Study first
Review the ideas behind the questions
Lifecycle personalization is useful only when the data use is fair, clear, and limited. Before a win-back or next-best-action campaign launches, check what customers were told, what they objected to, and whether an automated score needs extra review.
Explain The Profile Before Using It
Personalization often uses purchase, behavior, interest, or predicted-risk data. Customers should not be surprised by how that data shapes the direct marketing they receive.
- People must be told clearly when their information will be collected and used for direct marketing.
- Profiling for direct marketing should be fair, transparent, accurate, and not excessive.
- If a new use is hard to explain or would make people object, the activity should be reconsidered.
In Practice
Name The Data Use
A lifecycle notice should say if purchase history, behavior, or other profile data will shape marketing messages.
Do Not Hide Third-Party Enrichment
If the team adds outside data to a customer profile, the customer-facing explanation needs to cover that use.
Common mistakes
Calling a profile score internal because customers do not see the score.
If the score shapes direct marketing, review whether people were told about the profiling and whether the data is fair and limited.
Q&A
Can purchase history shape lifecycle messages?
Yes, but the customer should be told clearly that purchase history will be used for that marketing purpose.
What makes a profile use risky before launch?
A use is risky when the team cannot explain it plainly, expects customers would object, or cannot show the data is fair and not excessive.
Respect Preferences Across Profile Use
A lifecycle preference is not only about one send button. If the customer objects to direct marketing, the team must stop using their information for that purpose, including profiling that supports it.
- People have an absolute right to object to or opt out of direct marketing at any time.
- A direct-marketing objection covers profiling related to direct marketing.
- Suppression lists help prevent future direct marketing to people who have objected.
In Practice
Stop The Profile Use Too
If the churn score exists only to target win-back marketing, a direct-marketing objection should stop that marketing use of the score.
Keep The Small Suppression Record
Deleting every trace can make accidental re-contact more likely. Keep only the limited details needed to honor the objection.
Common mistakes
Removing a customer from sends but leaving them in a win-back profile model that is used only for marketing.
Stop using their information for direct marketing purposes, including profiling for that purpose.
Q&A
Does an opt-out always mean delete every record?
No. The team may need a small suppression record so the person is not marketed to again by mistake.
Can the team ask an objected customer later if they changed their mind?
No. ICO guidance says they cannot be contacted later to ask if they have changed their mind about direct marketing.
Limit Data And Review Automated Decisions
A richer lifecycle profile can create new risks. Review whether the data is necessary, whether sensitive or inferred data is involved, and whether an automated decision needs human review or a way for the customer to challenge it.
- Direct-marketing planning should consider what information is necessary, proportionate, accurate, and not kept longer than needed.
- Profiling that uses special category data for direct marketing is likely to need explicit consent.
- Solely automated decisions with legal or similarly significant effects need extra safeguards.
In Practice
Do Not Collect For Maybe Later
A win-back idea is not a reason to keep every behavior, inferred interest, or old score forever.
Check Significant Automated Effects
A next-best-offer score can need extra review when it meaningfully affects what a customer can access or pay.
Common mistakes
Treating every profile score as low-risk because it is used by marketing.
Check the data used, the effect on the customer, and whether a solely automated significant decision needs extra safeguards.
Q&A
When should old lifecycle data be removed or anonymised?
When it is no longer needed for the direct-marketing purpose and cannot be justified.
What should a human-review route cover?
It should let someone ask for reconsideration or challenge a solely automated significant decision.