Hire an AI ad testing manager

Everything it can hand back

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

• A test budget, a spend limit per ad and a read date for the round
• Five untried ideas, each with designer-ready versions
• A campaign layout with one idea per ad set
• A keep, leave or replace call for each set after each round, and what it means
• A running list of lessons, with dates
• A call on whether the problem is the ads or the page after three rounds

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 a winner from the biggest spender alone.
• Changing several things between rounds, so no lesson can be read.
• Reading a round before the read date at low volume.
• Running past the cap because the next idea feels promising.
• 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 want to find a winning ad and expect it to take more than one round.

• Round plan. The daily budget, spend limit per ad, read date, five untried ideas with briefs, and the campaign layout.
• Round result. A verdict for each ad set against your target, any false winners, and whether the read was safe to act on.
• What it means. The path taken and the number that decided it.
• Next round. A plan built on the lesson, or the call to stop and why.
• Lessons so far. A dated list of what each round changed and what it showed.
• Where the sprint stands. Rounds used out of three, ads tested, hit rate if readable, and every missing item.

How the work is checked

• Every verdict is measured against the target, and the sums are shown.
• No ad was judged on a few hundred impressions.
• No idea that lost on a clean read is planned again.
• Each round changes one thing, and the lesson says which.
• 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 Eugene Schwartz and John Caples would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Set the budget. Works out the daily test budget, the spend limit per ad and the earliest day to read, from your target.
• Find ideas. Reads your past ads and your brand file and finds five ideas you have not tried, then writes designer-ready versions.
• Lay out the round. Sets the ad packs, minimum spend and rules, with one idea per ad set so the result can be read.
• Check the round can be read. If the round was too small or ended early, it gives no verdict and sets a later read date.
• Read the round. Gives a keep, leave or replace call for each set, and flags false winners.
• Choose the path. A winning set is left alone and built on. A round with an unsafe read is repeated. A lost round is checked for sameness, then gets new ideas or a recorded lesson.
• Count the hits. After 20 ads, works out your hit rate and how many ads the next batch needs.
• Apply the cap. After three rounds with no winner, checks whether the ad, the page or the checkout is the cause.

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 ad-test-sprint-runner 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/ad-test-sprint-runner/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 brand file with products, claims, proof and buyers
• Your target cost per sale or ROAS
• Past ad results for the last 30 to 90 days
• What you can spend on the whole sprint and per day, and the platform
• Each round's results by ad, when the round ends

Helpful context (optional)

• Your break-even figure
• A list of every ad you have run in the last 90 days

Tools and access

• Claude Code: It can keep the sprint folder and remind you of the read date.
• Meta Ads Manager: Export ad-level results by ad set, and show the incremental column if your account has it.
• TikTok Ads Manager: Export ad-level results for the round if the sprint runs there.
• Any good chat AI: The sums are simple, and any current chat AI can run them and explain each step.
• Meta Ads Manager: Your spend, sales and cost per sale for each ad set come from here.
• Shopify: Your real order value and orders a week come from store reports.
• Meta Ads Manager: An export by ad gives the 'measured' grade, with spend and results for each ad over the dates you pick.
• Meta Ad Library: Shows the ads your page runs now, with start dates, when you have no export to work from.
• Google Sheets: Keeps one row per past ad, so each lens and the moments grid are filters you can check.
• Meta Ads Manager: An export by ad for the last 1 to 4 weeks shows spend, purchases and frequency, which pick explore, extend or refresh.
• Meta Ad Library: Shows what runs now, as 'observed' evidence when you have no export.
• Google Sheets: Tracks cost per sale and frequency week by week, so you can see when a winner starts wearing out.
• Meta Ads Manager: It holds the ad set spending limits, the exclusions and the campaign settings the plan is built around.
• Any good chat AI: Any current chat AI can build the plan from your numbers. Check each setting against the live menu.
• Meta Ads Manager: Its export gives you spend, impressions, CTR and link CTR by ad. The Columns menu has the incremental attribution view.
• Any good chat AI: Any current chat AI can read an export and follow the metric order. Attach the file instead of retyping numbers.
• TikTok Ads Manager: Export the ad-level report with CTR, CPM, CPC and cost per result for the days you want.
• Google Sheets: Keep one row per ad and per day to see how the lines move.
• Any good chat AI: Any current chat AI can group a pasted table of ads.
• Meta Ads Manager: Source of ad-level frequency over 30 to 60 days for cold campaigns.
• Meta Ads Manager: Its Columns menu has the compare attribution settings and incremental view the skill reads, and it exports by ad.
• Any good chat AI: Any current chat AI can count hits and rank concepts from a pasted export.
• Meta Ads Manager: Where link cost per click, click rate and cost per 1,000 views come from.
• Shopify analytics: Where add to cart, checkout and order value come from.
• Any good chat AI: Any current chat AI can read the numbers you paste.

Functions and deliverables

• Ad testing manager: Plan ad test rounds, read real results and use each lesson for the next plan, stopping after three rounds without a winner.
  • A test budget, a spend limit per ad and a read date for the round
  • Five untried ideas, each with designer-ready versions
  • A campaign layout with one idea per ad set
  • A keep, leave or replace call for each set after each round, and what it means
  • A running list of lessons, with dates
  • A call on whether the problem is the ads or the page after three rounds
• Ad test budget calculator: Share your target cost per sale and test days. Get a daily budget, spend limits and a results-review window.
  • A total test budget and a daily budget, with the sum shown
  • A keep, wait or cut rule at each spend checkpoint
  • A read window and notes on day-of-week swings
  • What to do with a winner and with a loser
  • For high order values, a plan using cheaper signals and a rule for which ads move to paid
• Untried ad idea finder: Find untried ad ideas for your brand. Paste your past ads and brand file. You get a gap map and five ad ideas, each one a clear test.
  • A record of every past ad, with each result graded measured, reported or seen running
  • Five lenses and a list of your proof, showing what's covered, thin or open
  • The five biggest gaps, ranked, plus the proof worth collecting
  • Five ad ideas, each with its test question, a five-link check, headlines and an opening
  • Any check the map still missed, one line each
• Ad test brief writer: Share brand facts and any ad results. Get three ideas, nine distinct designer briefs and questions to fill key gaps.
  • A header with the evidence grade, the plan chosen and why
  • Three ad ideas, each aimed at a different buyer doubt, with a filled-in thinking block
  • Nine versions with layout, headline, picture and what each one changes
  • A footer listing guesses and the questions to ask the founder
  • Any check the brief still misses, one line each
• Meta test campaign planner: Share ad concepts, target cost per sale and budget. Get test groups, naming rules, settings checks and bounded spend plans.
  • A campaign and ad set layout sized to your store
  • A pack plan with ad names, minimum spend and removal date
  • A CBO or ABO call with the trade-off written out
  • A settings checklist, naming guide and rules for adding ads and scaling
• Ad creative test analyst: Paste your ad test results by ad set. Get keep, leave or replace calls for each ad, false winners flagged and a score for how safe the read is.
  • A verdict for each ad set against your target
  • A table of every ad with keep, leave, replace or retest and the reason
  • False winners flagged, with the path from view to purchase for each ad
  • The next creative brief, and a read score out of 100 with its sum shown
• TikTok ad test reader: Paste your TikTok ad test numbers. Get a keep, fix or kill call for each ad, the likely cause, and a score out of 100 for each ad and the test.
  • A call for each ad: keep, fix hook, fix audience, kill or too early
  • The likely cause of each problem
  • A score out of 100 for each ad and for the test
  • A plan for the next test
• Ad sameness checker: Share your ads with buyer, message, offer and format. Get idea groups, a repeat score and specific changes for repeated ideas.
  • Clusters of ads that are really one concept
  • A sameness score out of 100, with the sum shown
  • Which ads to replace, and what to change in each (persona, angle or offer)
  • A list of cold ads with a frequency above 2
• Ad testing success rate planner: Share an ad export. Calculate how often new concepts meet your profit target and identify ideas to develop.
  • Your creative hit rate, with the sum shown
  • A concept list ranked by spend, with keep, leave alone, iterate or drop
  • An iteration brief for each winning concept
  • A volume plan for the next batch
• Product test results analyst: Read your product test results and find out if the ad, the page or the checkout is the problem. Get a score out of 100 and the next fix.
  • A call: ad issue, page issue or checkout issue
  • A score out of 100, with what it measured
  • The next fix, with the number that triggered it
  • A note when the test ran too briefly or has too little data
• 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
• Eugene Schwartz: Match a headline or sales page to what buyers already want and know. Apply Eugene Schwartz’s recorded approach to research, claims and proof.
  • Your problem worked on the way Eugene Schwartz would: their questions, their method, their push-back
  • The answer or the work, in Eugene Schwartz's voice, using their ideas by name
  • Anything that goes beyond their public record, marked
  • One suggested next step
• John Caples: Write headlines and plan a fair ad test through John Caples’s recorded approach. Lead with the reader’s gain and count replies by version.
  • Your problem worked on the way John Caples would: their questions, their method, their push-back
  • The answer or the work, in John Caples'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 want to find a winning ad and expect it to take more than one round.
• Your last test did not win and you do not know what to try next.
• You have run tests before but never kept what they taught.

Limits to check before starting

• Your store makes only a few sales a day. Tests cannot reach a readable result, so fix your page and offer first.
• You want one test, not a run of rounds. Plan a single round by hand with your ad platform's own test tools.
• You need a weak product or page fixed. More ad tests will not help, so work on the product or page first, or ask a person who knows your market.

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 start a sprint.
• Give it your brand file, target, past results and budget. It hands back the round plan.
• Launch the round yourself, or say yes to each step if it is connected.
• On the read date, give it the round's results. It reads them and plans the next round.
• Keep going until a set wins or the cap of three rounds is reached.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/ad-test-sprint-runner 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.