Relevance Beats Personalization: Why Talking About Their Dog Won't Close Deals

Personalisation proves you looked. Relevance proves you understood. Sales teams have spent a decade getting very good at the first and it has not moved reply rates, because a message that mentions a prospect's marathon and then pitches something unrelated to their work is not personal, it is surveillance with a pitch attached. The test is not whether the detail is about them. It is whether the detail earns the sentence that follows it.

The Distinction, Precisely

These two words get used interchangeably and mean almost opposite things in practice.

Personalisation is the insertion of information about the recipient. Their name, their company, their recent post, their dog. It demonstrates effort and it is easy to automate, which is why it is everywhere.

Relevance is the demonstration that you understand something true about their situation, and that what you are proposing relates to it. It is harder to produce and it is the only one of the two that gives a recipient a reason to reply.

A message can be highly personalised and completely irrelevant, and this combination is now extremely common. It is also the worst-performing category, worse than a plainly generic email, because it burns credibility: the recipient can see that effort was spent and that none of it went into understanding their problem.

Why the Dog Line Fails

Consider the pattern that has become the default in tooling-heavy outbound. The tool finds something about the prospect from a social profile. The email opens with it. The second paragraph pitches.

From the recipient's side, three things happen in sequence. They register that this person has looked at their personal profile. They wait for the connection between that and the message. No connection arrives. What is left is an impression that somebody ran them through a process.

The specific failure is a non-sequitur. "I saw you ran a marathon last month, and I wanted to talk about your email deliverability" contains two facts and no relationship between them. Human conversation does not work like this, and the mismatch is what registers, before any judgement about the offer.

There is a second cost that is easy to miss. Referencing someone's personal life in a commercial cold email slightly changes who you appear to be. It is not a catastrophe, but it makes the sender seem willing to use anything available, which is precisely the wrong first impression when you are about to ask for trust.

The One-Sentence Test

A practical filter for any opening line:

Can you write "which means" after it, and have the next clause be both true and relevant to your offer?

"I saw you ran a marathon last month, which means..." goes nowhere. Nothing follows that connects to a business problem.

"You have had an SDR role open since June, which means someone is carrying that quota on top of their own job." That follows. It is an inference the recipient will recognise, it names a real pressure, and the pitch now has somewhere to land.

If the "which means" clause requires a stretch, the detail is decoration. Cut it and find one that implies something. The tiers of data that pass this test are covered in which data points are worth personalising on.

Where Relevance Actually Comes From

Three sources, in descending order of quality:

• Their own words about the problem. A job description, a post, a conference talk, a public job posting. When someone has described the difficulty themselves, you can be relevant by taking them seriously rather than by guessing. Highest credibility available, and the hardest to fake.

• A structural inference from their situation. Company shape, function, stage. A team that doubled in a year has an onboarding problem whether or not anyone has said so, because that is what doubling does. Reliable and available at scale.

• A pattern from comparable companies. "Most teams at this stage hit this" is weaker because it is about a category rather than them, and it is still relevance rather than decoration, because it is about their work.

All three are about the professional situation. None require knowing anything personal, which is the point: relevance and privacy are not in tension. The information that makes a cold email work is all public professional context.

Relevance Scales Better Than Personalisation

Counterintuitive, and it is the practical reason to prefer it.

Deep personalisation is inherently per-prospect. Reading someone's posts to find a detail is work that cannot be reused, so quality falls as volume rises, and teams end up with tooling that produces the appearance of research without the substance.

Relevance is largely per-segment. If you understand that heads of operations at companies that recently doubled have a specific problem, that understanding applies to every company in that segment. You write it once, well, and it is genuinely relevant to all of them. The research investment is in understanding the segment, and it amortises.

Which inverts the usual advice. The scalable path is not better personalisation tooling; it is narrower segments with a real insight behind each one. Five segments with a genuine understanding of each outperform five thousand individually personalised emails built on details that imply nothing.

What This Looks Like in the Copy

The structural change is small and the effect is not. Relevance goes in the first line and does the work of justifying the email's existence; everything after it is about the problem, not about them.

A useful discipline: after the opening line, your email should not mention the recipient's company again except where necessary. If their name appears four times, you are still performing research rather than making an argument. One reference, then the substance.

This also means the relevance line should not be flattering. Praise is a form of decoration and it reads as a preamble to an ask. State the observation flatly and move to the point. A neutral, accurate observation about their situation is more persuasive than a compliment, because it signals understanding rather than effort.

The space available for all this is limited, which is itself useful: how long a cold email should be and where the ask goes.

The Uncomfortable Implication

If relevance is the requirement, then a segment you do not understand cannot be written to well, and no tooling fixes that.

This is worth sitting with, because it means the constraint on outbound quality is domain knowledge rather than software. A team that deeply understands two buyer types and has nothing to say about a third should probably not be emailing the third, however well it fits the firmographic filters.

The honest version of scaling outbound is therefore to expand the number of segments you genuinely understand, one at a time, rather than to expand the list. That is slower than buying more contacts and it is the only version that holds up.

FAQ

Is personalisation useless then? Not useless, just insufficient and frequently misapplied. Personalisation that implies a problem is relevance. Personalisation that only proves you looked is decoration, and decoration performs worse than nothing because it shows where the effort went.

Should I ever mention something personal in a cold email? Rarely, and only when it is genuinely professional context, such as a talk they gave or something they published. Personal life details from social profiles read as intrusive and connect to nothing.

How do I make an email relevant at scale? Segment more narrowly and understand each segment properly. Relevance is reusable within a segment, which is what makes it scalable in a way per-prospect personalisation is not.

What is the fastest test of whether my opening line works? Add "which means" to the end of it. If the following clause is both true and connected to your offer, the line earns its place. If it requires a stretch, the detail is decoration.

Does complimenting the prospect help? Generally no. Praise reads as a preamble to an ask. A flat, accurate observation about their situation signals understanding, which is what you actually want to convey.

What if I have no information about a prospect at all? Then write to the segment honestly and well. A clear, relevant, non-specific email outperforms a fabricated specific every time, and inventing a detail is the one option that is worse than both.

Want outreach where the first line earns the second, on segments someone actually understands? That is the point of human review in Lidgen's model. Book a demo.