How it works

One growth system. Five capabilities. One shape they all share.

The whole system, explained without a pitch. What each capability does, the workflow they all share, where it sits in your week, and who answers for it when the output is wrong.

How we work

Four steps before anything gets built.

The system below is what a team ends up owning. This is how we work out which part of it you actually need, and in what order. Each step is tagged with what covers it.

01Free, no commitment

Start with what you are accountable for

Before anything is drawn or built, we get clear on the number you carry this quarter and what is standing in front of it. Most engagements that fail were scoped against a wish rather than an accountability.

Thirty minutes, then a short written summary
What we look at
  • The number on your name this quarter
  • What leadership already believes is wrong
  • Which constraints are real and which are habit
  • Who is free to actually do the work
What comes out

A single page your leadership already agrees with, naming the outcome and the two or three things blocking it.

02Agentic Partner

Map how the work actually moves

Not the process as documented, the process as run. We follow a real piece of work from the moment it enters to the moment a buyer sees it, and mark every handoff and every wait along the way.

One to two weeks, depending on how many people touch it
What we look at
  • How work enters, and who decides it is worth doing
  • Every handoff between people and systems
  • Where things sit waiting for a person
  • What is rebuilt each time instead of reused
What comes out

Your growth workflow drawn end to end, with the waiting time on it. Usually the first time anyone has seen the whole thing on one page.

03Agentic Partner

Measure where it stalls

One constraint sets the pace of everything downstream of it. Until it is named with evidence, effort spreads evenly across a system where only one part is actually binding, which is why a year of improvements can leave the number unchanged.

Falls out of the map, usually within days
What we look at
  • Time in stage, not time in total
  • The step everything else queues behind
  • What that queue costs per month
  • Whether the constraint is capacity, decision, or information
What comes out

The binding constraint, named, with the measurement beside it. It is frequently not the step anyone expected, which is the argument for measuring rather than asking.

04Team Training or Agentic Partner

Choose the smallest thing that unblocks it

Then, and only then, the question of what to build. Sometimes that is an agentic workflow. Sometimes it is ordinary automation. Sometimes the constraint is a decision nobody has made, or a person who is missing, and no software fixes either.

One session, and the recommendation follows in writing
What we look at
  • What an agentic capability would actually change
  • Whether plain automation is enough
  • Whether the blocker is a decision or a hire
  • What it costs to do nothing for another quarter
What comes out

A recommendation in writing with an owner and one number against it. When the honest answer is that AI is not the fix, that is what it says.

Nothing on this page gets built before step four, and step four is allowed to conclude that you do not need us.

The shape

Every capability is the same four boxes.

This is the teaching claim the whole system rests on. Learn the shape once and the fifth capability is an afternoon rather than a project. Hover any step to see what it looks like on the day.

  1. Input

    Approved truth

    The knowledge layer, plus your own systems. Never a blank context.

    On the dayOne priority account, its history, and the proof your team has already approved.

  2. Agent

    The volume work

    Inside boundaries you set. It drafts and scores, it does not decide.

    On the dayFive roles mapped, a narrative and a first touch drafted for each, in minutes.

  3. Gate

    A named human

    Reviews before it acts. Operator, Reviewer, Approver, named people.

    On the dayRed pen on the pack. What is right, what is wrong, what is missing.

  4. Output

    Work, and one number

    Something that ships, and the single measure it reports.

    On the dayThreads out per seat, and stakeholders engaged per account on the scorecard.

The human frames the work upstream and reviews it downstream. Nothing in between runs unsupervised. Governance did not disappear. It moved to the two ends.
The capabilities

Five lists, one operating model.

Open any of them. The workflow underneath is the same four boxes every time, which is the point.

00

The knowledge layer

Not a capability so much as the floor the other five stand on, which is why it is built first and built in the room. Positioning, ICP, proof, objections, regulatory context. Every capability reads it before it writes anything.

InputYour real material, as it exists today
AgentStructures it, flags what contradicts
GateYou approve what counts as truth
OutputOne governed source everything reads
The one numberGap log items closed per week

Built first. Nothing else works without it.

Capability 02 compounds hardest, because coverage is the lever a lean team can never buy with headcount. Capability 01 pays inside the current year, because it works demand that already exists. That is the build order.

The rhythm

Where it sits in the week.

A capability nobody has time for is a capability that stops. The loop goes in your own calendar before anyone leaves the room.

MondayPrioritize

What the numbers say to work on this week. The scorecard decides, not the loudest request.

WednesdayDraft and review

The system produces, you judge. This is where most of the volume work lands, and where the gate does its job.

FridayActivate and learn

What shipped, what landed, and what returns to the knowledge layer as evidence rather than opinion.

MonthlyGovern

Definitions reviewed, account tiers reviewed, and anything that stopped earning its place gets retired.

Governance

Three questions, three hats, on every capability.

Before anything runs: what does this produce and what does it not touch, who reviews the output before it acts, and what happens when it is wrong.

The Operator

Runs the capability day to day. Knows what it produces and what it deliberately does not touch.

The Reviewer

Checks the output before it acts. Not a rubber stamp: the person whose judgment the work actually needs.

The Approver

Releases it. Answers for it when it is wrong, which is the question that makes the other two roles real.

On a small team one person often wears all three. Name that explicitly rather than leaving it implied. Skip it and the result is not automation, it is expensive shadow work.

What it runs on

Three things, and you already have most of them.

No stack is sold here and none is mandated. Every team arrives on something different, and the architecture is the same either way.

Where your truth already lives

Your CRM and databases

Pipeline state, account history, the lead list, whatever already holds the record. Capabilities read from your systems and write back to them. Nothing gets migrated.

Where the work gets orchestrated

An agentic workflow builder

Whatever you use to put a model behind a schedule or a trigger, pull from your systems, and hold the result at a gate. Several do this well and they change every quarter, so we teach the shape rather than the buttons.

Who is accountable

A named human on the team

The part nobody can buy. Three hats on every capability, held by people with names rather than roles in a policy document.

The boundary

What this is not.

Not a platform.

There is nothing of ours to subscribe to. Everything built sits in your workspace and keeps working if we never speak again.

Not done for you.

Your team builds it. If nobody on your side will touch it, the system has no owner and it stops the week after handover.

Not a cure for every stalled deal.

Licensing, accreditation, procurement sequencing, and internal stalls move for nobody. Those get named, with owners, and the system answers for the rest.

When you want it on your accounts

The workshop builds the first one with your team.

Two sessions, four hours each, up to five seats, on your own accounts. Your team architects all five and builds the first one live.

Book the workshopTalk to Grace first