a plan, and you set it up yourself.

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 quarter's offer check.
• Give it a closed date window for each offer, and say if offers ran at different times.
• Include every cost line. Costs left out make every offer look better than it is.
• Read the one change it proposes and the test plan. Set the test up yourself, or say yes to each step.
• Run it again next quarter. It reads its own log.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/offer-refresh-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.