Personalisation is often sold as a content problem.
Use the customer’s name. Mention the product. Adjust the message. Recommend the next action. Trigger the journey at the right moment.
Those things matter.
But personalisation fails when the data behind it cannot be trusted.
If customer attributes are stale, duplicated, incomplete or disconnected from current behaviour, personalisation becomes decoration. It may look sophisticated, but it does not make the customer experience more reliable.
The mistake is starting with the message
Many teams begin personalisation discussions by asking what the message should say.
That is the wrong starting point.
The better question is: what do we reliably know about this person, right now, and are we allowed to use it for this communication?
That question forces the business to confront the foundation.
Which record represents the person? Which product context is current? Which quote or interaction should be used? Which preference state applies? Which attributes are fresh enough to trust? Which fields are used for segmentation and which are only operational noise?
Without those answers, personalisation becomes guesswork with a better subject line.
Bad personalisation creates distrust
Poor personalisation is worse than no personalisation.
A generic message may be forgettable. A wrongly personalised message damages trust.
The customer receives a message about the wrong product. A follow-up refers to an outdated interaction. A preference is ignored. A journey treats a previous quote as current. A supposedly tailored message makes it obvious that the business does not have a coherent view of the customer.
The issue is not the content block. It is the data model.
Marketing Cloud can only personalise from the data it receives and the rules it has been given. If that data is fragmented, the message will inherit the fragmentation.
Identity comes before personalisation
Personalisation depends on identity.
A person can appear in multiple places. They may request several quotes, interact with more than one product, subscribe to a newsletter, opt in again after a previous suppression state, or use different contact details across channels.
Each record may be valid in its own context, but that does not mean each record is the right basis for personalisation.
If the environment cannot distinguish the person from the transaction, product interaction, subscription or communication event, personalised journeys become fragile.
The business needs to know which data should drive which decision.
A quote-level record may be useful for a specific follow-up. A person-level view may be better for preference management. A product interaction may be relevant for timing. A channel preference may determine whether the message should be sent at all.
Personalisation improves when those layers are clear.
Current context matters more than clever copy
A message can be well written and still be poorly personalised.
The real question is whether the context is current.
Does the customer still qualify for the journey? Is the product interaction still relevant? Has the person converted elsewhere? Has a newer record replaced the old one? Has the customer opted out of the channel? Has the audience been refreshed since the last interaction?
If the answer is unclear, the message should not pretend to be intelligent.
Personalisation is not only about inserting variables. It is about using the right variables, from the right source, at the right time, for the right purpose.
Consent shapes what personalisation should do
Personalisation and consent cannot be separated.
A business may know something about a customer, but that does not automatically mean every channel or message should use it.
Email, SMS and WhatsApp preferences can behave differently. A customer may opt out of one channel but remain eligible for another. A later opt-in may change the position. A blocked-contact signal may require suppression. A newsletter preference may not be the same as permission for a journey follow-up.
Personalisation needs a clear consent model because the message must respect both relevance and eligibility.
The question is not only “can we personalise this?” The question is “should we use this data for this communication through this channel?”
Personalisation without reporting is theatre
If personalisation matters, the business should be able to measure whether it changed anything.
Too often, personalised journeys are launched without a clear tracking model. The team can see that messages were sent, opened or clicked, but cannot confidently connect the personalised experience to a meaningful business outcome.
That creates a dangerous gap.
The business invests in personalisation but cannot tell whether the operating model improved. The campaign looks more advanced, but decision-making remains weak.
Good personalisation design should include the reporting question from the start. What are we trying to improve? Conversion? Follow-up completion? Product engagement? Channel response? Customer retention? Sales handoff?
If the outcome is not clear, the personalisation may be decorative.
Data freshness is an operating discipline
Personalisation depends on freshness.
A customer attribute that was accurate six months ago may be misleading today. A product interest may no longer apply. A quote may have expired. A customer may have already converted. A preference may have changed.
Marketing Cloud teams need to know how data is refreshed, which sources are current and which fields should no longer drive decisions.
This is not a purely technical issue. It affects customer trust and business performance.
If the environment keeps using stale values because they are easy to access, personalisation becomes a liability.
What better personalisation design looks like
Better personalisation starts with data discipline. That means:
- A clear view of which record represents the person
- A defined relationship between person, product, quote and communication event
- Current channel preference and suppression logic
- Fresh customer attributes from reliable sources
- Journey eligibility that reflects current context
- Tracking that connects the message to a business outcome
- Monitoring that shows when the audience or source data stops behaving normally
Once those foundations are in place, personalisation becomes more than a content tactic. It becomes part of a reliable customer engagement model.
The questions every Marketing Cloud team should ask
Before adding more personalisation, ask:
- Which customer attributes do we actually trust?
- How fresh is the data used in this journey?
- Which record represents the person and which represents the transaction?
- What consent or preference state applies to this channel?
- Could this message be wrong if the customer has a newer interaction?
- How will we measure whether the personalised experience improved the outcome?
- Who owns the data quality behind the personalisation?
If those questions are difficult to answer, the personalisation problem is really a data foundation problem.
Business takeaway
Better personalisation starts with better data discipline.
Personalisation is not only about what a message says. It depends on identity, freshness, consent, product context, reporting and operational control.
How Cloud Genii helps
Cloud Genii helps organisations stabilise and improve Salesforce Marketing Cloud by strengthening the foundations behind customer engagement.
That includes audience structure, identity logic, consent and suppression handling, journey design, reporting visibility, monitoring and the operating model needed to make personalisation trustworthy.