How it works
From first call to a running engine in 21 days
What happens in what order, who does what, where AI does the work and where a person checks it, who holds the keys, and how leaving works. Read it before the call, not after.
The lifecycle
Eleven steps, each with something in writing
Strategy call
Thirty minutes on your offer, your buyers and your markets. If outbound or content is the wrong tool for what you sell, this is where we say so.
No document — a conversation
Proposal and order form
The plan, a first draft of your ideal customer profile, the markets, the start date and the term. The agreement and a Data Processing Agreement come with it.
Proposal, order form, agreement, DPA
Kickoff and access
One onboarding form collects the people, accounts and approvals we need, so nothing is chased halfway through launch. Access is granted to accounts in your name.
Onboarding form, access record
The result definition, in writing
Before the first email goes out we agree what a billable meeting is: the ICP, the buyer roles, the channels it can come from, and the monthly cap on result fees. It is attached to your Order Form, so nothing is argued about after the fact.
Result definition (Order Form schedule)
21-day launch
ICP and messaging agreed. Sending domains registered to you, authenticated and warmed. First lists built and verified, first sequences drafted, and a content plan on Growth and Full-Funnel.
ICP & messaging document, launch checklist
Your approval, before anything sends
You approve the ICP, a sample of the list and every sequence before the first email goes out. Nothing goes out under your name that you have not seen, and any new angle later is approved the same way.
Signed-off ICP, list sample and sequences
Weekly cadence
Deliverability and replies watched daily, copy and targeting adjusted weekly. Interested replies reach you within one business day with the context attached.
Reply hand-offs; weekly report on Full-Funnel
Monthly report and review
What was sent, what was published, what came back and what changes next month — reported as it came in, against the result definition agreed at launch.
Monthly report
Monthly billing
The base fee is billed in advance. Results are billed in arrears at $200 per qualified meeting held, itemised meeting by meeting, never above your cap. No-shows are never billed, and you have 48 hours after a meeting to dispute it; we review it together against the definition.
Invoice with itemised meeting list
Quarterly review
On Full-Funnel, a strategy review of ICPs, channels and budget. On other plans the monthly review covers the same ground at a smaller scale.
Quarterly plan
Month to month, or a clean exit
After the initial term every plan runs month to month on 30 days' notice. Leaving is documented, never silent.
Offboarding record
The operating rhythm
One owner. Weekly, monthly, quarterly.
One named owner runs your account from kickoff to exit — the person on the strategy call is the person you hear from every week. No hand-off to someone you have not met.
Weekly
Lists, copy and deliverability
List quality, reply themes and inbox health reviewed; copy and targeting adjusted; anything new goes to you for approval first.
Monthly
Report and itemised invoice
What was sent, published and booked, the meetings held against the result definition, and an invoice that lists every billed meeting.
Quarterly
Strategy review
ICPs, markets and channels reviewed against pipeline in your CRM, and the next quarter's plan agreed in writing.
Who does what
A plan is done for you, not done without you. These are the parts only you can do.
We do
- Research prospects and build, enrich and verify the lists
- Set up, authenticate and warm the sending domains and inboxes
- Draft sequences, LinkedIn messages, articles, posts and ad creative
- Send, monitor deliverability daily and handle the reply inbox
- Build scoring, nurture and dashboards in your CRM
- Manage campaigns and budgets on Full-Funnel
- Report monthly, and change what is not working
You do
- Agree the ICP, then approve a list sample and every sequence before anything sends
- Agree the result definition and set your monthly cap on result fees
- Take the meetings — the conversation is yours to run
- Flag any meeting that did not meet the definition within 48 hours
- Give a short monthly interview or notes for content (Growth and up)
- Send the LinkedIn messages we draft (Growth and up)
- Approve articles before they publish
- Pay LinkedIn and Google directly for ad spend (Full-Funnel)
AI-native, person-checked
AI does the volume. A person makes the calls.
The engine is designed around AI from the start: research, enrichment and first drafts at a volume no one could do by hand. What is never delegated is judgement about who to contact, what is true, and what goes out under your name.
| AI does | A person does |
|---|---|
| Researches each account and summarises what it does | Decides who goes on the list, and who is excluded |
| Drafts first lines, follow-ups and LinkedIn messages | Checks every claim made about a prospect and approves copy before it sends |
| Sorts incoming replies by intent | Reads every reply and decides what goes to you |
| Drafts articles and posts from your material | Edits for accuracy and voice; you approve before publishing |
| Generates ad creative and copy variants | Chooses what runs and moves budget against pipeline |
The automation underneath — list building, enrichment, sending, CRM sync, reporting — is built and maintained by us on n8n, and every workflow is checked against 476 known node types before it runs. That is the same practice we sell as custom automation builds. The full policy — what AI does, what it never does, and where a person signs off — is on how we use AI.
Where your data sits in all of this
Two different things use AI here, and they are governed differently. The marketing work itself — researching prospects and personalising outreach — processes prospect data because that is the service: on your instructions, under a Data Processing Agreement, through named sub-processors, and never to train a model. The tooling we use to build the automation behind it is held to the four commitments below, which are the same ones written into the privacy policy.
- Your data does not go into our AI tooling
- Machine assistance operates on workflow structure, code and documentation. Your live records, your customers' personal data and the contents of your systems are not fed into it. Automations are built and tested against sample or synthetic data until the logic is right.
- Nothing you give us trains a model
- Client data is never used to train or fine-tune AI models, by us or through us. This has been our position since before we described how we build, and describing it has not changed it.
- A named engineer is accountable for the whole build
- However a line was drafted, a person reviewed it, tested it and put their name to the handover. AI-assisted is not AI-decided, and there is no point in the process where nobody is answerable for what shipped.
- The same rules apply to anyone we bring in
- Contract engineers work to the same encoded standards and the same data boundary, under written agreements that include IP assignment. The standard does not loosen because capacity was added.
The full position is in the privacy policy.
What this does not buy you
- It does not make a weak offer sell. If the market does not want what you sell at your price, better outbound only finds that out faster — which is useful, but it is not pipeline.
- It does not replace your sales conversation. We create and hand over the conversations; closing them is yours.
- Content and answer-engine visibility compound over months. The first quarter is mostly foundation, which is why it sits alongside outbound rather than instead of it.
Security
Access and accounts
Running your marketing means working inside your CRM, your calendar and your ad accounts. How a provider handles that access deserves more scrutiny than its tool list.
Accounts in your name
Your CRM, ad accounts and sending domains are registered to you and billed to you. We work inside them; we never hold them.
Credentials
Held only in a platform's own credential store or an encrypted password manager. Never in plain text, chat or email. Anything sent to us insecurely is treated as compromised and rotated.
Least privilege
The narrowest role that lets the work happen. We do not ask for an admin seat to run a campaign.
Recorded and revoked
Every item of access is logged with the date granted, and revoked against a written checklist when a plan ends.
If you leave
The easiest way to judge a monthly service is to ask what you keep when it ends. With us, everything we built runs in accounts you already own, so leaving is a handover rather than a rebuild.
Our terms- Give 30 days' notice at any point after the initial term
- You keep the sending domains, every list, sequence, article and dashboard
- You get the suppression list — the people who asked not to be contacted — so you stay compliant after we leave
- Our access is revoked against the access record, and you get written confirmation
- Prospect data we processed for you is returned or deleted under the DPA
Due diligence, answered
How much of our time does this take?
Most of it lands in the launch: agreeing the ICP and messaging, granting access, and approving the first sequences. After that it is taking the meetings, a monthly review, and on Growth and above a short monthly content interview, approving articles and sending the LinkedIn messages we draft.
How are results counted and billed?
Against the result definition agreed in writing before launch: the right ICP, a role on the agreed buyer list, booked through a channel we run, and the meeting actually took place. Each one is billed at $200 per qualified meeting held, monthly in arrears, itemised so you can check every line, and never above the cap you set. You have 48 hours after a meeting to dispute it.
What exactly is AI doing, and what is a person doing?
AI does research, enrichment, reply sorting and first drafts — the high-volume work. A person decides who is contacted, checks every claim about a prospect, approves copy before it sends, reads every reply, edits every article, and makes every budget decision.
Does prospect data go through AI models?
Yes, when researching prospects and personalising outreach is the service you have asked for — that is the work. It happens on your instructions, under a Data Processing Agreement, through sub-processors named in it, for your campaigns only, and never to train a model. Separately, the tooling we use to build the automation behind the engine works on structure and code, not your records.
Can we see the agreements before committing?
Yes. The agreement and the DPA are sent with your proposal, so your lawyer can read them on your timetable rather than under a deadline.
What if we want to leave?
After the initial term, 30 days' notice. You keep the domains, content, lists and dashboards, you get the suppression list, and our access is revoked against a written record. Nothing is held hostage.
Do you carry professional indemnity insurance?
Not currently. We would rather tell you that than imply otherwise. Liability is capped in the agreement you sign. If your procurement process requires insurance, say so on the first call so we can tell you honestly whether we can meet it.
Start with a strategy call
Thirty minutes on whether your market is a fit. Repomanta Information Technology Services contracts in its registered name, and Repomanta Digital Lab is the trade name on the work.