- A data cleanse usually covers deduplication, address and postcode formatting, trading checks, contact updates, email validation, TPS/CTPS screening, suppression and data append.
- Returned post, email bounces, 'she left last year' calls and duplicate CRM entries are all signs a database needs cleansing.
- Removing duplicates, tidying spaces and capitals, splitting names and standardising postcodes can be done free in Excel or most CRMs.
- UK GDPR Article 5(1)(d) says personal data must be accurate and, where necessary, kept up to date.
- If a cleansing service processes your contacts' personal data for you, UK GDPR requires a written contract between you and them.
- If an old list is badly out of date, fresh records can be the simpler option: 5,000 new records cost £1,225 + VAT.
01
What do data cleansing services actually do?
A data cleansing service compares your records with up-to-date reference data, corrects what has changed, removes duplicates and flags anything it can't verify. Most B2B data cleansing runs through the same eight checks, usually in this order.
- Deduplication. Finds the same business or person entered more than once, even when the spelling differs ("J Smith Ltd", "J. Smith Limited", "Smith J Ltd"). Good matching uses several fields together, such as name, postcode and phone number, and lets you choose which version of a record survives.
- Address and postcode standardising. Puts addresses into a consistent layout, fixes postcodes that are missing a space or have the wrong case, and corrects towns and counties against an address reference file. This makes post deliverable and lets you select by area reliably.
- Trading checks. Flags businesses that appear to have closed, been dissolved or moved. For limited companies this often includes a check of status on (Companies House); remember a registered office can differ from where the business actually trades.
- Contact updates. Identifies named contacts who have left or changed role, and where possible replaces them with the current person in that job. This step usually needs fresh research or a reference database of people, so it's where services differ most.
- Email validation. Checks that each email address is correctly formed, that the domain exists and accepts mail, and where possible that the mailbox exists, without sending a marketing message. It flags role addresses (info@, sales@) separately from named ones.
- TPS and CTPS re-screening. Checks every phone number against the (TPS) and (CTPS) again. Registrations change all the time, so a number that was clear when you got it may not be now.
- Suppression. Removes or marks anyone who has opted out, objected or asked not to be called, using your own do-not-contact list, plus any customers you don't want included in a prospect campaign.
- Data append. Fills gaps: adds a missing phone number, website, employee band, industry code or named contact to records that lack them. Appended data is new personal data in many cases, so the same rules apply to it as to any other data you collect.
You normally get back two things: the cleaned file, and a report showing how many records were changed, flagged or removed at each step. Ask for flagged records in a separate column rather than silently deleted, so you can decide what to do with them.
02
How do you know your marketing database needs cleansing?
The clearest signs are practical ones you'll notice in day-to-day use: letters coming back, emails bouncing, phone calls ending with "they don't work here any more", and the same business appearing twice in your CRM.
- Returned mail. Envelopes marked "gone away", "addressee unknown" or "business closed". Each one usually means the whole record is out of date, not just the address.
- Email bounces. Hard bounces (the address doesn't exist) and rising soft bounces. Your email platform shows these per campaign; if they creep up from one send to the next, the list is ageing.
- "No longer here" calls. Callers being told the contact left, the number belongs to someone else, or the line is dead. Log the outcome on the record every time, or the next caller will make the same call.
- Duplicates in your CRM. Two sales people chasing the same company, or one prospect getting the same letter twice. Duplicates often appear after importing several lists or merging systems.
- Gaps and odd formatting. Postcodes in a mix of formats, names in capitals, phone numbers without area codes, blank job titles. These make selection and personalisation unreliable even when the business is still there.
- No source or date. Records where nobody knows where they came from or when they were last checked. You can't answer "where did you get my details?" without this, and you can't judge how old the data is.
If you see several of these at once, start with the checks you can do yourself (next section) and then decide whether a full cleanse or fresh data is the better use of money. Our complete guide covers routine upkeep between campaigns; this page is about the bigger clean-up.
03
Can you cleanse a database yourself in Excel or a CRM?
Yes, a lot of database cleansing is free to do yourself in Excel or your CRM: remove exact duplicates, tidy spaces and capitals, split full names into first and last, standardise postcodes and filter out blank or obviously broken records. Always work on a copy of the file.
Step by step in Excel
- Save a copy first. Keep the original untouched in a separate file. Add a column called "Source" and another called "Date checked" if you don't already have them.
- Remove exact duplicates. Select the data, go to Data > Remove Duplicates and tick the columns that identify a record, for example company name and postcode, or email address. Excel only catches exact matches, so do this after the tidying steps below and run it again at the end.
- Trim stray spaces. In a new column use
=TRIM(A2)to remove leading, trailing and double spaces, copy it down, then paste the results back as values. Hidden spaces are a common reason duplicates aren't spotted. - Fix capitals.
=PROPER(A2)turns "JOHN SMITH" into "John Smith". Check the results by eye: it also turns "McDonald" into "Mcdonald" and "IT Services" into "It Services", so don't run it on company names blindly. - Split names. Use Data > Text to Columns with a space as the delimiter, or in current versions
=TEXTBEFORE(A2," ")for the first name and=TEXTAFTER(A2," ",-1)for the last name. Look out for titles (Mr, Dr) and middle names. - Standardise postcodes.
=UPPER(SUBSTITUTE(A2," ",""))removes spaces and capitalises; then=LEFT(B2,LEN(B2)-3)&" "&RIGHT(B2,3)puts the single space back before the last three characters, giving a consistent format such as SW1A 1AA. - Filter blanks and junk. Turn on Data > Filter and use "(Blanks)" on key columns (postcode, phone, email) to find incomplete records. Filter emails that don't contain "@" or contain spaces, and phone numbers with too few digits.
- Suppress before you use it. Match the list against your do-not-contact list and remove those records (for example with
=COUNTIF(Suppress!A:A,A2)>0to mark matches), and keep the suppression list itself so opt-outs are never re-added.
In your CRM
Most CRMs have a duplicate check or merge tool: use it rather than deleting, so call notes and history are kept. Set required fields and a fixed postcode format on import forms, and make "lead source" mandatory. Run a saved report of hard bounces and "gone away" outcomes each month and deal with them in one go.
What you can't easily do yourself: confirm a business is still trading, find who replaced a contact who left, validate mailboxes without sending to them, or screen numbers against the TPS and CTPS without a licence to the registers. That's the part a data cleansing service or fresh data covers.
04
Does UK GDPR require you to keep marketing data accurate?
For personal data, yes. UK GDPR Article 5(1)(d) says personal data must be "accurate and, where necessary, kept up to date", and that every reasonable step must be taken to erase or correct inaccurate data without delay. (legislation.gov.uk: UK GDPR Article 5)
In plain English, that means:
- It applies to records that identify a person, such as a named contact or an email like jane.smith@company.co.uk. A generic info@ address with no name isn't personal data, though keeping it accurate is still good practice.
- "Having regard to the purposes" matters: how accurate data needs to be depends on what you use it for. Data you use to contact people should reflect who is actually there now.
- "Erased or rectified without delay" means acting when you find a problem, for example updating a record when a call reveals the contact has left, not waiting for the next annual clean.
- The law doesn't set a fixed cleansing schedule. It asks for reasonable steps, so a simple routine you can show you follow (bounces removed after each send, returns logged, periodic checks) is the practical answer.
Accuracy sits alongside the other rules on marketing to businesses, such as screening calls and honouring opt-outs. Our complete guide summarises the law on B2B marketing by channel.
05
What should you ask a data cleansing service?
Ask what they check your data against, how often that reference data is updated, what happens to your data during and after the job, and whether they'll sign a written data processing contract. If a cleansing service handles your contacts' personal data on your behalf, UK GDPR requires that written contract.
About the checks
- What do you check against?. Ask for each step in the list above: which source confirms trading status, which address reference is used, how contacts are verified and how emails are validated.
- How often is your reference data updated?. A cleanse is only as current as the data it compares against. Ask when each source was last refreshed and how changes are picked up.
- What do I get back?. Ask for a sample report and the file layout: corrected fields, flags, reasons for removal, and whether flagged records are returned separately.
- Can you run a test?. A small sample of your own records shows how many matches, corrections and flags to expect before you commit the whole file.
- How is it priced?. Per record submitted, per record matched or per field appended? The answer changes the cost a lot on an old list.
About your data
- What happens to my data?. Where it's stored, who can see it, whether it's added to their own database or used for anyone else, and when it's deleted or returned at the end of the job.
- Will you sign a data processing agreement?. Under UK GDPR Article 28, processing by a processor must be governed by a contract that sets out the subject matter, duration, nature and purpose of the processing, the types of data and the controller's rights, and Article 28(9) says it must be in writing, which can be electronic. (legislation.gov.uk: Article 28) The ICO puts it simply: whenever a controller uses a processor, there must be a written contract in place. (ICO: contracts)
- Do you use sub-processors?. If they pass your data to another company, the ICO says they need your authorisation and a written contract with that company too.
- How is data sent and returned?. Encrypted transfer rather than a plain email attachment, and a clear route for returning the cleaned file.
Here you're the controller (you decide why and how the data is used) and the cleansing service is the processor (it works on the data for you, on your instructions). The ICO's minimum contract terms include processing only on your documented instructions, confidentiality, security, and deleting or returning the data at the end.
06
Should you cleanse an old list or buy fresh records?
Cleanse data that has history you want to keep, such as customers and past enquiries. For an old prospect list with no history, buying fresh records is often simpler, because a cleanse can only shrink the list while new data starts current.
| Cleansing an existing list | Buying fresh records | |
|---|---|---|
| What you start with | Your existing list, whatever its age and source | New records selected to your exact criteria |
| Cost | Quoted by the cleansing service; depends on list size, checks chosen and how many records match | £1,225 + VAT for 5,000 records; £2,200 + VAT for 10,000 (whole order at one rate) |
| Records that fail | Removed or flagged, so the usable list shrinks | Not applicable: you choose how many records to buy |
| Contacts who have left | Flagged; replacing them depends on the service | Records are checked by telephone researchers every day |
| Phone screening | Re-screened against TPS and CTPS as part of the cleanse | Screened against TPS and CTPS before supply; re-screen after 28 days |
| History and relationships | Kept: notes, past orders and conversations stay on the record | None yet: new prospects with no history with you |
| Source and date | Only as good as your original records | Known source and supply date to record in your CRM |
| Best for | Customer lists and CRMs with valuable history | Prospect lists that are old, of unknown origin or badly out of date |
Fresh records are priced per 1,000 by the size of the whole order: 5,000 to 9,999 records at £245 per 1,000, and 10,000 to 19,999 at £220 per 1,000, with a minimum order of £300 + VAT. See the full price table and calculator in the complete guide.
Many businesses do both: they keep and cleanse their customer and enquiry records, and replace the cold prospect part of the database with fresh data once a year. Whichever you choose, deduplicate the new or cleaned file against your CRM and suppression list before the first campaign.
If you'd like to compare, ask us for a free count of current records matching your target sectors and areas. You'll get sample records and a written quote before you pay anything.
07
Data cleansing services: FAQs
What do data cleansing services do?
They check your records against up-to-date reference data and fix or flag what has changed: duplicates, address and postcode formatting, closed businesses, contacts who have left, invalid emails, TPS/CTPS registrations and opt-outs. Many also append missing fields such as phone numbers or named contacts.
What is the difference between a data cleanse service and data append?
A cleanse corrects or removes what is already in your records. Data append adds information that is missing, such as a phone number, website or job title. Many services offer both, priced separately.
What is data append?
Data append means filling gaps in your existing records by matching them to another data source, for example adding a named contact or phone number to a record that only has a company name and address. Added personal data must be handled under UK GDPR like any other personal data you hold.
How often should I cleanse my marketing database?
The law doesn't set a schedule. Deal with bounces, returned post and "no longer here" calls as they happen, re-screen phone numbers against the TPS and CTPS if 28 days have passed since the last check before calling, and review the whole database before any large campaign.
Is data cleansing a GDPR requirement?
UK GDPR doesn't mention cleansing by name, but Article 5(1)(d) requires personal data to be accurate and, where necessary, kept up to date, with reasonable steps taken to correct or erase inaccurate data without delay. (Article 5) Regular cleansing is one way to meet that.
Can I cleanse my B2B data myself?
Partly. In Excel you can remove duplicates, use TRIM and PROPER to tidy text, split names and standardise postcodes. Checking that businesses still trade, finding replacements for contacts who left and screening against the TPS and CTPS usually need outside data or a service.
Do I need a contract with a data cleansing service?
Yes, if they process personal data on your behalf. The ICO says that whenever a controller uses a processor there must be a written contract in place, and UK GDPR Article 28 sets out what it must include. (ICO)
Is it cheaper to cleanse an old list or buy a new one?
It depends on the list. Cleansing keeps valuable customer history; buying fresh avoids paying to check records that will be removed. For comparison, 5,000 fresh records from us cost £1,225 + VAT and 10,000 cost £2,200 + VAT. Get a written quote from a cleansing service and compare like for like.
Do you offer a data cleansing service?
No. We supply fresh B2B records selected to your criteria, checked by telephone researchers every day and screened against the TPS and CTPS before supply. You can ask for a free count, sample records and a written quote.
What happens to duplicates in my CRM when I cleanse?
A good cleanse or CRM merge tool combines duplicates into one record and keeps the notes and history from each, rather than deleting one. Decide which version's details win, for example the most recently updated.