Grafton.
Project Barbara · Investment Strategy

Professionalise · Enhance · Transform

How we create value and underwrite risk · 5 June 2026

Every acquisition is underwritten on three escalating tiers, each pulling a different value lever. This governs how we price deals and feeds the three cases in every model. A paper for a colleague or an investor.

1Why a tiered strategy

We buy owner-operated lettings books and create value on three escalating tiers. They are cumulative — you always professionalise first — and each pulls a different value lever: the multiple, then earnings, then both plus scale. Framing every deal this way sequences the work after completion and, more importantly, lets us underwrite with a margin of safety.

TierLeverWhat we doArtefactCase
1 · ProfessionaliseMultiple (de-risk → transferable)Clean entity to sell; fix every compliance gap; proper accounts + monthly reporting; contracts on standard form; client money clean; fix operational failings; prove the book survives owner→professional transition; assess & default-outsource non-core salesProfessionalisation checklistDownside
2 · Enhance marginEarnings (EBITDA growth)Best-in-class off-the-shelf tech; remove family / right-size headcount; professionalise the management structure; strip owner-discretionary cost; fee & ancillary optimisationEnhance-margin playbookBase
3 · TransformEarnings ↑↑ + multiple ↑↑ + scaleThe Intelligent Letting Agency — custom AI-native central brain, hopper/centurion, automate the τ≈0 back office, roll-up operating leverageIntelligent Letting Agency + process mapUpside

2The governing principle — floor-protected underwriting

Underwrite to the Base (Tier 2). Require the Downside (Tier 1) to protect capital. Treat the Upside (Tier 3) as optionality we are not paying for.

This is the discipline that keeps a fast-moving roll-up out of trouble. We do not pay for the AI transformation in the entry price. We pay a price at which, if every clever thing we plan fails and all we do is professionalise the business and sell it, we still get our money back (plus the earnings yield we collected while we held it). The Base case is the return we expect; the Upside is asymmetric — free.

The corollary: a deal only clears if the Downside protects capital. That single test does most of the work of keeping us disciplined on price.

3The three tiers

Tier 1 — Professionalise (lever: the multiple)

Take a lifestyle / owner-operator business and make it an institutional-grade, transferable asset. This rarely grows earnings much — its job is to de-risk and re-rate: a clean, compliant, well-documented book whose income demonstrably survives a change of owner trades at a materially higher multiple than the messy version, and can be sold to a consolidator. Each item below becomes a checklist row we must tick:

Downside case = professionalise & sell. We model the floor as: fix it, make it transferable, exit to a consolidator at an institutional multiple. If that returns capital, we have our margin of safety.

Tier 2 — Enhance margin (lever: earnings)

These are owner-operated businesses, so there is real margin slack. Tier 2 is the competent professional-owner playbook — no custom AI, just good operating discipline and best-in-class off-the-shelf systems.

Base case = enhanced. What we expect with competent ownership and good systems — roughly half of the full cost-and-margin opportunity, with none of the AI risk. It is the return we underwrite to.

Tier 3 — Transform (lever: step-change + scale)

Re-found the business as an Intelligent Letting Agency — the custom AI-native architecture in that memo and mapped process-by-process in the process map. A central data brain, the hopper/centurion operating model, automation of the τ≈0 back office, humans only on the physical and the accountable — and, decisively, operating leverage across the roll-up as one governed brain runs many books.

Upside case = transformed. The post-automation run-rate plus roll-up leverage (RPE → £200k+, margin 55–60%). We strive for it; we don't pay for it.

4The three cases, in the model

The cases are one spectrum of how much of the cost-and-margin opportunity we capture, which the model already parameterises (normalisation → enhancement → automation). A case selector scales the levers:

5Worked example — Glenn Flegg (indicative)

Entry ~£1.01m ex-pipeline. Reported net profit £172k flatters; normalised it is ~£143k.

CaseWhat it assumesEarningsRead
Downside — ProfessionalisedClean, compliant, transferable; sold to a consolidator at ~2.3–2.5× recurring fee~£143k cleanpar on our money + the ~14% gross yield while held — capital protected
Base — Enhanced+ off-the-shelf tech, family removal, right-sizing, fee optimisation~£200k EBITDA~5.0× entry — solid double-digit IRR (the case we underwrite)
Upside — Transformed+ the Intelligent Letting Agency + roll-up leverage~£265k run-rate, then scale~3.8× entry; ~25%+ IRR standalone, more across the roll-up

The shape is the point: we don't lose in the Downside, we make a good return in the Base, and the Transform is free.

6How it runs — the standard artefacts

Every opportunity, through the DD & triage process, produces all three:

7The discipline, restated

Buy for the floor. Build to the base. The transformation is the asymmetry. Professionalisation protects the downside, enhancement earns the return, and the Intelligent-Business transformation — compounding across the roll-up — is the prize we get for free.