Fixed-price building work.

For discussion · March 2026

Hal & Graham

What you need to know
02 / 22

What this business is

We're not a marketplace. We're the counterparty to both sides.

Customer
"Fix my roof"
Pays flexible or fixed price
Gets financing option
→
Pitch
Prices the job
Manages delivery
Guarantees outcome (fixed tier)
→
Builder
Consistent work
No lead fees
Delivers quality
Flexible PriceFixed Price
Customer getsCheaper quote. Pays any overages. We ensure builders are reasonable.Premium quote. One guaranteed price. Any overrun is our risk.
We earn15% margin, always. Zero pricing risk.28% margin target. We absorb overruns.

Revenue streams: 1. Construction margin · 2. Financing (via Kanda initially, own lending later) · 3. Future: insurance products, maintenance contracts

What you need to know
03 / 22

What that means legally

The model above makes us a principal contractor. The law treats us as the builder.

CRA
Consumer Rights Act 2015
We sub-contract, we remain legally responsible for the entire contract. We cannot separate the price guarantee from the quality guarantee.
CDM
CDM Regulations 2015
Domestic client duties transfer to us. We're accountable for site safety, planning, and welfare.
CIS
Construction Industry Scheme
Must register with HMRC, verify every subcontractor, deduct 20-30% tax from labour payments.
HGCRA
Housing Grants, Construction & Regeneration Act 1996
Must pay builders on schedule regardless of whether the customer has paid us. Cash flow risk sits with Pitch.
Legal
Legal position
We are the principal contractor. Liable for quality regardless of tier. Insurance: £7.5-19.5k/year (ELI, PLI, PI, contractors' all risks).

Sources: HSE CDM 2015 guidance, Consumer Rights Act 2015 Chapter 4, GOV.UK CIS Guide

What you need to know
04 / 22

We need a builder

We're legally on the hook for everything - quality, safety, and delivery. This isn't optional.

What we can't answer today
  • What does a quality inspection of a completed re-roof actually involve?
  • When a builder finds rotten timbers, who decides if it's a £500 fix or £5,000?
  • CDM regulation obligations when we're the principal contractor?
  • How do you know flashing, ventilation, and felt lapping are properly done?
  • CIS compliance, payment scheduling, subcontractor verification?
Who we need
  • Someone who's run residential roofing crews. 10+ years on tools, then managing.
  • Understands quality, estimating, and builder psychology.
  • Can own CDM compliance and site safety obligations.
  • This isn't a year-two hire. It's a day-one requirement.
The question
Co-founder (equity, commitment, decision-making power) or founding employee (salary, less equity, hired not chosen)?
What you need to know
05 / 22

Unit economics

Based on 10 hypothetical roofing jobs with realistic cost variance. 50/50 flexible/fixed split. No financing revenue in base case.

Avg job value
£9,600
Blended flexible/fixed
Gross profit / job
£1,422
Before CAC & opex. No financing.
CAC
£400
Central estimate (£300-500 range)
Break-even
28 jobs/mo
333 jobs/year
Two-tier pricing
FlexibleCustomer pays overages. We always earn 15%.£11,849 P&L / 10 jobs
FixedOne guaranteed price. We absorb overruns.£16,583 P&L / 10 jobs
Blended50/50 split£14,216 / 10 jobs
Monthly opex
Founders (x2)£16,666
Builder hire (all-in)£6,200
Overheads (inc. insurance)£5,500
Total£28,366/mo
Open interactive pricing model
What you need to know
06 / 22

Annual projection

At 30 jobs/month steady state. No financing revenue - see Kanda slide for why.

Revenue
£3.46M
£9,600 × 360 jobs
Gross profit
£512k
£1,422 × 360
CAC
-£144k
£400 × 360
Opex
-£340k
£28,366 × 12
Net profit
£28k
0.8% net margin
What this means
At steady state without own lending, this is barely profitable. The construction model needs to work at this level to justify the Series A that unlocks financing revenue. That's the real prize.

Ramp period not modelled. Months 1-6 are pre-revenue. Months 7-18 are partial volume. Steady state is month 19+. Financing revenue excluded from base case.

What you need to know
07 / 22

Sensitivity - what if we're wrong?

Same 10 jobs, four overrun scenarios. Flexible-tier jobs always make money (customer pays overages) - they contribute £11,849 per 10 jobs in every scenario. Only the fixed tier changes.

ScenarioFixed P&L (10 jobs)Blended P&L (10 jobs)Per-job contribution
Current model£16,583£14,216£1,022
Optimistic fewer overruns£21,572£16,711£1,271
Pessimistic two major overruns£13,770£12,810£881
Nightmare three blow-ups£10,473£11,161£716
Key finding
Per-job contribution = blended P&L per job minus £400 CAC. Even in the nightmare scenario it stays positive at £716. The flexible tier acts as a hedge - on 50% of jobs we have zero pricing risk. But margins are thin without financing revenue. Nightmare scenario doesn't cover opex.
What you need to know
08 / 22

Size of the prize

Our best guess, not a forecast. Team sizes are bottom-up. Key assumption: one ops person manages ~30 active jobs with AI-automated comms and scheduling (vs ~12 without). This is testable in month one.

Year 3: Single trade, regional
Volume
100
jobs/month, 2-3 regions
Revenue
£11.5M
Roofing only
Construction gross profit~£1.7M
Lending book~£10M outstanding
Spread income~£300k/yr
Team6-7 people
2 founders + ops lead + 2 regional ops + 1 admin
Opex~£550k
CAC (£350/job)~£420k
Net~£1M
Net margin~9%
Year 5: Multi-trade, scaling
Volume
200+
jobs/month, 2-3 trades, 4-5 regions
Revenue
£18M
Multi-trade
Construction gross profit~£2.9M
Lending book£30M+
Spread income~£900k/yr
Team9-10 people
+ 4 regional ops, quality/compliance, finance
Opex~£750k
CAC (£300/job)~£720k
Net~£2.3M
Net margin~13%
The real play
The construction business is the distribution channel. The lending business is where the real margin lives. This mirrors the Gen H playbook: build distribution first, layer financial services on top.

Scenario assumptions, not forecasts. Key unknown: jobs per ops person with AI automation. At 30/person we need the teams above. At 15/person, double them.

What you need to know
09 / 22

What if we go all-in?

The previous slide is the disciplined version. Here's what it looks like if we raise more and move faster.

The aggressive plan
  • Multi-trade from the start. Roofing + driveways from day one. Both satellite-measurable, per-sqm priceable.
  • Multi-region from month 6. London, Manchester, Birmingham with dedicated builder panels.
  • Own lending from day one. Apply for FCA immediately. Don't give 9-11% to Kanda.
  • 400+ jobs/month by year 2. Aggressive marketing, national brand.
  • Insurance products. Latent defects insurance, maintenance plans, home warranty.
Aggressive P&L
Year 3Year 5
Volume400 jobs/mo800+ jobs/mo
Trades / regions3 / 4-54-5 / national
Revenue£38M£70-80M
Construction GP£5.7M£10-12M
Lending + insurance£1.2M£3.5-5M
Team12-1419-22
4-5 regional ops leads + quality + finance + marketing + tech
Opex£1.2M£2M
CAC£1.4M£2.9M
Net£4.3M£8.6-12M
Margin~11%12-15%
What this requires
Seed: £2-3M (multi-region, FCA, warehouse equity, aggressive marketing). Series A: £10-15M (national expansion). VC pressure, board oversight, dilution. Every risk from slide 18 hits harder at this scale - Angi's quality control failure is more relevant, not less. The question is whether the prize justifies the risk and the personal commitment.
What you need to know
10 / 22

Financing strategy

How we think about customer financing - and why it's a two-stage play.

Stage 1: Kanda (day one)
  • Kanda (YC W21) is an FCA-authorised credit broker with a lender panel
  • We can offer customer financing from day one without our own FCA authorisation
  • The catch: Kanda takes 9-11% of financed job value as subsidy fee on 0% finance deals
  • At 40% financing take rate: £345-420 per financed job goes to Kanda - roughly a quarter of gross profit
  • We don't earn financing spread - Kanda's lenders do
Stage 2: Own lending (Series A)
  • FCA authorisation: £25-55k, 8-15 months
  • Warehouse facility: 15-20% first-loss equity (vs 5-10% for mortgages)
  • On a £10M loan book: £1.5-2M equity required
  • Forward flow deals achievable 12-24 months after first origination
  • Gen H track record helps with credibility but unsecured credit has different loss characteristics
Recommendation
Prove the construction model first with Kanda. Apply for FCA in parallel. Add own lending as Series A.
What you need to know
11 / 22

What we'd raise

Seed Round
~£700k
Prove the construction model. Defer own lending to Series A.
  • Pre-revenue burn (6 months)Founders £16.7k/mo, builder hire £6.2k/mo, overheads £5.5k/mo (inc. insurance)£170k
  • Ramp period losses (12 months)Scaling to 30 jobs/mo. Gross profit growing but doesn't cover opex until ~month 15£190k
  • MarketingCAC £300-500/job, Google Ads £5-10k/mo, local SEO, leaflets£100k
  • Technology build£50k
  • Pricing mistakes reserveOnly fixed-tier jobs carry risk (50%). Covers ~40 bad jobs.£60k
  • Contingency (20%)£115k
Funding sources: Self-fund · Angels · Institutional seed
Series A
~£2-3M follows once the construction model is proven. Funds FCA authorisation + warehouse equity for own lending.
The market
12 / 22

The market nobody serves

£3-4bnUK residential roofing, annually
50,000+businesses. ~90% sole traders or micro
0dominant national residential roofing brand
  • No legal requirement to be qualified or registered
  • Only ~1,100 NFRC members (2-3% of market)
  • Skills shortage: ~10,000+ roofers short, ageing workforce
Nobody does what we'd do
  • CheckaTrade, MyBuilder - marketplaces. Discovery, not certainty.
  • TrustMark, Which? - endorsement schemes. No price guarantee.
  • HomeServe - managed service, but only simple repairs.
  • Thumbtack (US, $3.2bn) - only books estimate visits.
  • Houseace (AU) - tried algorithmic fixed pricing. Won awards 2021. Silence since.

Nobody offers fixed-price managed construction. Not in the UK. Not globally.

The market
13 / 22

What customers actually say

Evidence from MoneySavingExpert, Reddit, consumer forums.

"I got three quotes - £4,000, £6,500, and £9,000 - how do I know which is right?"
50-100% quote variation for identical work is routine
"If someone could just tell me 'this is what it costs, guaranteed' I'd bite their hand off."
MoneySavingExpert forum
40%of homeowners postpone necessary roof repairs due to cost uncertainty and trust
10-15%premium consumers will pay for certainty (TrustMark research)
The market
14 / 22

What builders actually say

Platforms are hated
"You pay for leads, half don't respond, and then you're competing with 5 other firms"
  • One tradesperson spent £21,409 over ~14 months on CheckaTrade for almost nothing
  • CheckaTrade: £90-140+/month with 12-month lock-in
  • MyBuilder: 2.9/5 on review sites, 795 formal complaints from trades
Builder economics
Self-employed income£30k-60k/yr
Day rates£200-350
Gross margins25-40%
What Pitch offers builders
Consistent, paid work. No lead fees. No chasing quotes. Payment terms. Just build.
The market
15 / 22

Pricing risk: the real data

20-35%of roofing jobs experience some cost overrun
Causes (ranked)
  • Hidden timber rot (~40% of overruns). Only found once tiles stripped. £1k-5k+.
  • Additional work discovered (~20%). Chimney, flashings, valleys.
  • Structural issues (~15%). Sagging, inadequate structure.
  • Material miscalculation (~10%). Complex geometry.
  • Access complications (~10%). Scaffolding more complex than quoted.
5-10%of jobs have severe overruns (40%+). This is where you lose money.
Key insight
Property age is the strongest predictor. Post-1960 concrete tile roofs are predictable. Pre-1940 (Victorian/Edwardian) are significantly riskier. In year one, overweight post-1960 properties and price pre-1940 conservatively.
How it works
16 / 22

AI is a superpower, not the USP

It lets a team of 5-10 execute at the speed that would normally require 50+.

Estimating team→Satellite imagery + per-sqm pricing models
Call centre→Automated customer comms
Sales team→AI-generated content, SEO, ad creative at volume
Dispatcher→Builder matching and job routing
Back office→Ops automation

The competitive edge isn't "we have AI." It's "we execute at much lower fixed cost than anyone in this space, because AI handles the work that normally requires headcount."

How it works
17 / 22

We already built it

A working prototype that generates instant two-tier quotes from a postcode.

How it works
  • Enter a postcode - that's all the customer does
  • OS Data Hub returns the building footprint polygon via API
  • Roof sections detected - area, material, pitch calculated
  • Regional pricing applied - materials, scaffolding, labour
  • Two-tier quote generated - flexible and fixed price, instantly
Start with roofs, then expand
  • Roof area measurable from satellite. £5-18k tickets. Fragmented market.
  • Then: driveways (satellite), plastering (photos), fencing (linear metres)
  • Never: rewiring, plumbing, structural work with hidden conditions
What the prototype proves
Instant, remote quoting is technically feasible using government mapping data. No site visit needed.
The stress test
18 / 22

What could kill this

01
Pricing risk
We're writing insurance on construction costs. A few bad jobs (£15k quoted, £25k actual) wipe months of margin.
02
Adverse selection
Customers most attracted to fixed-price guarantees have the riskiest properties. Victorian terraces self-select.
03
Quality control vs lean team
We guarantee outcomes but subcontract execution. When the roofer does a poor job, the customer comes to us.
04
Builder supply
Good roofers don't need lead gen. Our panel may skew mid-tier. Quality control question gets louder.
05
Financing dependency
Construction margins alone are thin (0.8% net). Kanda takes 9-11% of financed amounts as interim. The real economics depend on Series A unlocking own lending.
The stress test
19 / 22

The Angi post-mortem

The closest anyone has come to our model. A multi-billion dollar company. They failed.

What happened

Angi launched "Angi Services": fixed-price, managed model, act as principal, dispatch subcontractors, guarantee price and quality.

-71%Services revenue drop, Q4 2023
-19%Overall revenue decline by Q1 2025

Failure mode: quality control. Dispatching subs you don't closely manage means you own every bad outcome.

Why we might be different
  • Curated panel, not dispatch marketplace. Relationships, not algorithms.
  • Start narrow (one trade, one region) vs scaling nationally from day one.
  • Builder co-founder who understands the supply side from the inside.
Why we might not be
  • "Different" is what every startup says. Angi had billions and couldn't make it work.
  • Quality control doesn't get easier with volume - it gets harder.
The stress test
20 / 22

Why us

What we bring from Gen H
  • 6 years building a regulated lender
  • Risk modelling and pricing - our actual day job
  • FCA authorisation process - we've done it
  • Consumer credit and financing product design
  • Operations at scale in a regulated environment
  • Product and technology leadership

Most construction tech founders don't have any of this. The financial architecture of this business is what we're genuinely good at.

What we don't bring
  • Construction management experience
  • Builder relationships
  • On-site quality assessment capability
  • Knowledge of CDM regulations, building standards
This is why slide 4 matters
Our financial services DNA is the foundation. Their construction DNA completes it. Without both, this is a finance company that doesn't understand its product.
The stress test
21 / 22

Decisions for Graham

Split into what we decide together and what we need to figure out.

Decisions to make together
  • Full-time timing. Day one, or 3-6 months alongside Gen H? Saves ~£130k but slows everything.
  • Founder pay. Model assumes £100k each. Lower extends runway. Both have young families.
  • Scale ambition. Profitable-before-Series-A discipline, or raise bigger and scale faster?
  • Builder person. Co-founder (equity, commitment, decision-making) or founding employee (salary, hired not chosen)?
Questions to answer
  • Ops workflow: accepted quote to completed job - what does it look like?
  • Insurance specifics: PI, public liability, product liability, contractor's all-risk?
  • Builder recruitment: how do we find our first 5 roofers?
  • Geographic launch: London (highest margin, highest risk) or suburbs?
  • IR35 structure: are subcontractors inside or outside?
  • Mid-job discoveries: who decides scope changes?
What's already answered
  • Financing route: Kanda day one, own lending Series A
  • Legal structure: principal contractor, CRA/CDM/CIS obligations mapped
  • Pricing model: two-tier flexible/fixed, 50/50 split
  • Seed ask: ~£700k to prove construction model

Let's stress-test this together.

Open interactive pricing model

Hal & Graham · March 2026