The Investment Process

1The governing principle — work follows conviction

We see far more opportunities than we will ever buy, so we spend effort in proportion to conviction. Most opportunities should die cheap, after a couple of hours' reading; a few earn the full underwrite. We never want a 14-tab model built for a deal we were never going to pursue — and we never want to pass on a good one because the work felt too heavy to start.

Everything up to the offer runs on the broker pack and the public record only — we don't touch the live business data until our offer is accepted and diligence opens. Two companion documents bound this one:

• Investment Strategy is the lens we judge through at every gate — the three-tier value ladder (Professionalise · Enhance · Transform) and the floor-protected underwriting rule.
• Diligence → Onboarding is the back half: once our offer is accepted (Under Offer), full diligence opens, and in our model that is the start of building the operating system — we begin ingesting the actual portfolio as the kernel of the operating-system data. This process delivers a live deal to its front door.

2The spine — one living deal memo that matures

Every target gets one deal memo, and every memo is a LiveDoc (a target-<name> wiki page + branded HTML — never bare markdown). The memo carries a status and a version + date, and climbs three rungs as we choose to invest more work:

It is the same memo throughout — the status and version say how cooked it is. A Preliminary v2 has been refreshed on new information but not modelled; an Investment Case v1 is ready to act on. The 10-section memo format is owned by the deal-memo skill; the rungs just fill in more of it.

3The Preliminary stage — read the pack, scan the public record

The cheap, public work is done up front, before any modelling, because it is exactly what tells us whether an opportunity is worth the heavier analysis. Two sources, both public:

• The pack — the broker's Business Sale information sheet and headline accounts: a qualitative read of the business, the book, the mix and the rough economics.
• The public record — what the outside world can see, independent of what the seller chose to disclose. Four strands, each summarised into structured data:

1. Reviews — across Google, Trustpilot, allAgents, Facebook. Per platform: rating, count, recency; a blended score; the recurring themes (top praise / complaints — slow repairs, deposit disputes, communication); whether the owner replies; and the landlord-vs-tenant skew (landlord sentiment matters most for a fee book). A read on operational health the accounts can't give.
2. Website — services and published fees (corroborates the pack), positioning and target market (prime vs volume), branding, portal/tech signals, the team page (corroborates the staff list and key people), and the branch & coverage footprint.
3. Listings — Rightmove / Zoopla / OnTheMarket, lettings and sales, split by available · under offer / let agreed · recently let / sold. Counts, asking rents and prices (validates avg-rent and the sales-desk read), and the let-agreed / under-offer ratio as a live demand-and-velocity signal. Visible let stock is also a sanity check on the claimed managed book.
4. Key individuals — the owners, directors and partners as people. Companies House (filing history, other and former directorships, disqualifications), LinkedIn and the web (background, tenure, age corroboration), any press or prior ventures. Surfaces who really runs the business, competing interests (other agencies, referral/mortgage entities — directly relevant to non-compete), and whether the operator we'd need to retain is credible and findable.

The scan's structured outputs are stored against the opportunity and feed the Preliminary memo's book-quality, market-position, ownership and risk sections — and the verdict that decides whether we climb to Analysis.

4The Investment-Case offer package — readied before we offer

The work of the Investment Case stage is not only to mature the memo to its final rung — it is to assemble the offer package: the three documents that make us ready to transact the moment we decide to proceed. They are kept separate and individually editable, because each is a substantial piece of work:

• Heads of Terms — the offer: structure, price, conditions, exclusivity. (Worked example: hot-glenn-flegg.)
• DD Plan — the prioritised list of what we must verify before completion, shaped by the structure (the HoT) and by what must be true to operate the book (the onboarding plan).
• Onboarding Plan (90 Day Memo) — how we take control and run the first 90 days: the Tier-2 enhance-margin playbook made concrete for this book — people, systems, the first professionalisation moves. (We'll settle its exact shape on the first real one.)

With the package ready, the proceed decision is simply: make the offer. The deal then moves Offer Made → Under Offer → Completion — and Under Offer is where Diligence → Onboarding runs: due diligence and the ingestion of the real portfolio as the kernel of the operating system are one continuous process during the offer period.

5The staircase — stages, deliverables, gates and skills

An opportunity carries one thing: its stage — its position in the funnel. Everything else — each memo rung, the model, the term sheet, the 90 day memo, the DD plan — is a deliverable linked to the opportunity: a separate, versioned, individually-editable document. The stage is the position; the deliverables are the evidence. Each stage has a skill that produces its deliverable, so we invoke exactly the increment we want ("run the analysis skill on Flatman") rather than an all-or-nothing triage.

Stages 2–4 are the heart of this process — the evaluation staircase, run on the pack and public record, each rung a deliberate decision to spend more. Stage 4 assembles the offer package; deciding to proceed means making the offer (→ Offer Made). When the offer is accepted (Under Offer) we move into Diligence → Onboarding — diligence and data-ingestion as one process — through to Completion. The old monolithic opportunity-dd-triage is retired — or kept only as a convenience that chains the stage skills in order.

6The decline skill

When we abort at any gate, the decline skill drafts a courteous decline email to the broker or seller — referencing the opportunity, giving a real but diplomatic reason (off-thesis / sub-scale / price / timing), and keeping the door open where we'd want it. It is drafted for review and never sent without sign-off. On send, the opportunity moves to Passed and the reason is logged. Brokers talk to each other; a graceful, prompt no is part of being the buyer they bring the next deal to.

7The current pipeline

Where the five live opportunities sit today, and the single next decision each is waiting on. (None has had the §3 public-record scan yet — it's a new Preliminary input we can add to any of them.)

The outstanding work collapses to: complete Glenn Flegg's offer package and make the offer; promote Flatman to Investment Case; publish Martindon + Hoopers' preliminary memos and park them. No new spreadsheets for the lukewarm ones — they sit at Preliminary until they earn the next step.

8How the deal-room represents this

The deal-room app mirrors the model:

• One stage per opportunity — its position on the staircase (§5): Sourced · Info Received · Preliminary · Analysis · Investment Case · Offer Made · Under Offer · Completion (+ Passed).
• Deliverables as linked records — each memo, model, term sheet, 90 day memo and DD plan a separate item with its own type, status, version and link, editable on its own, surfaced as link chips on the opportunity's pipeline card.
• A memo_status chip — Preliminary / Analysis / Investment Case, with version + date.
• The market-scan signals stored against the opportunity (reviews / website / listings / individuals) so they're comparable across opps.
• A league-table view — sorted by thesis-fit, so the board reads as "working vs. parked."

9The staircase at a glance

                 ── pack + public record only ──
Sourced         → opportunity + broker contact
Info Received   → document register
Preliminary     → memo (PRELIMINARY) + market scan ..... interesting? → up / park / decline
                  ├─ pack        → qualitative read + headline economics
                  └─ public rec. → reviews · website · listings · key individuals
Analysis        → financial model; memo → ANALYSIS ..... like it?     → up / park / decline
Investment Case → memo → INVESTMENT CASE
                  + offer package: Heads of Terms · DD Plan · 90 Day Memo ... PROCEED? → make the offer / decline
──────────────────────────── (offer made) ────────────────────────────
Offer Made      → Heads of Terms submitted ............. accepted?
Under Offer     → = diligence-onboarding: DD + ingest the REAL portfolio
                  as the KERNEL of the operating system
Completion      → done
                                          ↘  (abort any gate) → decline → Passed