Removing client details from construction documents
Strip customer names, site addresses, payment details and worker names from construction records, swapping in consistent tokens, before anything leaves your systems. Buyers that describe their process de-identify as well, but going first keeps the decision with you.
Which fields must go?#
| Field | Replace with | Why |
|---|---|---|
| Owner or client name | OWNER-014 | Client identity is the client's, not yours |
| Project name | PRJ-2019-031 | Names often include the client or street |
| Site address, parcel | State and building type | Addresses identify owners and occupants |
| Bank, card, account numbers | Nothing: delete | No buyer needs them |
| Worker names | Trade and role: ELEC-FOREMAN-2 | Personal data of your staff |
| Phone numbers and emails | Nothing: delete | Direct identifiers |
| Design firm and sub names | ARCH-03, SUB-MECH-07 | Third parties' identities |
| Signatures and stamps | Nothing: delete | Identify a person, may be licensed seals |
Where do these details hide in each record?#
- RFIs
- Header fields, the "from" and "to" lines, attached sketches with title blocks, email chains pasted into the response.
- Change orders
- Owner name and address block, signature lines, contract number, references to the owner's internal approvals.
- Daily logs
- Header address, visitor lists, crew lists with names, inspector names, photo metadata with GPS position.
- Pay applications and invoices
- Remit-to details, lien waivers, retainage notes naming parties. Most firms leave these out entirely.
- Schedules
- Activity names such as "Smith residence kitchen" and file names; owner milestones that quote the contract.
Why use tokens instead of blanking fields?#
A blank loses the link between records. If OWNER-014 appears on the RFI, the change order and the schedule, the buyer can still follow one project from question to cost to delay without learning who the owner was. Keep the mapping table on your side only, and never send it.
PracticeRun the replacements from one mapping file for every system, so the same client gets the same token in Procore, accounting and the spreadsheets.
What do buyers say they do with personal details?#
| Buyer | Published process |
|---|---|
| micro1 | Done by micro1 (de-identifying, transforming sensitive fields). Seller can review samples. Location of the step: unstated. |
| Scale AI | Runs inside the seller’s own environment, pre-extraction. Personal data stripped. Seller signs off package by package. |
| Nyne | De-identified prior to release. Who performs it: unstated. |
| Labelbox | Labelbox side: ingestion, normalization, personal-data scrubbing. |
From buyers' own sites, checked 2026-11-07. Paid-referral notice: a listed buyer may pay this site if one of our introductions ends in a signed license. Rows are sorted by published pay floor; fees play no part in order or wording.
What is the final check before anything leaves?#
- Search the cleaned set for every client name, street name and project nickname on your mapping list.
- Search for patterns: phone numbers, email addresses, account numbers, ZIP codes.
- Open a random sample of files from each system and read them in full.
- Strip file metadata: author, GPS, revision history.
- Have a project manager who knows the jobs read the sample, then sign off in writing.
Questions#
Do I have to remove client details if the buyer de-identifies?
Not by every buyer’s rules, but you should. Removing customer names, addresses and payment details yourself means less identifying information ever leaves your systems, and it shows you exactly what a buyer would see.
Is replacing names with "Client A" enough?
Only if nothing else points back to the client. A building type, a city and a contract value together can identify a project. Coarsen locations to state level and round or remove figures that are publicly known.
What about photos?
Site photos carry faces, vehicle plates, signage and location metadata. Most firms leave them out of a first license; including them needs a separate review.
Who checks the result?
Someone who knows the projects. A project manager will spot an owner’s name in a file path or a nickname for a job that a tool misses. micro1 lets sellers look over representative samples after processing.
Sources and read dates: 4 pages
- micro1.ai/data, as of Nov. 7, 2026
- scale.com/data-partnership, as of Nov. 7, 2026
- nyne.ai/sell-company-data, as of Nov. 7, 2026
- prism.labelbox.com, as of Nov. 7, 2026
Get a quote
ZIP code first, then a reply address, then a handful of optional facts about your project records. Our answer names each buyer whose posted entry rules your firm clears, and nothing moves further without your say-so.
Disclosure: introduction service, not a buyer. A buyer that signs a license after our introduction may owe us a referral fee; your company is never charged.
Keep project files, exports and samples out of this form. We never take custody of company data or open it.
Got it. Here is what follows.
- Your headcount, country, records and ownership answer get compared with every posted buyer rule we track.
- A reply follows by email: buyer names that fit, each with its posted terms, or a plain note that none fits yet.
- Your details stay with us unless you approve one named buyer; only that buyer then hears from us.