Skip to the contentMapped Margin

AI implementation

No business is too small or too big for us to make an impact on.

We set up AI agents inside your own Claude or ChatGPT account. Each one drafts the work in the tools you already run and waits for your approval.

30 minutesNothing sends until you say so

You approve every send

Sample: a new enquiry by email, after hours

Front desk · waiting for you

Hello, and thank you for getting in touch. We have an opening on Thursday morning or Friday afternoon. Which suits you? If neither works, send a time and we will confirm it.

Ready for your review

A sample. Nothing here was sent.

Set up inside the tools you already run

What we do

One team for AI agents, from one job to a company-wide rollout.

01

AI agents set up in your account

Front desk, Get paid and Lead response, set up inside your own Claude or ChatGPT account.

  • Tested on twenty of your own past cases
  • Your team trained in one session
  • Checked every month
See how it works
02

Audit and build

We map every workflow, forecast what an agent returns, and build the ones you pick.

03

Enterprise pilots

One seat, your own hardware, a controlled rollout.

See pricing

We also build lead generation agents: each one finds prospects that fit, drafts the first message, and holds every send for your approval.

  • Diagram: the Front desk agent sits between your mail, calendar, files and contacts.

    Front desk

    Inbox, calendar, files and CRM as one job.

  • Diagram: the Get paid agent sits between your invoices, reminders, payment links and cash report.

    Get paid

    Invoices, reminders, payment links and your cash report, moving as one flow.

  • Diagram: the Lead response agent sits between your leads, search, calendar and CRM pipeline.

    Lead response

    Inbound leads from your website, phone and email, answered fast and followed up.

How it works

Four steps to AI that works.

  1. Thirty minutes on what takes your time.

  2. One job, inside your own Claude or ChatGPT account.

  3. Nothing sends until you say so.

  4. We review results and tune it.

Book a call1/4
Book a callThirty minutes on Google Meet, pick any weekday slot.
30 minutesNothing sends until you say so
Book a call

Or write to hello@mappedmargin.com. We reply within one business day.

Front desk

Nothing sends until you say so.

A sample. Nothing here was sent.

Front desk · waiting for youSample: a new enquiry by email, after hours

Hello, do you have any openings this week? Mornings work best for me.

  • Read the enquiry in Gmail
  • Checked Google Calendar
  • Drafted the reply, waiting for you

Hello, and thank you for getting in touch. We have an opening on Thursday morning or Friday afternoon. Which suits you? If neither works, send a time and we will confirm it.

Ready for your review
Front desk agent screen with a drafted reply awaiting Approve, Edit or Hold, a calendar slot being booked and a CRM record being written. Sample data.

Illustrative

Sample Blueprint, by the numbers

Illustrative totals from the sample Blueprint: eight workflows at a company of 500 people, before build cost and the cost to run the agents.

Today 4.9 days1.6daysWith agentsAverage days, trigger to done
1,011hours a month returned
51,800of 80,800 a month in labor and error cost comes out

The four numbers we move

Margin up

We measure each agent against its forecast labor saving at 30 and 90 days.

Errors down

An agent runs the same check on the last invoice of the day as on the first.

More from your headcount

Each row shows its headcount effect in FTE, up or down, before you approve it.

Work faster

An agent starts each task when it arrives, so the days it sat in an inbox come out of the cycle.

Sample screens

Dashboard demo, measurements screen: measured and forecast run rate, and a chart of the hours each live agent returned at 30 and 90 days against its Blueprint forecast.Dashboard demo, agents screen: four sample agents, each with its status, owner, cases and exceptions today, a threshold slider, and a Pause or Resume button. The paused agent shows who paused it and the runbook your team follows.

Your account stays yours.

Four facts about where the agent lives and what happens if you stop.

  • 01

    In your own account

    The agent lives inside your own Claude or ChatGPT account and works in the tools you already run.

  • 02

    You approve every send

    Nothing sends until you say so.

  • 03

    Checked every month

    Each month we rerun your saved cases and tune the rules.

  • 04

    Yours if you stop

    If you stop working with us, you keep the copy you installed.

A drafted reply waiting for review, with Approve, Edit and Hold, and the line "Nothing sends until you approve it."

All sizes

Priced per workflow.

You see the number before we build.

01

Solo and small teams

Set up in your own Claude or ChatGPT account.

Quoted after the call

  • Mapping session
  • Monthly review quoted
  • Tested on twenty of your own past cases
  • Your team trained in one session
  • Checked every month
02

Mid-size companies

Mapped, forecast and built workflow by workflow.

Fixed fee per workflow

  • Audit quoted after the call
  • Blueprint included
  • Payback shown on every row first
  • Measured at 30 and 90 days
  • Agents go live one at a time
03

Enterprise pilots

One seat, your own hardware, a controlled rollout.

Quoted per workflow

  • One seat
  • Scope agreed in writing
  • Read-only before it can write
  • A named owner who can pause it
  • Your own hardware or cloud

Each workflow is quoted after the call.

FAQ

Questions owners ask first.

All questions
  • You decide that for each action, and the Blueprint row records it. On invoice matching the agent posts a clean match to the approval queue. Your controller still approves the batch. Payments, customer offers above your limit, and changes to pay records stay with a named approver. We do not build agents that approve spend, change bank details, release a payment, or sign a contract.

  • An agent will get some cases wrong. Each agent has a confidence threshold you set, and a case below it goes to a named person with the reason attached. Before go-live the agent runs in parallel with your team. We measure its error rate against theirs and print it on the Blueprint row. The log shows what the agent read and what it wrote, and the agent's owner can pause it. We measure again at 30 and 90 days. If a result lands below the forecast, we tell you the week we find it.

  • Yes, inside your own Claude. The agent reads your customers, orders and invoices through Oracle's own NetSuite connector, drafts the reminder or the record change, and waits for your approval before anything is written. Oracle's connector needs a Claude Pro plan or higher.

  • Yes, through our small-business team. The full Audit and Blueprint are built for companies of 200 people and up, where the volume carries the forecast. Under about 50 people a separate team runs a shorter engagement sized to your headcount, with the same agents and the same controls. Book the same discovery call and say your size, and that team takes the call. The starter path sets out the steps, from the mapping session to your first agent.

  • You tell us, and within 24 hours you have the files and complete control: the agent code and prompts, the runbooks, the logs, and the accounts each agent signs in with, transferred to a person you name. The agents keep running while the handoff happens. Nothing we build depends on a platform only we can operate.

  • We report the headcount effect the numbers show, up or down, on every row of the Blueprint. An agent takes the routine cases in a workflow, and the people who run it today keep the exceptions, the approvals and the vendor calls. The Audit also lists the work that has sat in a queue for months because your team had no hours for it, so you can see where the returned hours could go before you change a role. We will not write a document to justify a decision you have already made.

Let us set up your first agent.

Start with one workflow in your own account.

30 minutes · Nothing sends until you say so. We reply within one business day.

  • Your own account
  • You approve every send
  • Checked every month
  • Yours if you stop