PARTLY

As of 13 August 2026, AI can only partly clean your email list for a UK marketing campaign.

This still needs a person who signs their name to it.

Can you do it?

15 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsClay is listed as an AI-enriched prospect-list and personalised-outreach tool, but the supplied product information gives no price.

If this goes wrong, you may contact someone who opted out or damage the campaign's deliverability and compliance position.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice

  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Export the list from your CRM or email platform with the email address, unsubscribe status, bounce status, complaint status, consent or source fields, and any suppression fields included.
    2. Open the export in a spreadsheet, remove unrelated columns, and save an untouched copy of the original before sharing a suitably minimised copy with the chatbot.
    3. Paste the prompt and upload the list, then ask the chatbot to process the file without guessing missing consent or changing email addresses.
    4. Download the cleaned send list, suppression list, review list and change log, then compare their row counts and categories with the chatbot's summary.
    5. Check every suppression and review record against the current suppression and unsubscribe records in your email platform, resolving conflicts in the platform rather than by accepting the model's choice.
    6. Inspect a sample of retained records against the original export, including spelling, email addresses and subscription status, and reject the output if any source value has been invented or changed.
    7. Import only the checked send list into your email platform, keeping the platform's suppression controls enabled before scheduling the campaign.

    Prompt

    I am preparing a UK marketing campaign and need to clean an email-list export. I will provide the data as a CSV or pasted table, with these columns where available: [column names]. Do not send emails, contact anyone, enrich the list, guess missing consent, or decide that a record is legally usable. Preserve the original data and create a cleaned version with these rules: remove exact duplicate email addresses while retaining the most complete record; flag malformed or blank email addresses; separate unsubscribed, opted-out, bounced, complained-about and otherwise suppressed records into a do-not-send file; retain the original values in separate columns; normalise email addresses only by removing accidental surrounding spaces and comparing case-insensitively; do not alter the address itself; flag records with conflicting subscription or suppression statuses instead of choosing one. Return: (1) a cleaned send list, (2) a suppression list, (3) a review list, (4) a row-by-row change log with the reason for every removal or flag, and (5) a summary of input rows, output rows and each category. Never invent a status, consent record, person, company or email address. If the file is too large to process reliably, say so and give me a safe batch plan. I will compare your result with the source export and the records in my email platform before using it.

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

What caps this at PARTLY: consent and privacy, verification cost and legal accountability.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs1
Verification1
Liability1
Effort delta2
Total7 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT remove duplicates from my email list?
Yes. It can identify duplicate addresses, malformed entries and blank values in a supplied file, then produce a cleaned version and a change log. You still need to compare the result with your CRM or email platform before sending.
Can AI check whether my email list is GDPR compliant?
No. AI can organise consent, unsubscribe and suppression fields that already exist, but it cannot prove the history or legal basis behind each record. Your business remains responsible for checking the campaign under the rules that apply to it.
Can AI remove unsubscribed contacts from my mailing list?
It can separate records marked as unsubscribed or suppressed if those fields are present in the export. It cannot find an unsubscribe that is missing from the file, so check the result against your email platform's current suppression list.
Is it safe to upload my customer email list to AI?
Treat the list as personal data and check your organisation's data-protection and supplier rules before uploading it. Minimise the fields, use an approved service, and do not upload more information than the cleaning task needs.

Nearby answers

All Email & CRM campaigns rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.