AI use case · Is your data ready for personalisation?Personalisation needs to know who it’s talking to. Your systems disagree.
Recommendation and next-best-offer tools promise the right message for every customer. They work only when the shopper on the site, the subscriber in email and the customer in finance resolve to one person, with one consent status. Otherwise the model personalises for fragments of people who don’t exist.
Starts with a 30-minute call. Commits you to nothing.
What it needs
What personalisation needs from your data.
Each requirement, the systems it lives in, and what we check. The AI is the easy part; this list is the hard part. Each check feeds the Truth Score™, a score out of 100 for how far your data can be trusted.
Truth Score pillars that gate it mostConnectivityConsistencyGovernanceThe six pillars
Where it breaks
Every system looks fine. The gaps between them don’t.
Where personalisation breaks: not inside one system, but where two of them disagree about the same customer, deal or invoice.
-
Web ↔ Marketing
The same shopper as several marketing profiles
What it does to the AIEach profile carries part of the history, and gets a different offer.
-
Marketing ↔ CRM
Consent withdrawn in one system, still active in another
What it does to the AIPersonalisation becomes a compliance problem.
-
Finance ↔ Web
Returns and refunds recorded in finance, not in the store
What it does to the AIThe model recommends more of what customers sent back.
Gaps to test for this use case, not client findings.
Before you sign
Questions to ask the vendor selling it.
Four questions that separate a demo from a deployment. Bring the answers, or the proposal, to the Truth Score Read and we test them against your records.
- Q1
How do you decide two profiles are the same person?
Exact email alone misses many duplicates; loose matching merges strangers. Ask what they match on and how they test it.
- Q2
When our systems disagree about consent, which one wins?
The safe answer is the most recent opt-out anywhere. Make sure the tool can see all of them.
- Q3
What do you do with guest checkouts?
If they are dropped, a large share of real behaviour never reaches the model.
- Q4
How will we measure lift, and against what control group?
Without a hold-out group, any revenue rise will be credited to the tool.
How a Truth Score Read tests it
Personalisation, tested against your own records.
Our engine matches the customers, people, deals and invoices across the exports from the systems behind the decision. Every red flag is checked by hand against sample records before sign-off, and the verdict comes in writing. Here is what that means for this use case.
- 01
Resolve the customer
Store, email and CRM exports matched by email, phone and customer ID, so you see how many profiles each real person has.
- 02
Check consent
Subscription and consent status compared person by person across every export.
- 03
Test the catalogue and history
Product data quality, and whether orders, returns and behaviour attach to a resolved customer.
- 04
Give the verdict
Deploy now, Needs work or Blocked, with identity and consent gaps counted.
Two systems are enough to start. We send the exact field list after the call.
How the Truth Score Read worksWhat you get back
See what comes back: a real sample memo.
One fictional B2B software company, Example Co: 72,837 records in 7 exports from its CRM, finance and HR systems, run through the same engine and the same checks as a paid Truth Score Read. Each system passes its own checks. Matched across systems, only 13% of deals tie to an invoice.
Questions
Personalisation, answered plainly.
Anything else, ask on the 30-minute call. It commits you to nothing.
Book the callWhat does personalisation need from my data?
One identity per customer across the systems that touch them, purchase history attached to that identity, consent that agrees everywhere, and a clean product catalogue. Identity is the foundation; everything else hangs off it.
Why don’t our store, email and CRM records join?
Each system creates its own records on its own terms: guest checkout, a second email address, a phone number in one place only. Without a shared customer ID, matching falls back to fuzzy rules, and that is where duplicates and false merges both appear.
Is consent really a data quality issue?
Yes. If someone has opted out in one system and not in another, a personalisation engine will message them. That is a data problem with a regulatory consequence, and it shows up clearly when exports are compared.
We already have a customer data platform. Isn’t that solved?
A customer data platform is only as good as the matching rules and sources behind it. We test its output against the source systems’ own exports, which is where the merges it missed, or got wrong, become visible.
Is your data ready for personalisation? Find out before you sign.
Exports from two or more systems in. A signed verdict on this use case within 5 working days of complete exports. Starts with a 30-minute call. Commits you to nothing.
Starts with a 30-minute call. Commits you to nothing.