Reports in your firm's voice.
Your voice took years to earn — the hedging you trust, the words your clients recognise, the exact sentence that follows a Condition 2. Upload 5–10 past reports and the AI reads them forensically, extracting a 25-dimension writing fingerprint it applies to every report it helps draft. AI speed; your signature.
The voice fingerprint
Twenty-five dimensions. One voice.
A firm's voice isn't one thing — it's a stack of habits across structure, tone, vocabulary, patterns and formatting. The AI measures all twenty-five independently, then writes inside the shape it finds.
How training works
Five steps from filing cabinet to fingerprint.
Training isn't a settings page — it's a forensic read of your firm's actual work. One upload session, and every future AI-assisted report starts in your voice.
Upload 5–10 past reports
Drop in DOCX, PDF or plain text files from your archive — anonymised reports are actively encouraged, because the analysis reads your writing patterns, not the client or property details. Mix Level 2 and Level 3 documents freely: the RICS level of each is detected automatically from its structure and content, and the profile keeps separate guidance for each level.
Each file becomes clean text
Every document is converted to searchable text before analysis. Very long documents are automatically trimmed to stay inside the per-document cost cap — a 200-page building survey can't run up a surprise bill. Each extraction is also logged against a daily firm-level spending ceiling, checked before anything is dispatched.
The 25-dimension forensic read
Our AI engine reads each report in around 30 seconds, working through the five categories in order — you watch real progress labels stream from the analysis itself: Extracting structural patterns → Mapping tone and voice → Building vocabulary fingerprint → Identifying writing patterns → Cataloguing formatting conventions.
The read is strictly evidence-based: for every dimension it must quote verbatim from your text, and if a dimension isn't clearly evidenced in a document it records nothing rather than guessing. A null is always better than a vague label.
Merge — frequency-ranked, not averaged
Observations from all documents merge into one firm fingerprint. Vocabulary lists are ranked by frequency across the whole set — a word present in eight of ten reports outranks one seen once. For the critical voice dimensions (section openings, defect descriptions, recommendation language) the most detailed observation wins outright, and verbatim exemplars are kept at full length. One verbose surveyor on one job can't skew the firm's voice.
Live on every run — automatically
From that point on, the profile is injected into every Otto auto-mode report, every AI Dictation routing session and every inline "Rewrite with AI" in the report editor. There is nothing to select per report. Add more samples whenever you like — the profile is cumulative, refining rather than replacing what it already knows.
The training pipeline
A five-report training run typically completes in a few minutes from first upload to a live, active profile — each document read independently, then merged into a single firm-wide fingerprint.
Illustrative example — one dictated observation, two voices
"The roof is in poor condition. Some slates are damaged and it should be repaired soon."
"The main roof is covered with natural slate. Several slipped slates were noted to the front elevation. This is likely attributable to weathered fixings and, if left unattended, could result in water ingress. We recommend that repairs are instructed within the next 12 months."
What the profile changes
The observation always comes from your inspection — the profile shapes how it reads on paper. Sentence structure, hedging, recommendation timing: the habits your clients recognise as yours. Example above is illustrative.
Verbatim anchors
Not labels. Quotations.
Generic AI tools describe writing in the abstract — "formal", "passive". The Style Profile quotes your reports back to themselves: exact passages, captured word-for-word, used as the template for every draft that follows.
Section B, captured in full
Section B — the overall opinion of the property — is the most personal passage in any RICS report and the first thing a client reads. The training quotes your Section B verbatim: your assessment opening, how you list priority items, your summary structure. It is the single most important extract in the fingerprint, stored at full length.
Condition-rating language, per rating
The exact text your firm writes after a Condition 1, 2 or 3 is captured as a direct quotation — including the phrase you use when an element is satisfactory and no defect needs recording. If your Level 3 profile lacks an example for a rating, the system borrows it from your Level 2 profile: same voice, different format.
Your defect structure, as a template
Many firms describe defects in a consistent arc — factual observation, probable cause, risk if unattended, recommendation with a timeframe. The training detects and records your firm's actual sequence, in your firm's actual words, and every auto-drafted defect description follows it.
Length discipline, measured — not guessed
The profile measures how long your firm's element comments actually run for each condition rating, and anchors every draft to those averages. A firm that writes two crisp sentences on a satisfactory element gets two crisp sentences — not three padded paragraphs of AI enthusiasm. This is the single biggest lever for making drafts feel like yours.
Strict priority order. The extraction works in tiers: verbatim voice anchors first (section format, openings, defect style, recommendations, signature words, Section B), register and tone second (formality, hedging, urgency, rating language), formatting conventions third. The things that make a report unmistakably yours are the things it fights hardest to capture.
Why voice matters
A voice that took years. Taught in an afternoon.
Every report in your archive is a lesson in how your firm writes. The Style Profile turns that archive into a working instruction — so AI speed never costs you the signature your clients trust.
1 · The archive
Years of judgement on paper
Your past reports already contain your firm's voice — the phrasing, the caution, the structure. Training reads 5–10 of them and finds the pattern.
2 · The problem
Same firm, different voices
Put two surveyors' reports side by side and clients notice: one hedges, one doesn't; one writes paragraphs, one writes lists. Consistency shouldn't depend on who got the job.
3 · The fix
One signature on everything
With the profile live, every AI-assisted first draft — whoever dictated it — arrives in the firm's voice. Your surveyors review judgement, not punctuation.
Honest by design
The profile knows how much it knows.
Every profile is graded by how many documents support it — and the AI is explicitly told how much weight to give the patterns. One report never masquerades as a firm's definitive voice.
You set the dials
A voice you control, not a black box.
The fingerprint is extracted automatically — how it's applied is up to you. Every control below ships today.
Where it applies
Trained once. Speaking everywhere.
The profile is injected wherever the platform generates or reshapes report text on your behalf — automatically, on every run.
Otto auto-mode
Every auto-mode report is drafted inside your fingerprint — section openings, defect structure, rating language and element length all anchored to your firm's own examples.
AI Dictation routing
When your spoken site notes are routed to RICS fields, the resulting field text is shaped by the profile in the same pass — your hedging and phrasing from the first draft.
Rewrite with AI, in the editor
Highlight any block in the editable report and ask for a rewrite — expand, condense, formalise — and the refinement carries your full style guide with it. Even edits speak in your voice.
Before you train anything: firms with no profile aren't left with raw AI. A professional corpus-wide baseline keeps untrained output in credible surveyor register — training simply replaces the generic baseline with your voice.
Safeguards
Cost-capped. Fault-tolerant. Yours.
Training reads your firm's most personal documents. The engineering around that is deliberately conservative — on cost, on failure, and on privacy.
Two-level cost protection
Each extraction has a per-document cost ceiling — very long documents are automatically trimmed before analysis rather than running up a bill. A daily firm-level spending ceiling is checked before any extraction is dispatched. Bulk-uploading a filing cabinet cannot burn a fortune.
Circuit-breaker fault tolerance
If the AI provider has an outage, repeated failures trip a circuit breaker: extraction fast-fails instead of hammering the service, then recovers automatically when it's back. Failed documents are marked clearly so you can simply re-run them.
Never blocks a report
Auto-mode works with or without a profile. If style extraction fails — or your firm hasn't trained yet — Otto continues without the style block rather than holding your report hostage. Style is an enhancement, never a dependency.
Privacy by default
Reports are processed only to extract your style fingerprint, using AI providers that do not train on customer content by default. The profile belongs to your firm and is applied only inside your firm's account. Anonymised samples are actively encouraged.
Why style profiling
Four ways to a consistent voice. One that scales.
Most practices already fight style drift — with a red pen, a locked template or a phrase list. Each helps. Only one produces a firm-voice first draft on every report without a senior surveyor in the loop.
| Capability | Style Profile | Senior red pen |
|---|---|---|
| Setup effort | One upload session | Every report, forever |
| Tone & hedging consistency | Automatic, firm-wide | If time allows |
| Structure consistency | Detected from your reports | Manual |
| Element length control | Measured from your averages | Manual cuts |
| Scales with report volume | Every report, no bottleneck | No — bottleneck grows |
| First draft in the firm's voice | Yes — that's the point | Never — voice added after |
These approaches aren't enemies — the profile and a Standard Phrases library are designed to run together, and your senior review always remains the final quality gate.
The quality trio
Voice. Language. Substance.
Three tools teach the AI your firm's standard, and they deliberately divide the job. Together they give Otto the most complete picture of how — and what — your firm writes.
Voice
Style Profile
How your firm writes: the 25-dimension fingerprint — sentence rhythm, hedging, structure, formatting — extracted from past reports and applied to everything the AI drafts.
Language
Standard Phrases
The exact words: discrete, pre-approved wording for specific elements and defects, inserted by click or dictated code — and offered to Otto during routing.
Substance
AI Training
What your firm says: rating philosophy, caveats and recommendation judgement learned from your past reports. Style Profile is the pen; AI Training is the opinion.
Otto — AI report writer
The auto-mode engine that reads your profile on every run and drafts the full RICS report from dictation and photos.
Level 2 Report Writer
The Level 2 Home Survey workspace your profile writes into — editable web report and branded PDF.
Level 3 Report Writer
Building Surveys with extended commentary — where a trained voice makes the biggest practical difference.
RICS Standards Check
Your voice, then the compliance pass — flag gaps against the Home Survey Standard before sign-off.
Who it's for
Every practice has a voice worth keeping.
Style drift costs editing time and client trust. Whether you run a multi-surveyor firm or work alone, the profile pays back on the first report.
Multi-surveyor practices
Several surveyors, one brand — and without a shared profile, noticeably different reports. A single firm-wide fingerprint brings every surveyor's AI-assisted draft to the same standard: same hedging, same rating language, same structure.
Clients hear one professional voice from your practice, whoever conducted the inspection.
Sole practitioners scaling up
You've spent years earning a voice clients trust. When AI Dictation and Otto let you turn around more instructions per week, the profile makes sure every one of them still sounds like you — not like a template.
Volume goes up; your signature stays on every sentence.
Growing teams & APC candidates
New associates absorb a firm's drafting conventions slowly, through months of red pen. With the profile active, their first AI-assisted drafts already carry the firm's vocabulary, hedging and condition-rating phrasing — before senior review.
Supervision focuses on site judgement, not tone correction — and drafting standards hold from day one.
Getting started
Your firm's voice, live in minutes.
No configuration, no technical setup. The Style Profile is included with every verified Home.co.uk surveyor listing.
Claim your firm's listing
Search for your firm by name or postcode — most RICS-regulated firms are already listed, so you just claim yours. The full toolkit, Style Profile included, activates once your RICS membership is verified.
Upload 5–10 past reports
Open Style Profile in your tools, drop in DOCX, PDF or plain text files — anonymised is fine, and a mix of Level 2 and Level 3 is ideal if you produce both. The RICS level of each document is detected automatically.
Watch the fingerprint build — then forget it
Each report is read in about 30 seconds, with live progress labels streaming from the analysis. Once merged, the profile applies itself to every Otto run, dictation session and AI rewrite from then on — nothing to select per report.
One upload session. Included with your verified listing. Retrain any time.
FAQ
Common questions about the Style Profile
Training & set-up
How many sample reports do I need to train the style profile?
5–10 reports give a robust profile. Each report is read in around 30 seconds, and the combined firm-wide fingerprint is ready within a few minutes of the first upload. The profile is graded by how many documents support it: one report produces a directional early profile, two to three a developing one, four to six an established one, and seven or more a mature profile the AI applies with high confidence. You can start with what you have and add more at any time.
Which file formats are supported for style training?
DOCX, PDF and plain text files are all supported. Most past reports are already in DOCX or PDF, so no conversion is needed — each file is turned into clean text before the AI reads it.
Does the style profile work with anonymised sample reports?
Yes — anonymised reports are actively encouraged. The training reads only writing patterns: sentence structure, vocabulary, hedging, tone and formatting. Redacted client names, property addresses and reference numbers have no effect on the quality of the extracted fingerprint.
What if I only have one or two sample reports to start with?
One or two reports produce a working fingerprint you can use immediately — but the system is honest about the difference. An early profile built from a single report is treated as directional guidance rather than definitive style, so one surveyor's unusually verbose day can't dominate your firm's voice. As you add documents the profile is re-graded automatically: patterns confirmed across multiple reports carry progressively more weight. Five or more reports is the recommended floor for client-facing work.
How long does the style profile last?
Until you replace it. The profile stays active on every run, and the underlying per-document observations are retained in your firm's account — so adding more samples later refines the existing fingerprint rather than starting from scratch. Upload a fresh set at any time to rebuild it cleanly.
The fingerprint
What exactly does the AI read in my sample reports?
Every section — but with a strict priority order. The highest-priority extractions are verbatim: the exact structure of your section openings, how your firm describes a defect from observation through to recommendation, your recommendation phrasing, your most frequently recurring vocabulary, and the full text of your Section B overall opinion — the most personal passage in any RICS report. Next come register and tone: formality, hedging, urgency expression and the precise words that follow each condition rating, including the phrase your firm uses when no defect is found. Formatting conventions — lists, numbers, measurements, abbreviations — complete the 25-dimension fingerprint.
What are the 25 dimensions, exactly?
Five in each of five categories. Structure: section format, heading style, subsection depth, paragraph length and overall document structure. Tone: formality level, passive/active balance, hedging style, authority tone and urgency expression. Vocabulary: signature words, technical jargon level, British English markers, word-repetition patterns and sentence starters. Patterns: section openings, section closings, defect reporting style, condition-rating language and recommendation format. Formatting: list usage, emphasis patterns, numerical style, measurement format and abbreviation patterns. If a dimension cannot be confidently determined from your samples, it is recorded as absent rather than guessed.
How are multiple reports merged into one profile?
Not by averaging. For vocabulary lists such as signature words and sentence starters, candidates are counted across every document and ranked by frequency — words present in eight of ten reports outrank words seen once. For the critical voice dimensions — section openings, defect descriptions, recommendation language — the most detailed observation across all documents wins. Verbatim exemplars are kept at full length. The result gets richer with each new sample, not noisier.
Does the AI detect whether a sample is a Level 2 or Level 3 report?
Yes. The RICS Home Survey Standard level is detected from each document's structure and content — you don't label or tag files before uploading. Your firm holds separate style guidance for Level 2 and Level 3 outputs, and each report you generate reads the guidance for its own level. If a Level 3 profile is missing a condition-rating example, the system falls back to your Level 2 profile's example for that rating — same voice, different format.
Can I retrain the profile if my firm's writing approach changes?
Yes. Upload a new set of sample reports and rebuild the profile at any time. Practices commonly retrain after bringing on a new senior surveyor whose style they want to formalise, or after a deliberate decision to modernise the firm's reporting tone.
Using it day-to-day
Where does the style profile actually apply?
Three places: every Otto auto-mode report, every AI Dictation routing session, and the inline "Rewrite with AI" feature inside the report editor. Anywhere the AI is generating or reshaping report text on your behalf, your profile shapes the output. There is nothing to switch on per report — once trained, it applies automatically.
Can different surveyors in my firm have different style profiles?
Yes. Alongside the firm-wide profile, individual surveyors can train a personal profile from their own reports. Each user chooses a priority: automatic (the firm profile applies by default, with your personal profile taking over where you've trained one), company-only, or personal-only. Most practices run a single firm profile for a consistent client-facing voice, with personal profiles for surveyors who have a distinct established style.
Can I control how strictly the AI follows the profile?
Yes. A per-user style adherence setting has three levels: high instructs the AI to follow your style guide with maximum fidelity; medium applies it where naturally applicable and adapts where the content requires; low treats it as general inspiration. You can also choose between structured and prose output for the D–G condition sections, and add free-text style notes of your own.
Does the style profile affect how condition ratings are described?
Yes — this is one of its most practical benefits. The training captures the exact text your firm writes after a Condition 1, 2 or 3 rating, as direct quotations from your sample reports — including the phrase your firm uses when an element is satisfactory and nothing needs recording. Every auto-mode report starts its condition commentary from those verbatim anchors, so hedging, urgency framing and recommendation timing stay consistent across every surveyor.
Does the style profile replace the need to review drafts before sending?
No — and it is not designed to. The profile shapes the first draft so it sounds like your firm from the outset, which materially reduces the editing that follows. A qualified surveyor still reviews and takes professional responsibility for every report. The profile reduces correction time; it does not remove your judgement from the process.
Trust, privacy & safeguards
Will my reports be used to train the underlying AI?
No. Your reports are processed only to extract your firm's writing style, and we use AI providers that do not train on customer content by default. The extracted profile belongs to your firm and is applied only within your firm's account.
What does training cost, and is spending capped?
Style training is included with your verified Home.co.uk surveyor listing — there is no separate subscription for the Style Profile. Extraction is cost-protected at two levels: each document has a per-extraction ceiling (very long documents are automatically trimmed before analysis rather than running up a bill), and a daily firm-level spending ceiling is enforced before any extraction is dispatched. You cannot accidentally bulk-upload a filing cabinet and regret it.
What happens if style extraction fails or the AI is unavailable?
Nothing blocks. If the AI provider has an outage, a circuit breaker fast-fails the extraction rather than hammering the service, and recovers automatically when it is back — you simply re-run the upload. And if a report is generated while your firm has no working profile, Otto auto-mode continues without the style block rather than holding up the report.
Does a consistent house style help with professional indemnity?
The style profile is not a PI tool in itself, but consistent language around Condition 2 and Condition 3 observations removes one source of discrepancy. When every surveyor in the firm uses the same urgency framing and hedging conventions, identical levels of diligence read identically — which matters if a report is later scrutinised by a solicitor or claims assessor. The surveyor's final review remains the safeguard; the profile ensures the first draft already reflects the firm's standard of care.
Can the style profile be used for snagging reports?
The profile is designed for RICS Home Survey Standard reports — Level 2 and Level 3 — where section structure, element codes and condition ratings are the primary outputs. Snagging inspections run through the separate Snagging Report Builder workflow and are not shaped by the style profile.
More tools
Explore the rest of the surveyor toolkit
Otto
AI auto-mode RICS report writer
Dictate on-site, upload photos, get a finished, fully-routed Level 2 or Level 3 report in minutes.
Level 2 Report Writer
RICS Home Survey Level 2
Field-by-field RICS Level 2 reports with inline AI suggestions, image gallery, comparables and one-click branded PDF.
Level 3 Report Writer
RICS Building Survey Level 3
Comprehensive Level 3 Building Survey reports with deep technical sections, custom rating scales and AI legal-adviser notes.
Surveyor Comparable Tool
Sold + rental comps in one click
Auto-fetch sold and rental comparables within a 1-mile radius, weighted valuation range, branded PDF appendix for any survey.
Snagging Report Builder
New-build defect tracker
Build professional new-build snagging lists with location-based defect tracking, photos, urgency, status and PDF export.
AI Dictation
Voice-to-fields routing
Speak observations on-site; AI routes every sentence to the correct RICS field with confidence scores and replayable audio.
Image AI
Auto-categorise survey photos
Upload up to 200 photos; vision AI tags each to the right RICS element, rates condition, extracts defects and picks the best 35 for the report.
Property Viewer
Pre-survey property intelligence
Look up any UK property and instantly see EPC, energy renovation costs, broadband coverage, environmental risks (flood, radon, mining, landfill), invasive plants, sales history and listing status — all in one dashboard.
Property Data Drawer
EPC + risks + comps in one panel
Open the property data drawer in any report to see EPC, environmental risks, listing history and comparables — copy-paste ready.
Standard Phrases
Reusable language library
Personal, firm-wide and industry-preset phrase libraries — insert with one click and use to train AI routing for voice consistency.
RICS Standards Check
Home Survey Standard compliance pass
One-click AI compliance pass against the RICS Home Survey Standard — flags missing content, weak ratings and structural gaps before you sign the report.
Home Payments
Paid faster, no invoicing
Home Payments mission control — customers pay at booking through a Stripe-verified checkout, payouts land on a schedule you can see, payment links cover your private work, and there is no invoice chasing ever again.
Bookings CRM
Every job, one pipeline
The surveyor CRM built into the platform — Homemove jobs land pre-paid with customer and property attached, enquiries convert to bookings, and the whole pipeline runs from one screen.
Calendars & Requests
A diary that fills itself
Per-surveyor diaries with live availability — keep your dates current and we book real jobs straight in, with inbound availability requests answered in a tap.
The firm inbox that knows the job
A full mail client scoped to your firm — @home.uk purpose mailboxes and per-user addresses, with every thread landing against the booking it belongs to.
Documents & e-sign
Signed before you set foot on site
Terms of engagement sent with the booking, tracked from sent to signed, chased automatically and stored against the job — no more chasing clients to sign.
Telephony
The firm number that never misses a job
A phone system built into the platform — take calls in the browser, forward to mobiles, ring the whole team at once, and keep a shared call log with optional recording.
Customer portal
Your client, never in the dark
A white-label portal under your firm's branding and domain — booking status, reminders and secure report delivery, so customers stop asking where things are.
Website builder
Built by asking, not designing
A firm website built and edited by talking to Inigo — brand and theme sync, a real section library, custom domains via CNAME, and enquiries that land in your pipeline.
Campaigns
Win the next job from the last one
Firm email campaigns with per-recipient tracking — sends, opens, clicks and read time — built on consented lists with one-click unsubscribe handled for you.
Reviews & badges
Proof that wins the next booking
Reviews earned on real completed jobs — scored by the homemover, shown on your directory profile — plus embeddable trust badges that carry the proof to your own website.
Performance
See the pipeline, beat the clock
Firm analytics without the spreadsheets — pipeline conversion, team performance and your own turnaround speed, the fourth lever of your rate.
Inigo
The AI assistant for surveyors
Ask in chat, dictate by voice, or click any element for contextual help — grounded answers on valuations, sales history and market data, inside the platform.
AI Training
Teach it your standard
Upload past reports and press Train — the AI learns your firm's defect language, ratings philosophy and phrasing, and Otto writes with it on every job.
Environmental Report
Every risk, one report
A one-click environmental risk report for any property — flood, ground stability, environmental factors and planning constraints, delivered as a PDF.
Free directory listing. Pay-as-you-go toolkit.
Every RICS-regulated surveying firm gets a verified profile, direct bookings from home.co.uk consumers and the surveyor toolkit included with the account. No subscription. No pay-per-lead. No commission on the job. Pay only when you use the premium AI tools.
2 minute setup · Verified against the RICS register · Trusted UK property platform since 1995