Service
Personalised direct mail, written not printed
“Dear [FIRST_NAME]” is not personalisation; it is a mail merge that everyone can see. Personalisation is a sentence in the middle of the letter that could only have been written to one person — and it has to be written in the same hand as the sentence around it, or the effect inverts.
Personalised direct mail uses data held about each recipient to change the content of their individual mail piece. When the piece is handwritten, the merged fields are written by the same pen in the same pass, so there is no visible difference between the personalised words and the rest of the message.
Fields worth merging
The test is not “do we have this field?” It is “would the reader be surprised that we knew this?”
- Their street or suburb — the single most effective field in property and local services. “Two houses on Bayview Terrace sold this month” is a different letter to “houses in your area are selling”.
- A specific past transaction — what they bought, when, and from whom. Naming it proves the letter is not a template.
- A date that matters to them — renewal, expiry, service due, anniversary. It gives the letter a reason to exist today rather than any day.
- A person’s name on your side — the adviser, agent or account manager they actually dealt with.
Fields not worth merging
Their title. Their industry, if you inferred it. Anything sourced from a data broker that they never gave you. Getting these wrong is worse than not personalising at all, and getting them right is faintly unsettling.
Write the sentence, not the slot
The commonest failure in variable-data copy is writing a sentence that only works for the median row of the spreadsheet. Then row 4,182 has a business name in the first-name column, row 900 has “NSW” where the suburb should be, and four hundred letters go out sounding wrong.
We proof merged copy against the extremes in your file, not the middle: the longest value, the shortest, the empty one, the one in capitals, the one with an ampersand in it. Then we show you what the worst-case letters actually say.
Your data, handled properly
Used only for your campaign
Your list is used to write your mail and nothing else. It is never merged with another client’s data, never enriched, never sold.
Returned or destroyed
On request at the end of a campaign, with confirmation in writing. Otherwise held only as long as we need it to reprint or reconcile the run.
No certification claimed
We do not hold ISO 27001 and we are not going to imply that we do. Exactly what we do and do not claim.
Questions about personalisation
How many fields can be personalised in one letter?
Technically as many as your data supports. Practically, two or three is the limit before the letter starts to read like a mail merge showing off. The best campaigns personalise one thing that genuinely could not have been in anyone else’s letter.
What happens when a field is missing or wrong?
Every merge field needs a fallback, and we will not run a campaign without them. A missing first name becomes a generic greeting rather than a blank; a missing suburb means the sentence containing it is swapped for one that does not. We report the fallback rate before writing begins so you can decide whether the data is good enough.
Is this different from variable data printing?
Variable data printing changes what is printed on each piece. This changes what is written on each piece — the merged text is produced by the same pen, in the same hand, at the same time as the rest of the sentence, so there is no visual seam.
What does personalisation cost?
$0.12 per piece for merged variable fields. The bigger cost is usually getting the data clean enough to be worth merging.
Keep reading
Handwritten direct mail
A direct-mail channel that does not look like direct mail. Handwritten letters and envelopes …
Bulk handwritten letters
What changes when a handwritten campaign goes from 100 letters to 20,000: data, proofing, con…
Personalising at scale
Which fields are worth merging, how to write a sentence that survives every row of your data,…
Pricing
Work out an indicative per-piece and total cost for a handwritten campaign. Change quantity, …
Automotive
Service reminders, lapsed-customer reactivation, trade-in approaches and delivery thank-yous,…
Send us a sample of your data
Twenty rows with the identifying details removed is enough. We will tell you which fields are worth merging, which are risky, and what the fallback rate is likely to be.