Remote data operations — worldwide

Send us your messiest export. It comes back clean.

Duplicates everywhere, an import gone wrong, reports nobody trusts. We take the mess off your plate — fixed scope, every change logged and reversible, reconciled to ±0. You only pay after acceptance.

You get: half-page scope with fixed price + free sample on ≤1,000 rows, reply within 24h. Your sample file is deleted on your request — deletion is verified against the disk and confirmed in writing (see data handling). Prefer email? [email protected]

See the method first

Sample reconciliation plot: 5,210 input records resolved to 4,988 clean, 201 merged, 21 exceptions — identity ±0

Sound familiar?

Three situations we were built for.

The inherited mess

You just inherited the CRM.

New role, new system — and years of other people’s imports, duplicates and half-finished cleanups. Everyone expects you to fix it. Nobody can tell you where to start.

We start where the data actually is — and clean it so every change is logged and reversible
You get a base you can trust, plus the paper trail to show for it

The migration deadline

You’re migrating — and the export is a mess.

New platform next month, and the export is full of duplicates, broken fields and mystery values. Whatever gets imported now, you’ll live with for years.

We clean and map before the import — so the mess stays behind, not comes with you
Conformance-checked staging file your team imports in the sandbox

The numbers argument

Nobody trusts the numbers.

Sales quotes one customer count, marketing another, and the dashboard a third. Every meeting starts with an argument about whose data is right.

We reconcile to ±0 — input = clean + merged + exceptions, scripted, not “trust me”
Every figure traceable, every fix logged — the argument ends

Capabilities

Three fixed-scope units. Nothing open-ended.

Acceptance criteria are agreed before work starts — and never softened afterwards.

WU-1 · Cleanup pass

Data hygiene sprint

Your exported CRM or e-com data — yours or a client’s — comes back import-ready. Judgment rows are flagged with reason codes — never silently “resolved”.

Dedupe — zero duplicate pairs survive
Change log — every edit reversible
Reconciliation — before = after, ±0
Exceptions — flagged, never guessed

WU-2 · Migration prep

Pre-migration pack

The mechanical 70% of a migration, packaged so your seniors keep only the decisions. Local conformance is ours; the sandbox dry-run stays on your side.

Mapping workbook — unmapped fields explicit
Staging file — imports clean in your sandbox
Issue log — severity-classified
Rollback notes — the exit is written down

WU-3 · Reporting

Build-to-spec dashboards

Your analysts define what numbers mean; we build, reconcile and document them so they stay alive without tribal knowledge.

Reconciled to source on spot-checks
Refresh doc your team re-runs alone
Source notes — no tribal knowledge
Repeat specs → zero-touch refreshes

Process

How an engagement runs

Eight steps, three phases, one rule: you only pay after acceptance.

Phase 01 — Scope · steps 01–03

Agree before work starts

You send a spec-shaped question A messy export, a migration, a slipping report.
Half-page scope in return Inputs, outputs, criteria verbatim, fixed price, date.
Written go NDA on your paper. Sandbox or file access only.

Phase 02 — Run · steps 04–06

Executed and checked

Sample slice first ≤2h of work so you sanity-check the format.
Full run + independent review Scripted checks plus a separate review pass.
Delivery you can audit Outputs, change log, README a junior can revert with.

Phase 03 — Close · steps 07–08

You only pay on acceptance

One revision round included Our defects fixed free; scope changes re-quoted.
Invoice after acceptance Nothing owed until criteria are confirmed met.

Proof

Trust is a checklist, not a promise.

Every batch carries a machine-checkable evidence trail. Shape of a real change log entry (dummy data):

# changelog.csv — cleanup pass WU1-sample
row_id,field,before,after,action
00412,email,"  [email protected]  ",[email protected],trimmed_whitespace
00788,notes,=2+3+cmd|' /C calc,'=2+3+cmd|' /C calc,formula_injection_neutralized
01033,[whole row],dup_of:00412,,merged
02177,email,[email protected],EXCEPTION,plus-tag vs base address → your decision
03402,email,,EXCEPTION,null-key → excluded

# scripted identity, ±0:
input(5210) = clean(4988) + merged(201) + exc(21)

An honest note instead of testimonials: this practice is new, so there is no client list — and we won’t invent one. What we offer instead is inspectability: criteria agreed before work starts, a full change log after it, and a free 100-row sample on your own anonymized export.

If a check fails twice on our side, you get the defect analysis — not a quiet deadline slip.

Terms

Working terms

Pricing

Cleanup: $300–800 per ≤10k-record batch · pilot $150–300 · larger volumes via scoped pilot
Migration: $500–1,500 by volume · single-object $250–400
Dashboard: $250–600 build · refresh $200–500/mo
Model: Fixed per unit, quoted before start — never hourly drift
Currency: All prices USD, invoiced net — taxes per your jurisdiction

Data & confidentiality

Access: Sandbox / file exports only — no production writes
Retention: Deleted on your request — verified against the disk, confirmed in writing
Tooling: File-based work, never touching your systems; nothing undisclosed touches your data
NDA: We sign yours, freely

Working rules

Turnaround: Pilots 24h · batches 1–3 days · prep 2–5 days
Payment: Deliver-then-invoice; transfer after written acceptance
Out of scope: Certifications, cutover sign-off, production writes, on-site
Incidents: Work stops same-day; you are told, never quietly patched

$0Free sample · ≤1,000 rows · inside 24h

Start with the messiest export you have.

One anonymized dataset in — cleaned file, change log and exception report back inside 24 hours, free. If it’s useful after that: pilot slices run $150–300, deliver-then-invoice.

Reply within 24h · or write directly: [email protected]