The same customer. Three systems. Four spellings.
Duplicates, dead records and nineteen versions of the truth. We clean it, merge it and hand back data you can act on, with every change documented and reviewed.
What the job looks like
Exports from the systems that disagree: CRM, ERP, spreadsheets. Redacted is fine.
Profile, match, score, merge. Matching rules written down, exceptions listed, nothing changed without sign-off.
One clean master dataset, a report of what changed, and the rules that keep it clean.
What changes
- One record per customer, supplier and product
- Reports that stop contradicting each other
- Migrations that reconcile at cutover
- Data your next tool can load
49 jobs in Data Rescue.
Every one is a real job we do. Find the one that sounds like your week.
Duplicate Detection & Merge
Finds duplicate customers, suppliers, products or employees scattered across systems and spreadsheets, scores each match with a confidence level, and produces a reviewed merge plan.
“We have the same customer in three systems under three different names.”
Ask us about this →Data Cleansing & Validation
Fixes the everyday grime in business data: typos, inconsistent formats, missing values, impossible dates and out-of-range numbers, checked against agreed validation rules.
“Nobody trusts the numbers in our spreadsheets any more.”
Ask us about this →Data Wrangling & Transformation
Reshapes awkward data into the structure you need: pivoting, unpivoting, splitting combined fields, restructuring exports and converting between layouts and formats.
“Every report we receive is in a different shape and we waste hours rekeying it.”
Ask us about this →Data Consolidation
Pulls related data from multiple systems, exports and files into one coherent dataset with consistent columns, formats and definitions.
“Our data lives in six places and none of them agree.”
Ask us about this →Spreadsheet Consolidation
Takes dozens or hundreds of overlapping workbooks - regional copies, versioned files, personal trackers - and consolidates them into one governed workbook or dataset.
“We have 80 versions of the same spreadsheet and no idea which one is right.”
Ask us about this →Master Data Cleanup
Systematic cleanup of the records your business runs on - customers, suppliers, products, materials, employees - applying naming conventions, deduplication and completeness checks with enterprise master-data discipline.
“Our master data has grown into a jungle and every report built on it is suspect.”
Ask us about this →Master Data Governance Setup
Puts lightweight, practical rules around who can create and change master records, what a complete record looks like, and how quality is checked - governance sized for a real business, not a textbook.
“Anyone can create a new customer record and they all do it differently.”
Ask us about this →CRM Cleanup & Consolidation
Untangles a CRM full of duplicates, dead contacts, half-filled fields and abandoned pipelines so the sales team can trust what they see.
“Our CRM is so cluttered the sales team has stopped using it.”
Ask us about this →Customer Database Cleanup
Cleans the customer file itself - names, duplicates, statuses, contact details and account hierarchies - so billing, marketing and reporting all run off the same accurate list.
“Invoices go to one version of the customer and marketing emails go to another.”
Ask us about this →Supplier Database Cleanup
Cleans and deduplicates the supplier/vendor master: duplicate vendors, inconsistent payment terms, outdated bank and contact details flagged, and dormant suppliers identified.
“We cannot tell how much we spend with each supplier because they appear under multiple names.”
Ask us about this →Product Catalogue Cleanup
Standardises product data: names, descriptions, categories, units of measure, attributes and identifiers, fixing duplicates and gaps so the catalogue works for sales, web and stock alike.
“Our product data is so inconsistent that customers cannot find what we actually stock.”
Ask us about this →ERP Data Cleanup
Cleans the data inside an ERP - masters, open items, statuses and orphaned records - so the system reflects reality instead of ten years of workarounds.
“Our ERP is full of junk records nobody dares to touch.”
Ask us about this →ERP & System Migration Support
Does the data heavy lifting for a system change: extracting, cleaning, mapping and preparing load files so the new system starts life with trustworthy data.
“Our new system goes live in months and nobody owns getting the data ready.”
Ask us about this →Data Migration Projects
End-to-end handling of the data side of a migration: scoping what moves, cleansing it, transforming it to the new structure, loading in waves and proving nothing was lost.
“We are terrified of losing data when we switch systems.”
Ask us about this →Legacy System Data Migration
Rescues data out of old, awkward systems - end-of-life software, Access databases, bespoke tools, even flat-file exports - and gets it into a modern, usable form.
“Everything is trapped in an old system that only one retired employee understood.”
Ask us about this →Migration Validation & Cutover Reconciliation
Independently checks that a migration worked: record counts, value totals, field-level sampling and side-by-side comparison of source and target before and after go-live.
“We went live last month and we still do not know if everything came across.”
Ask us about this →Historical Data Reconstruction
Rebuilds usable history from fragments: old exports, PDFs of reports, backup files and paper-era spreadsheets stitched into a coherent historical dataset.
“We lost our history in the last system change and now we cannot see trends.”
Ask us about this →Data Quality Audit
A structured health check of your data: completeness, accuracy, consistency, duplication and timeliness measured across your key datasets, with a plain-English findings report.
“We suspect our data is bad but we cannot say how bad or where.”
Ask us about this →Ongoing Data Quality Monitoring
Sets up recurring automated checks that catch data problems as they appear - new duplicates, broken formats, missing fields - and reports them in a simple periodic digest.
“We cleaned our data once before and within a year it was a mess again.”
Ask us about this →Data Profiling
Systematically examines a dataset before anyone builds on it: value distributions, null rates, outliers, format patterns and hidden surprises, summarised in a readable profile.
“We keep starting projects on data we have never actually examined.”
Ask us about this →Data Enrichment
Fills the gaps in existing records by adding verified information from reputable public and licensed sources: company details, registration numbers, industry codes, geo-coordinates and more.
“Half the useful fields in our database are empty.”
Ask us about this →Record Linkage
Connects records that refer to the same real-world entity across datasets that share no common key, using deterministic rules plus AI-assisted fuzzy matching with transparent scoring.
“These two lists clearly describe the same people but nothing joins them.”
Ask us about this →Data Mapping & Field Mapping
Produces the definitive map of how fields in one system correspond to fields in another, including transformation rules, default values and the awkward fields that fit nowhere.
“Nobody can tell us which field in the old system feeds which field in the new one.”
Ask us about this →Schema Transformation
Converts data from one structure to another: flat files to relational tables, one system’s schema to another’s, denormalised exports into clean normalised form and back.
“The data we received is one enormous flat file and our new system needs it structured.”
Ask us about this →Import/Export File Preparation
Builds files that load first time: data shaped, encoded and validated to a target system’s import specification, with rejected-row handling sorted before you press the button.
“Every import we try fails with errors we do not understand.”
Ask us about this →Batch Corrections & Bulk Updates
Applies large-scale corrections safely: thousands of records updated in controlled, logged, reversible batches instead of weeks of manual editing.
“We need to change thousands of records and the thought of doing it by hand is unbearable.”
Ask us about this →Data Validation Rules Design
Defines and implements the rules that keep bad data out: field formats, allowed values, cross-field logic and duplicate checks, built into spreadsheets, forms or system settings.
“Bad data gets in faster than we can clean it out.”
Ask us about this →Metadata Documentation
Writes down what your data means: field definitions, allowed values, units, owners and quirks captured in a plain-English data dictionary anyone can use.
“The meaning of half our fields lives only in one person’s head.”
Ask us about this →Data Inventory & Catalogue
Maps every dataset the business holds - systems, databases, spreadsheets, shared drives - into one inventory showing what exists, where it lives, who owns it and how sensitive it is.
“We honestly do not know what data we have or where it all is.”
Ask us about this →Data Lineage Documentation
Traces where a number comes from: the journey from original entry through every system, export, transformation and spreadsheet to the report where it finally appears.
“Three reports give three different totals and nobody can explain why.”
Ask us about this →Reference Data Standardisation
Cleans up the small lists everything else depends on - country codes, categories, units, statuses, cost centres - into one agreed standard applied consistently across systems.
“Every system spells our categories differently so nothing joins up.”
Ask us about this →Date, Currency & Unit Standardisation
Fixes the silent killers in mixed data: ambiguous date formats, mixed currencies without rates, and quantities recorded in different units, converted to one documented standard.
“Half our dates are American, half are British, and nobody noticed for two years.”
Ask us about this →Free-Text Field Cleanup
Brings order to free-text chaos: notes fields, descriptions and 'other’ boxes classified into structured categories using AI-assisted reading, with every classification reviewable.
“All the useful detail is buried in free-text fields nobody can report on.”
Ask us about this →Address Validation & Standardisation
Checks, corrects and standardises postal addresses against authoritative references, fixing format inconsistencies and flagging undeliverable or incomplete entries.
“Parcels keep bouncing back because our addresses are a mess.”
Ask us about this →Contact Normalisation
Standardises names, phone numbers, email addresses and job titles: consistent casing, international phone formats, syntax-checked emails and duplicate contact points flagged.
“Our contact list has phone numbers in five formats and names in capitals, lowercase and everything between.”
Ask us about this →Stale Record Identification & Archiving
Finds the dead weight in your systems - dormant customers, obsolete products, ex-employees, expired agreements - and moves it into a safe, searchable archive under agreed retention rules.
“Most of what is in our system is ancient history clogging up every search and report.”
Ask us about this →Business Acquisition Data Consolidation
When you buy a business, merges its customers, suppliers, products and history into your systems in a controlled way, preserving what matters and flagging what conflicts.
“We bought a company and now we have two of everything.”
Ask us about this →Post-Merger & M&A Data Integration
The structured data workstream after a merger: aligning masters, mapping between the two estates, deduplicating across entities and building the combined view leadership needs.
“Six months after the merger we still cannot produce one combined customer list.”
Ask us about this →Acquisition Due-Diligence Data Support
Analytical support during due diligence: profiling a target’s data for quality, consistency and hidden surprises - customer concentration, dormant accounts, irregular records - so the deal team asks better questions. Analytical support only, not financial or legal advice.
“The seller’s numbers look fine but we have no way to check the data behind them.”
Ask us about this →Cross-System Reconciliation
Compares the same information across two or more systems - stock, balances, orders, headcount - identifies every mismatch and traces each discrepancy to its cause.
“Two of our systems hold the same data and they never agree.”
Ask us about this →Customer, Vendor & Product Master Merging
Merges parallel master files - two customer masters, two vendor masters, two product masters - into single survivorship-based records with history preserved on both sides.
“We are combining systems and each has its own version of every customer, vendor and product.”
Ask us about this →Chart of Accounts Alignment
Maps and aligns differing charts of accounts across entities or systems so consolidated reporting adds like with like. Data mapping and preparation support for your finance team - not accounting advice.
“Consolidating group accounts takes weeks because every entity codes things differently.”
Ask us about this →Location Hierarchy Consolidation
Straightens out sites, branches, warehouses, regions and territories recorded inconsistently across systems into one agreed hierarchy that reporting can roll up correctly.
“Every system groups our sites differently so regional numbers never match.”
Ask us about this →Business Unit Consolidation
Aligns how departments, divisions and business units are represented across systems, mapping legacy structures to the current organisation so cross-company reporting finally adds up.
“Every reorganisation breaks our historic reporting.”
Ask us about this →Multi-Company Data Harmonisation
Brings the data of several companies in a group onto shared standards - common codes, formats, categories and definitions - without forcing everyone onto one system overnight.
“Our five companies cannot even agree on how to spell the product categories.”
Ask us about this →Web Scraping & Online Data Collection
Collects publicly available online data - prices, listings, directories, specifications, published registers - into clean structured datasets, respecting site terms and legal boundaries.
“The data we need is sitting on public websites but copying it by hand is impossible.”
Ask us about this →Scheduled Data Scraping & Monitoring
Turns one-off collection into an ongoing feed: scheduled scraping runs that track changes over time - price movements, new listings, availability, published updates - with change alerts.
“We check competitor prices manually once a quarter and we are always behind.”
Ask us about this →Single Source of Truth Creation
Builds the one dataset everyone agrees to trust: consolidated, cleaned, documented and refreshed on a schedule, ending the argument about whose spreadsheet is right.
“Every department has its own version of the truth.”
Ask us about this →AI-Readiness Data Preparation
Prepares your data so AI tools work on it: cleaned, deduplicated, consistently labelled, well-described and structured so chatbots, copilots and automations give sensible answers instead of confident nonsense.
“We bought an AI tool and it confidently tells us things that are wrong.”
Ask us about this →Show us the messy spreadsheet.
One file. See it come back clean and reconciled.
Send us your problem →Or write to hello@tasksmithai.com