Cover Letters When There Is No Job Posting

There’s a category of application most people never send: the one where nothing was advertised.

You’ve noticed a company doing work you want to do. Or you know a firm your skills obviously fit. Or someone mentioned that a team is stretched. There’s no posting, no form, no process — just an organisation and an idea.

Speculative applications feel presumptuous, which is why most people don’t send them. That’s also exactly why they can work: you’re not one of two hundred responses to a posting, you’re one of very few messages that week, and you arrive without competition attached.

The form is different from a standard cover letter, though, and sending a standard one is how these fail.

Why the normal letter doesn’t work here

A conventional cover letter answers why me for this role. Speculatively, there is no role, so that question is unanswerable and the letter has nothing to attach itself to.

What a speculative letter has to do instead is propose the role, at least in outline. You’re not asking to be considered for something; you’re suggesting something exists that they haven’t advertised.

That means you have to say what you think they need. Which requires actually thinking about them for twenty minutes — and that’s the real cost of this approach, and why it doesn’t scale, and why it works.

The structure

1. Why them, specifically. One or two sentences, concrete. Not “I’ve long admired your work.” Something you actually noticed: what they build, a change they’ve made, the market they’re in.

2. What you think the need is. Your hypothesis, stated with appropriate humility. “It looks like X, which usually means Y becomes a problem around this size.”

3. What you’d bring to it. The one relevant thing, with evidence.

4. A small, easy ask. Not a job. A twenty-minute conversation.

Shorter than a normal cover letter — 200 to 250 words. It’s an unsolicited email; length is presumption.

A complete example

Hypothetical, to the head of operations at a mid-sized specialist manufacturer.

Dear Mr Halvorsen,

I saw that Brightmoor opened the second site in Leeds this spring. Running two production sites usually breaks whatever reporting worked for one — the numbers stop being comparable, and month-end starts taking twice as long as anyone budgeted for.

I spent the last three years as the operations analyst at Northwind through exactly that transition, from one site to three. Most of the work wasn’t technical; it was getting the sites to agree on shared definitions before any consolidated reporting could mean anything. By the end we had one weekly view all three sites accepted and a month-end close of one day rather than four.

I don’t know whether this is a problem you have or one you’ve already solved, and I appreciate you may not be hiring at all. If it is on the list, I’d value twenty minutes to hear how you’re approaching it — and if it isn’t, no reply needed and apologies for the intrusion.

Best regards, Sam Okafor

215 words. It’s specific about the company, offers a genuine hypothesis about their situation, backs it with directly relevant experience, and makes an ask small enough to say yes to.

The last paragraph is doing important work. Acknowledging that you might be wrong and that they might not be hiring removes the pressure — and paradoxically makes a reply more likely, because it’s no longer a request that has to be handled.

Getting the hypothesis right

This is where speculative letters live or die.

A good hypothesis is specific, plausible, and derived from something observable. Sources worth twenty minutes:

Recent changes. A new site, a new market, a funding round, a product launch, an acquisition. Growth creates predictable problems and you can name them.
Their own job postings — for other roles. If they’re hiring five engineers and no one to support them, that’s a signal. If every posting mentions the same tooling, that tells you their stack.
The shape of the company. A 40-person firm has different problems from a 400-person one, and the transitions between sizes are well-worn.
Public writing. Their engineering blog, an interview with a founder, a conference talk.

A bad hypothesis is generic (“I imagine you’re looking to improve efficiency”) or presumptuous (“your website has several problems I could fix”). The second is a genuinely common mistake — leading with criticism of someone’s work is rarely the opening you want, however accurate.

Hold it loosely. “It looks like” and “I don’t know whether this is a problem you have” cost you nothing and make being wrong survivable.

Who to send it to

The person who’d manage the role, if you can identify them. Head of Operations, Engineering Manager, Head of Content. They feel the problem and often have influence over whether a role gets created.

A founder or director at a small company, where they make these decisions directly.

Not a general info@ address, and generally not HR — HR fills defined roles, and there isn’t one yet.

Finding the name usually takes a few minutes on a professional network or the company’s own site. Guessing an email address from an obvious pattern is normal practice; if it bounces, it bounces.

What response to expect

Most speculative letters get no reply. That’s the baseline and it isn’t a judgement — you’ve arrived unannounced in a busy person’s inbox.

But the ones that land tend to land well, because you’ve reached someone with a problem at a moment they were thinking about it, without a competitive process. Outcomes people report from this approach: a conversation that leads to a role being created, a referral to someone else who is hiring, or a “not now, but let’s talk in the autumn” that turns into something six months later.

Given the hit rate, treat these as a supplement — a few well-researched ones a month alongside your normal applications, not a replacement for them. Twenty minutes of research each is the cost, and it’s the part you can’t skip; a speculative letter without a specific hypothesis is just a nuisance email.

One follow-up, then stop

If there’s no reply, one short message after ten working days:

Hi Mr Halvorsen — following up briefly on the note below in case it got buried. Entirely understand if there’s nothing there; I won’t chase again either way.

Best, Sam

Then stop. Unsolicited outreach earns you one follow-up, not a campaign, and saying you won’t chase again is both courteous and true.

The variant worth knowing

The same form works for a referral request to someone in your network, with the ask changed:

…I saw Brightmoor has an Operations Analyst role open and it’s very close to what I did at Northwind. You’d know better than me whether it’s worth applying — and if you think it is, would you be willing to pass my name to whoever’s hiring? Completely fine if not.

Shorter, warmer, and much higher conversion, because the recipient already knows you. If you have people you could ask, ask them before you spend an evening on speculative letters to strangers.