What we've done, how well it worked, and what it taught us. Honestly, with the numbers in front of us.
Five people leaving the room with the same picture of the business — and the same answer to what we're not doing.
An 18-month plan we can actually execute — with the numbers behind it, not just the intent.
No scheduled breaks. We're all grown-ups.
Take five between sessions whenever you need it — grab a drink, step outside, take a call. We won't be timetabling it. Lunch is on the premises at 12:15. The one ask: when a session is running, phones down and be in the room.
One thing each about yourself that nobody in this room knows. You're allowed to stretch it. We'll guess whether you did.
A few months ago we agreed we wouldn't formalise the company until we'd proved there was something here worth formalising. That test has been met — so now we form the company properly and set the shareholdings to match.
Company 15338024 carries on — we just rename it, so the legal entity matches what we trade under. Still not VAT registered.
A proposal to adjust shareholdings in response to contributions.
Right now this is paperwork. With revenue, a valuation and a client book behind it, the same change gets slow and expensive.
The existing agreement gets nullified and replaced.
Adding a shareholder means the current shareholders' agreement no longer reflects the company. If everyone is happy with what we agree today, Scott will send round a new one for signing.
Scott reduces his shareholding to 47.5%
SibbertCo — Nigel and Libby — increases to 17.5%
The employee pot reduces to zero and is redistributed
We have made £7,886 of profit and hold £4,006 of cash.
The difference is the £4,000 Aura invoice raised on 4 September and not yet paid. Profitable on paper, and almost all of the year's profit is still sitting in someone else's bank account. At this size, getting paid matters as much as getting the work.
Source: Xero, accrual basis, as at 8 September 2026. No debt, no payables outstanding.
Worth asking properly while everything else is being settled. The legal entity is MIXE Group Ltd — "Lean Growth Labs" is a trading name, so changing it touches no shares, no filings, and no structure. It will never be cheaper to change than today.
The deciding question isn't whether we like it. It's whether it's the name we'd want on the ad that sells a £1,000 audit to someone who's never heard of us.
Source: Xero, accrual basis, as at 8 September 2026. The £2,000 shows against Will in the books but is Kingfisher's work. £4,000 of the Aura total was invoiced on 4 September and is still outstanding.
Phase 2, the customer quoting tool: built, finished, invoiced at £4,000. Phase 1, end-to-end job visibility across their business: £10,000 quoted, £5,000 received, now midway through its final stage.
The only engagement where we have meaningfully delivered.
Both our original product: lead generation for a monthly fee, with the client funding ad spend. Kingfisher paid one £2,000 handling fee and stalled. Optio is on profit share and has paid nothing.
We couldn't generate the leads we'd promised. Optio has now moved to Meta — three weeks in and running.
An AI version of Rob, the CEO, to coach when he isn't available. Quoted and agreed, then paused for his health. Realistically 2027.
Warm, not lost — but not revenue this year.
Five engagements. One has produced delivered value for a client. The rest are unconverted, unpaid or unstarted.
Optio was lead generation on a profit share. We built the micro-site, the sector landing pages, the voice qualification and the ad campaigns. Conversions to date: none. Revenue to us: none. Kingfisher, the same kind of work for a small fixed fee, has produced no conversions either.
We got paid only if someone else's sales process worked.
Not on whether we delivered well. Not on whether the campaigns performed. On conversions, in a market we didn't choose, through a channel we didn't rate, closed by people we don't manage. We funded the build and handed over the upside.
This was the original proposition of the business, and it was a good one: £2,000 a month, and we guarantee ten leads in your diary. Monthly, recurring, same shape for every client, priced clearly. Exactly the kind of product we spent this morning saying we need.
We guaranteed an outcome we couldn't reliably produce.
Niche sectors with no obvious audience, reached through channels we didn't rate — LinkedIn chief among them. The shape of the product was right. The promise inside it depended on things outside our control, so when the leads didn't come there was nothing to bill and nothing to show. Kingfisher stalled. Optio has only now moved to Meta, three weeks in, still unconverted.
We found a repeatable product. The guarantee inside it was the wrong one.
Same £2,000-a-month instinct. Same recurring shape. A promise we can actually keep.
We'd happily share risk where we control the outcome, or have real confidence in it — that's often the better deal for everyone. Three questions decide whether that's true. Optio failed all three.
If conversion depends on their sales team, their pricing and their follow-up, we're not sharing risk — we're absorbing it.
Optio meant sourcing niche B2B leads on LinkedIn — a channel we already knew wasn't effective. We took the deal anyway.
An incredibly niche sector with no obvious audience is a different proposition from B2C heating on Facebook, in demand, going into winter.
Profit share stays on the table. Fully deferred doesn't.
Where we control the outcome or have real confidence in it, sharing the upside is a good deal for both sides and we should be willing to do it — the returns are better than a day rate. What changes is the balance: an element up front to offset our risk, with the share on top. The only thing we rule out is carrying the whole cost of delivery against a wholly deferred return.
Sixteen weeks left. With the £5,000 still to bill on Aura Phase 1 we reach £16,000 — eleven percent of the number.
The £5,000 on Aura is its own lesson: a big project that doesn't finish is money we have already spent earning and have not been paid for.
Six honest answers per engagement. More than two nos, and we should be asking why we're still doing it.
| Engagement | How it reached us | Work | Commercial model | Billed | Would we take it again? |
|---|---|---|---|---|---|
| Aura · Phase 1 | Will's network | Job visibility system | Fixed, £10,000 | £5,000 £5,000 unbilled |
Yes, but smaller |
| Aura · Phase 2 | Existing client | Customer quoting tool | Fixed, £4,000 | £4,000 | Yes |
| Kingfisher Property | Will — brother is MD | Lead generation | £2,000/mo handling fee client funds ads |
£2,000 · stalled | Couldn't deliver the promise |
| Optio Systems | Network | Site, funnel, voice, ads | Profit share | £0 | Not on these terms |
| The Purposeful Leader | Network | AI coach for the CEO | Quoted and agreed | £0 · slipped to 2027 | Yes — this is the shape |
Every single engagement arrived through somebody we already knew. Not one came from a stranger finding us. That is the honest answer to "is our work self-generated?" — and it is exactly what the audit engine is meant to change.
An AI version of Rob to coach when he isn't there. Quoted, agreed, then paused for his health. Likely 2027 — worth keeping warm, not writing off.
Initial conversations had. Needs a follow-up.
Initial conversations had. Needs a follow-up.
The CRM shows one opportunity worth £0. Reality is one agreed project and two live leads. The demand isn't the problem — the recording of it is, and so is the follow-up.
Action for today: agree who owns follow-up, and that nothing counts as a lead until it exists in the CRM.
A year of trading, five lessons, and every one of them points the same direction.
This is the difference between a business that wins on timing and one that wins on substance. It also happens to be far easier to advertise.
Most firms sit on one side of it. They either understand the business and can't build, or they can build and can't hold the commercial conversation. We have both ends and the middle, in three people.
Sales background and the polish. Gets in the room, holds the relationship, and makes the offer land.
Technical enough to know what's really possible, commercial enough to know what's worth doing.
Deep technical expertise — the thing most of our competitors are quietly missing.
The USP is not that we do AI. It's that we hold both conversations without a handover in the middle.
We've spent real budget in two places this year and the difference is not subtle. This matters because the entire 2027 plan assumes paid acquisition — and we're not guessing about it.
Niche B2B targeting on a channel we already suspected was weak for this. It performed as feared. The proposal now is to move the next £2,000 to Facebook instead.
Markedly different engagement for the money. Still no conversions — so what we've learned is about reach and cost, not yet about closing.
Careful with this one: neither campaign has converted, so the honest claim is that Meta buys us far more attention per pound, not that it sells. Converting that attention is what the £1,000 audit is designed to do — and it's a much easier thing to sell on Facebook than a niche B2B service.
A large consultancy engagement is a good month and a dead end. It's shaped around one client's problem, it can't be templated, and when it finishes we start from nothing again. We can't build a factory out of bespoke.
We're off and running. The right team, the right pieces, pointed at the right part of the market. Money is coming in.
What we haven't found is the widget — the one thing we can build a thousand times over. The rest of this session is about naming it and building the line that produces it.
Small businesses Will already talks to who want lead generation or process optimisation. Relationship-led, starts as a conversation. Aura came from here.
Meta spend selling audits to businesses who have never heard of us. The only source that scales without a relationship in front of it.
Will's HubSpot referral route, and others like it. Cheap, warm, and we've barely touched it.
Our instinct — mine especially — is to take on the big thing, think it through properly, and land it beautifully. That instinct is the thing most likely to slow us down now.
Smaller and more frequent beats bigger and less frequent. A client who gets a real win every fortnight renews without being asked. A client waiting eight weeks for something perfect starts wondering what they're paying for. This is a mindset shift, and it needs actual discipline to hold.
This is also why the concierge model works and the big-build model doesn't: the product itself is built out of small, fast wins.
The audit reveals what they're missing — and that they haven't the capacity to fix it themselves. That's the moment they look to us.
Immediate value — recommendations, tools, what they should be doing. And it surfaces the uncomfortable truth that they have neither the capability nor the capacity to act on it. We're not selling at this stage. We're being useful in public.
A long-term partnership making incremental improvements, one problem at a time. Light touch, repeatable, and the only tier that compounds. This is the widget.
Chunky, transformational, expensive. They don't scale and they never will — but they're extremely profitable, and sitting inside the business every month is how we see them coming.
One warning: tier three is where our instincts pull hardest, and it's the tier that can quietly eat the other two. It works as the upside on top of a retainer book. It does not work as the business.
A monthly retainer for a set number of hours. They bring us problems, we solve them with them — live, on a call, one at a time. They watch it happen and learn to do it themselves, so they're part of the solution rather than dependent on us.
This is the thing we can build a thousand times. Same shape for every client, different problems inside it. It recurs, it compounds, and it's the only asset here that's worth something at exit.
The container is identical every time. Only the problems inside it change, which is where the interesting work lives.
We start each month with a book, not a blank page. That's the difference between a business and a run of good luck.
The format forces the discipline: a problem solved this week, not a project delivered next quarter.
Solving real problems with people, not writing documents about solving them.
It needs judgement and range, which is what we have. It doesn't need a delivery army, which is what we don't.
A retainer book is an asset. A list of finished projects is a CV.
Deciding what we stop doing is harder than deciding what we start, and it's the part most strategy days skip. After a year of trying both, the honest position is that neither of these is where our energy should go.
We promise inputs we control — our time, our expertise, problems solved in front of them. We don't promise a number of leads, a conversion rate, or a revenue figure that depends on someone else's business working.
We can do it. We've proved we don't want to. It's crowded, it's judged purely on a number we can't control, and it isn't AI automation. It moves out of the core — passed to partners, or declined.
These are the two things we've spent most of the year on. Putting them down frees up the capacity the concierge needs.
The model only works if we can say yes to the non-core work without doing it ourselves. That means a bench of contractors and agencies who are good, reliable, and ideally using AI to be efficient themselves — so the work lands well and we stay out of it.
Paid media, web and app development, CRM implementation, design. What else keeps coming up?
Existing relationships first. What does a trial engagement look like before we put a client in front of them?
Referral fee, margin on pass-through, or white label. Each changes who owns the client.
Transformation projects are where the profit is and where our capacity runs out first. Four ways to fulfil them, and they trade margin against management time.
| Model | Cost to us | Our margin | Management load | Who owns the client |
|---|---|---|---|---|
| Offshore contractors | Lowest | Highest | Heavy | Us |
| UK / near-shore associates | Mid | Good | Moderate | Us |
| White-label agency | Highest | Thin | Light | Us, nominally |
| Straight referral | None | Fee only | None | Them |
The trap: offshore looks like the best margin until you price our own management time into it. For a team this lean, attention is scarcer than cash — a cheap contractor who needs daily supervision costs more than an expensive one who doesn't.
It's the relationship and the recurring revenue. The moment someone else is on that call, we're a middleman and the book stops being ours.
A small bench of trusted UK or near-shore people we can put in front of a client. More expensive per day, far cheaper in our time.
Self-contained, well-defined, low-ambiguity build. If the spec needs a conversation to understand, it isn't an offshore job.
Whole workstreams outside the wheelhouse — paid media, design, CRM implementation — go white label or straight to referral. We take a fee or a margin and stay out of the delivery entirely.
Everyone on the bench has to be AI-native.
We cannot sell AI automation and have it quietly fulfilled by someone building it by hand. What they deliver has to be the thing we advertised — or we're selling a story our own supply chain contradicts. This rules out most cheap offshore capacity, and that's the point.
Decision needed today: do we start building the bench now, before we need it, or wait until a project forces it? Building it now costs a little time. Waiting costs us the first big project we can't say yes to.
These are ambitions, not forecasts. The next two slides ask what would have to be true — and what the arithmetic says back.
| 2026 | 2027 | 2028 | |
|---|---|---|---|
| Audit revenue | |||
| Retainer revenue | |||
| Project work | |||
| Total revenue | |||
| vs target | |||
| Retainer clients | |||
| Corporation tax | |||
| Distributable | |||
| Per quarter |
Engine starts October 2026 and ramps to full rate over 18 months. Marketing at 10% of revenue; project work at £2,500 per retained client per year. CT 19% to £50k, marginal relief to £250k, 25% above. Three directors at the personal allowance, GM from mid-2027. Illustrative — confirm every rate with the accountant.
At 18 audits a month, 25% converting, £1,500 retainers, the model lands 2028 almost exactly on £1.2m. The compounding is real and it works.
2027 comes in around £370k against a £600k target, and 2026 can't get near £150k — an engine started in October has three months to run and a book that's barely begun. Recurring revenue is slow to start and then very hard to stop.
On these assumptions, 2026 runs at a loss.
Three salaries and the equipment spend land before the revenue does. That's not a modelling error — it's what starting an engine costs. Either the cost base phases in behind the revenue, or we fund the gap deliberately and call it an investment. What we shouldn't do is discover it in December.
Accept the ramp. The targets hold, they just each slide about a year right.
£2,000 retainers instead of £1,500 gets 2028 to £1.2m at lower volume — and 2027 closer.
Every month earlier is a month the book compounds. This is the cheapest lever we have.
Move the sliders on the previous slide and see which combination the room actually believes.
Salary set at the personal allowance — tax-efficient and deductible — with the rest taken as dividend.
No salary. Return comes through quarterly distributions and, in time, an exit.
Declared each quarter out of distributable profit, so returns track the business rather than lag it a year.
A general manager gets hired as soon as revenue and profit support it — modelled from mid-2027. That hire reduces the distributable pool in the year it lands, and is what makes the year after it possible.
We are not currently on the right kit for an AI automation business. Machines, peripherals, and the time to get productive on them.
Software and AI credits, accountancy, PI and PL insurance, banking, domains and hosting, Companies House, legal and contract templates, training.
Winchester every other week, two strategy days a year, one dinner. Small money, and the thing that keeps five people one company.
Modelled at roughly £37k in 2026 rising to £90k by 2028, plus contingency. Tell me what's missing.
We took real steps towards the Entrepreneurial Operating System and parked it when everything became all-hands. That was the right call at the time. It stops being the right call the moment we're running a factory.
The question isn't whether to restart it — it's what triggers the restart, and who owns it.
Capture live. Nothing leaves this room as "someone should".