Hire an AI subscription revenue, churn and customer cost review

Everything it can hand back

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

• Recurring revenue, churn and net new customers under fixed definitions
• The last eight weeks in a row, with the target line
• The one thing to fix first, and whether it repeated
• Payback months, the lifetime profit to cost ratio and a scale-or-fix call, monthly
• Commit, best case and pipe-only forecasts, if you have a pipeline
• One growth target for the week and what to skip

How it does the job well

• Agree the question and the decision it supports before analysing.
• Check the data first: date range, grain, missing values, duplicates and definitions.
• Compare like with like: same period length, definitions and segments.
• Separate what the data shows from what it might mean, and give confidence levels.

Mistakes it avoids

• Changing the definition of churn to make a week look better.
• Reading a yearly plan as a month of revenue.
• Chasing the growth target with a discount that breaks payback.
• Reporting lifetime value from two months of data as if it were known.
• Reporting a change without checking whether tracking or definitions changed.
• Treating correlation as cause.

What a good result looks like

An example, not your task: You run a subscription business and read the numbers differently each week.

• This week. Recurring revenue, growth rate against the target, new and churned customers, and net change.
• The trend. The last eight weeks in rows, with the target line.
• Fix first. The one thing to fix, and whether it is new or repeated from last week.
• Economics. Monthly payback in months, lifetime profit to cost ratio, scores and a scale-or-fix call.
• Pipeline. Commit, best case and pipe-only forecasts with the maths shown, if you gave a pipeline.
• The week's target. One growth target, what it means in customers or revenue, and what to skip.

How the work is checked

• Every number follows definitions.md, and the definitions did not change.
• The growth rate is worked out from the saved rows, not estimated.
• The one thing to fix is named, and a repeat is called a repeat.
• The payback and ratio use the gross margin given, or are marked as guesses.
• 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 would. This is a way of reviewing the work. It does not mean they wrote or endorse it.

Capabilities

• Fix the definitions once. Writes what counts as recurring revenue, an active customer and churn, approved by you and used every week.
• Read the source. Pulls recurring revenue, active customers and churn from billing, adds the week's row and works out the growth rate.
• Read the numbers together. Revenue, churn, cost to win a customer and lifetime value read as one picture, with the one thing to fix first.
• Check the economics monthly. Payback months and lifetime profit against cost, scored, with a scale-or-fix call.
• Forecast the pipeline. Commit, best case and pipe-only forecasts with the maths shown, if you sell with a pipeline.
• Set the week's target. One growth target, what it means in sign-ups or revenue, and what to skip.
• Log and hand back. Saves the row and 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 weekly-saas-numbers-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/weekly-saas-numbers-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

• Last week's billing export with plan, price, start, cancel and payments per customer, or a read-only billing connection
• Ad spend, sales pay and tools for the week (advised)
• Your gross margin per customer (advised)
• Your open pipeline, if you sell with a sales team (optional)
• Last week's keeper file, once it exists

Helpful context (optional)

• Ad spend by channel for the week
• Your gross margin, checked by your accountant

Tools and access

• Claude Code: It can keep the weeks and definitions files between runs. You start each run by hand every Monday.
• Your billing tool: Export per-customer plan, price, start, cancel and payments. A read-only connection lets the tracker read them directly.
• Any good chat AI: It needs only your numbers and a clear table. Check one month by hand first.
• Shopify: Its cohort and customer reports give you first order and repeat order counts by month.
• Shopify cohort report: Gives revenue by first-purchase month at day 30, 90, 180 and 365.
• Your ad platforms' spend export: Total spend per month for CAC.
• A spreadsheet: Lets you check the sum by hand.
• Any good chat AI: Short sums on a table and a list of tasks are easy for any strong chat AI.
• A spreadsheet: Keeps the weekly figures and shows the trend as you add each week.

Functions and deliverables

• Subscription revenue, churn and customer cost review: Review recurring revenue and customer losses under fixed definitions. Get cost gaps, payback checks and one proposed weekly priority.
  • Recurring revenue, churn and net new customers under fixed definitions
  • The last eight weeks in a row, with the target line
  • The one thing to fix first, and whether it repeated
  • Payback months, the lifetime profit to cost ratio and a scale-or-fix call, monthly
  • Commit, best case and pipe-only forecasts, if you have a pipeline
  • One growth target for the week and what to skip
• Subscription revenue tracker: Work out your subscription revenue: monthly and yearly recurring revenue, active subscribers and churn, read from your Shopify store's contracts.
  • Monthly recurring revenue
  • Annual recurring revenue
  • Active subscriber count
  • Rolling churn rate (a rough proxy unless you give it last period's CSV)
  • A short session log
  Source: 40rty AI.
• SaaS metrics coach: Check your SaaS metrics: share your revenue and customer numbers and get ARR, churn, CAC and LTV rated healthy, watch or critical, with fixes.
  • A table of your key metrics
  • A status label for each metric
  • A short summary of your overall position
  • Priority actions for this month
  Source: Alireza Rezvani.
• CAC payback checker: Check your CAC payback: give it ad spend, margins and cohort profit. Get profit to CAC ratios, a scale, hold or fix call and a score out of 100.
  • Contribution per order and the first order ratio
  • The profit to CAC ratio on each window, with its band
  • A target, ceiling and floor CAC, and payback in days or months
  • Whether each new customer is more or less profitable than the last
  • A scale, hold or fix call and a score out of 100 for payback safety
• Customer acquisition profitability reviewer: Share ad spend, new customers and order profits. Get customer payback ratios, warning signs and a score out of 100.
  • CAC and LTGP by month, with the time window used and why
  • The LTGP to CAC ratio and its band: scale, hold or fix
  • Which side is moving, CAC or LTGP
  • A list of changes to timestamp, and a score out of 100 with its sum
• Bookings forecast builder: Build a bookings forecast with commit, best case and pipe-only numbers. Give it your pipeline and cohorts. The maths and assumptions are shown.
  • Commit, best case and pipe-only numbers
  • An assumptions block
  • Retention by cohort
  • A reliability rating for each funnel stage
  Source: Alireza Rezvani.
• Weekly growth target setter: Paste your weekly revenue or user numbers. Get one weekly growth target, a yearly view and which tasks to skip. Sets a clear weekly growth target.
  • The one measure it picked and why
  • Your recent weekly growth rate, worked out in the open
  • A target rate, with what it adds up to in a year
  • Each task marked do, defer or skip, with a reason
• Metric definition writer: Share a metric, calculation and business use. Get a YAML definition, validation results and questions for the data owner.
  • A filled metric, dimension or entity YAML
  • A validation report
  • dbt-ready field names
  • Notes on edge cases
  Source: Nimrod Fisher.
• 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

Tasks you can use it for

• You run a subscription business and read the numbers differently each week.
• You spend to win customers and are not sure whether each one pays back.
• You want one number to judge the week by, and permission to skip the rest.

Limits to check before starting

• You sell one-off products with no repeat billing: use a plain sales report from your shop or accounting tool.
• You want a deep read on why customers cancel: use the Subscriber cancellation and retention reviewer, and bring your cancel list with dates and plans.
• You have no billing export and no customer list yet: start by writing down each customer's plan, price and start date in a spreadsheet.

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 week.
• Export last week from your billing tool, or connect it read-only. Put the file in one folder, or attach it in the chat.
• On the first run, approve the definitions of revenue, customer and churn. They are saved and never drift.
• Add your spend and gross margin once. Update them when they change.
• Read the target and the one thing to fix. Run it again next Monday. Without file access, copy the weeks and definitions text it gives you and paste it in next time.

Read it as data

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