How an engagement runs
No onboarding deck, read access and a written diagnosis.
Most agency onboarding is a month of meetings that produces a deck. This produces a document you can forward, and it produces it by week three. What follows is what actually happens, in the order it happens, including the parts that depend on you.
01
Week one
I get read access
Analytics, Search Console, your CRM, your ad accounts, and the repo or CMS if you will have me. Read access only, and I will sign whatever your security process asks for.
What happens in that week is boring and it matters: working out what your numbers actually count. Two systems disagreeing about how many signups you had last month is the most common first finding, and very little downstream is worth much until that is settled. I would rather look at the data than be told about it, partly because the gap between those two is usually where the problem is.
02
Weeks two and three
You get the diagnosis in writing
A document, not a call you have to take notes in. What is working, what is leaking, what to ignore, and what I could not determine from the data you currently have. That last section is deliberate: a diagnosis claiming certainty it does not have is worth less than one that names its own gaps.
You can forward it to a co-founder, to your board, or to anyone else on your team. It is yours to take away, pick apart and come back to me with questions on, so the picture is clear before we start implementing anything.
03
Weeks four and five
We agree a 90-day plan
Sequenced rather than listed, because half of this work depends on the other half. Positioning before pages, tracking before paid, with the reason written next to each one. Every item carries what it should move and roughly how much, framed as a target with its assumptions named, never as a promise.
You sign off before anything runs. If the plan and your product roadmap disagree, this is where that surfaces, and it is far cheaper to find out here than in month three.
04
Then
I build
Async by default. Written updates you read on your own time, a standing call at a regular cadence that works for both of us, and documentation written as the work happens rather than assembled at the end. That last part is the difference between an engagement and a dependency: what I build should still run after I stop.
Work in flight is visible. You see the pages, flows and campaigns that are being built, not just summarised in a report after the month is over.
- The first quarter
What 30, 60 and 90 days
actually look like.
Day 30
The diagnosis is delivered and the plan is signed. Tracking is fixed, because almost nothing else can be judged honestly until it is. The first piece of work is in flight, usually positioning or a page, because most of the rest depends on it.
Day 60
Two or three of the service lines are running. The first numbers are arriving and they are early enough to be directional rather than conclusive, which is worth saying out loud. This is the month where something gets cut. That is normal and it is planned for.
Day 90
A full quarter of data against the targets set in week five, and a written review of what moved, what did not, and what we got wrong. Then we diagnose, plan further and agree on the next phase of implementation built on top of the system that was created in the first 90 days.
- Working together
What it looks like week to week.
- Written first. Diagnoses, plans and updates are documents. Calls are for decisions, not status.
- A standing call at a pre-agreed cadence. Every two weeks, or monthly. We can change it based on our calendars and availability.
- One place for questions, in whichever tool you already use. I am not in it all day, and I will tell you when I am away.
- Documentation as I go, written into your systems so that everyone has access.
- From your side
What I need from you.
- Read access in week one, not week four.
- One person who can make a decision. Not a committee, and not something that needs three calendars aligned.
- Honest answers about what has already been tried and did not work. That is usually the most useful hour of the whole engagement.
- Your product roadmap, or at least the version of it you actually believe.
- What this is not
I am one person. That is the trade, and it is worth naming here rather than letting you discover it in month two.
I am not available daily, I am not on your Slack on demand, and I do not turn work around in 24 hours. What you get instead is someone senior on the account every week, rather than a strategist at the pitch and a coordinator for the eleven months afterwards. It also means I take a small number of clients at a time, and there are months when I am not taking any.
I also do not write production code for clients. I read it, and I write the specification your engineers need. Taking a ticket in your sprint is a different job, and hiring me for it would waste what I am actually for.
- Ready when you are
Thirty minutes, no decks. Bring the problem and your analytics, and you will leave with a sharper next step on whatever is actually stuck.