Story & Founder POV ad angles

Barely a story, despite the family it opens: it is a position. One named person says something about the category that another competent person could disagree with, and that is the whole structure. What makes it belong here is the byline. The same sentence from a brand account is a slogan and from a person with a face is an opinion — the reader knows somebody can be held to the second one, and that accountability is the entire mechanism.

There is one test for whether you have a real POV and it is cheap to run: could a competent competitor put their name on this sentence and be perfectly comfortable? If yes, you have written a platitude. A position has a cost — it forfeits some part of the market on purpose. “Most brands should run eleven mediocre ads instead of one good one” loses everybody who sells craft, and losing them is exactly what makes the people who stay actually agree with you rather than merely read past.

Falls flat if the founder has nothing interesting or credible to say, says the taxonomy, and credible is the load-bearing half. An opinion is worth reading in proportion to what it cost the person holding it: they ran the campaigns, they lost the money, they shipped the thing and watched it fail. So attach the position to the experience that produced it in the same breath — one clause is enough. That is also why it works cold. A stranger can evaluate an opinion without knowing anything about your product, which almost nothing else in this library can claim.

The story of the beginning: who, when, and what they were doing instead. It works for the same reason a face works — attribution. A product from a company is a thing that appeared on the internet; a product from a person on a specific afternoon has a cause, and a cause is far easier to trust than an assertion. It is also the most memorable shape in the family, because people retain narratives at rates that plain claims never approach.

Keep it separate from Why We Built This, which sits directly below it and gets merged with it constantly — and merged is where both stop working. Origin is about the founding, and its job is warmth: this came from somewhere and somebody cared. Why We Built This is about the problem, and its job is product-market fit. The test is the last line. If the story ends “and that is how we started”, it is this angle; if it ends “which is why it works the way it does”, it is the next one, and it should have been written differently from the first sentence.

Self-indulgent if the customer is not central, says the taxonomy, and this is the most-hit weakness in the family for a structural reason: the origin is the story you most enjoy telling and the one the reader has the least stake in. The fix is a ratio you can count. Two sentences of before, one of turn, and then the reader appears — and every sentence in the first half that is not load-bearing comes out. Nobody has ever bought anything because of where a company started. Plenty of people have bought because of what starting there made it good at.

The origin story with the customer installed as the reason. It is the most useful shape in the family for a cold audience that already has the problem, because it does two jobs in one sentence: it names the pain, in the manner of Problem & Pain (issue #1), and it explains the product’s existence as the response to it. A feature list says what the thing does. This says why it does that and not something else, which is the question a sceptical reader is actually asking.

The mechanism is that it converts a design decision into evidence. Any claim about a product can be doubted; “we built it this way because that kept happening” is much harder to argue with, because the reader can check the second half against their own week. This is the angle where every strange thing about your product becomes an asset — the odd default, the step you removed, the option you refuse to offer. Those are the parts a competitor cannot copy without the story, and the story is what makes them read as judgement rather than as gaps.

Weak if the origin problem feels manufactured, says the taxonomy, and retrofitting is both easy and extremely visible. The tell is a problem stated at exactly the width of your feature set — a frustration nobody has ever complained about in those words. So source the sentence rather than writing it: the support ticket, the sales call, the thing somebody said on a demo before they caught themselves. Keep it in the customer’s vocabulary and not the roadmap’s. And if you cannot find a real one, that is information — the honest response is to run a different angle.

Frames the brand around something larger than the transaction. It is one of only two rows in the family marked Cold outright, and for a precise reason: a mission requires no product comprehension whatsoever. A reader can agree with what you are for several seconds before they have any idea what you sell, and the agreeing is what makes them read on.

What separates a mission from a value statement is a cost. A belief you have paid nothing for is free to hold, which is why every company has one and why none of them persuades anybody. The version that works names something you gave up: a customer segment you turn down, a feature you will not ship, a pricing model you refused because it would have made you behave differently. It is the Founder POV test applied to the company rather than to a person — if it forfeits nothing, it argues nothing.

Performative is the listed weakness, and the taxonomy is unusually specific about the trigger: “if product actions do not support the mission.” So write the mission out of the actions rather than the actions out of the mission. Look at what the product already refuses to do and what the price already implies, and say that. Readers are thoroughly inoculated against this angle — a mission ad that reads as marketing does not merely fail, it spends credibility you were going to need for the other sixteen rows on this page. If it came out of a positioning workshop rather than out of the way you already behave, do not run it at all.

Uses the low point as the engine. Structurally it is the only angle in the family with genuine narrative tension — a beginning that goes wrong before it goes right — and tension is what buys the second and third second of attention. Everything else on this page is told from a stable present, which is comfortable to write and much easier to scroll past.

It persuades rather than merely entertains because survival is evidence. A company that nearly died has been tested against something, which is a kind of proof no feature supplies and no competitor can claim on your behalf. It also inverts the reader’s default assumption: they know that anything in a paid placement is the polished version, so an ad that volunteers the worst part reads as standing outside the convention. That is a one-time credibility purchase per audience, which is why this is a rotation angle rather than an evergreen one.

Can feel overly dramatic, says the taxonomy, and there is a specific mistake underneath the drama — staging the crisis and never showing what changed. A near-death with no consequence is a story about you having a hard time, and the reader owes you nothing for it. Spend one beat on the low point and the rest on the decision that came out of it, and make sure that decision is visible in the product today. Keep the scale honest, too: “we almost quit” about a soft quarter is the kind of inflation a reader punishes retroactively, across everything else you have ever told them.

The mistake is the credential. It is worth separating carefully from the row above it, because they look adjacent and do different jobs: Why We Almost Quit is about the company nearly dying and trades on survival, while this one is about the founder being wrong and trades on the correction. The taxonomy’s phrase is exact — authority through lived experience rather than polished claims — and for a company with no logos to show, it is the cheapest authority on the market.

The structure is fixed and worth following literally: what I believed, what it cost, what I do now. The middle term is the one every draft skips and the only one doing real work. A lesson with no price attached is an opinion, which is angle 101 with extra words; an opinion with a receipt is a lesson. The cost does not have to be money — six weeks, a client, a launch that went out with the wrong thing on it — but it has to be specific, because specificity is the only available evidence that the experience happened at all.

Requires genuine insight; fake vulnerability is easy to detect. The standard tell is a failure that flatters — “I cared too much about quality”, “I over-delivered” — and readers decode a humblebrag faster than they decode almost anything else in advertising. Pick a mistake you would still rather not mention. The second tell is a lesson too general to have come from any particular experience: if it fits unchanged in a LinkedIn caption, it did not come from anywhere. And being caught here is not one wasted impression, it is a tax on the whole family, every row of which runs on the reader believing you.

Opens with something you would normally keep quiet. What separates it from the lesson above is tense and object: a lesson is about the past and safely resolved, while a confession is about now, and is often about your own product or category. “Most of what we sell is unnecessary for half the people who buy it” is a confession. It is the strongest pure hook mechanic in the family, because an admission opens an information gap the reader has to close.

The strongest version admits something against your own commercial interest, because that is the only kind a reader cannot explain away. Anything convenient gets reclassified as marketing inside a second, and the reclassification is automatic rather than considered. So the confession worth running is the one your own marketing would have edited out: a limitation of the product, a customer you are not right for, a fact about the category that makes your pricing look strange. Say it plainly, then let the rest of the ad be the reason to buy anyway.

Can become gimmicky if every ad is a confession, and the taxonomy has named the exact cause. The form is a rotation, not a voice. Three in a row and the audience has learned that “honestly” is a hook, at which point the mechanism is dead for you specifically and cannot be revived by writing a better one. The other failure is the bait: an admission that opens hard and then resolves into an ordinary product pitch. Readers remember being baited considerably longer than they remember what was being sold, so whatever you admit has to still be true by the end of the ad.

The customer is the protagonist and the founder is off-screen entirely. It is the only Mid row in this family, and the stage is a statement about what it needs rather than about what it does: a story about somebody who fixed the reader’s problem only works once the reader accepts that the problem is theirs and that fixing it is possible. On genuinely cold traffic it is a stranger’s anecdote, and it will lose to a callout every time.

Keep it distinct from Testimonial over in Proof & Trust (issue #4), because the two use the same raw material for opposite purposes. A testimonial is a verdict — it worked, here is what I got — and it is aimed at doubt. This is a journey, and the part that makes it one is the hesitation in the middle. That is the whole difference and the reason this converts where a testimonial stalls: the reader’s own objection gets spoken out loud by somebody who then went ahead anyway, which is an argument no first-person claim from you can make.

Requires a compelling, representative customer story — and representative is the harder of those two words. Your most spectacular result is usually your least useful one, because a reader who cannot see themselves in the protagonist is reading a lottery ticket. Choose the customer whose starting conditions look most like the audience’s, not the one whose numbers look best. And keep it recognisably one person’s account, in their own words and their own grammar. The moment it is tidied into brand voice it becomes a case study, and nobody in the history of the feed has read one of those.

Shows the company as a place with people and habits in it. There is no argument and no offer; the persuasion is familiarity, which is a real and well-documented effect rather than a soft one, and a face is processed by a reader on a different track from a logo. What it produces is not a decision but a slightly different starting position for every ad you run afterwards.

Its honest limitation is written into the taxonomy: may not be persuasive enough for direct response on its own. That makes it a supporting angle, and the mistake teams make is judging it as a closer and then killing it. It belongs in a sequence — the second impression on somebody a colder angle already caught — and its real return shows up in what a later, harder-selling ad achieves, because the same claim from a known entity is not the same claim. Read it on assisted conversions and frequency, and schedule it rather than benchmarking it.

For software there is a specific trap: “behind the brand” is a room with laptops in it, and nobody wants to see that. The material that works is the specific and slightly odd — what actually gets decided here, how, and by whom. A very small company has an advantage rather than a shortage. “The person who answers support wrote the thing you are emailing about” is a fact about operations that a competitor with two hundred employees literally cannot say. Show the process at the point where it is unusual, and skip the group photo entirely.

Turns how the thing is made into the argument. Visible effort is a quality signal and a price signal at once — watching something be built raises what people will pay for it — which is why the taxonomy files it under premium positioning rather than under education.

It also says “less relevant for invisible/digital products”, and that is the sentence that matters if you sell software, because there is nothing to film. The translation that works is to change the subject from the product to its output: not how the software was built, but how one ad got made — the concept, the style picked out of the gallery, the take that got kept and the ones that did not. That version has something to look at, and it doubles as a demonstration, which is the rare case of two angles reinforcing each other instead of blurring into one.

The other digital version is the decision rather than the craft: why the brand kit is a constraint on the model instead of a suggestion to it, why nothing can publish before a person has looked at it. Both versions share one failure, and the taxonomy calls it production-heavy. A making-of more polished than the thing being made reads as an ad about an ad, and the reader notices the extra layer. Keep it plain and let the roughness stay in — the persuasion is that the work is visible, and gloss is precisely the thing that hides it.

Moves the microphone from the founder to somebody who does the work. It buys two distinct things. The first is a fresh voice, which is the standard remedy when founder-led ads have fatigued an audience. The second is a different species of authority: the founder has an obvious stake in you believing them and an operator has a craft, and a reader discounts those two claims at very different rates.

“The speaker needs a clear reason to be trusted” is the entire angle compressed into a line. That reason is the work, and it has to be legible in the first three seconds — what this person does all day is the credential, not what their title says. A media buyer explaining how they judge a test is credible; “our Head of Growth” saying the product is great is a testimonial with a lanyard on it. The content has to match the framing, too: the operator ad is at its best teaching something slightly technical that a founder would never bother to mention.

Small companies read this row and skip it, and they should not. The operator does not have to be on your payroll. A customer who runs a hundred campaigns a month, a freelancer, an agency lead who opens the thing daily — the angle needs a practitioner, not an employment contract. And if the founder is also the operator, that is a framing choice rather than a disqualification: speak from the work instead of from the ownership, drop the word “we”, and the same person delivers this angle rather than Founder POV.

One day, in order, with the product appearing exactly where it actually appears. It is Cold outright — one of two in the family — because a routine needs no setup at all: the reader recognises the shape of a day before they have registered that they are being sold anything, which is about as low-friction an entry as advertising gets.

What it does that a demo cannot is show the product’s size. A demo shows what a tool is capable of; a day shows how much of a life it takes up, and for a great many purchases the unasked question is “how much of my week is this going to want” rather than “does it work”. Two minutes between the coffee and the first meeting is an answer no feature list can give. It is also the natural home for the small true details — the tab already open, the thing done one-handed on a phone — that make everything around them believable.

Can become lifestyle filler if the product role is weak, and there is a blunt test for that: would the day be different without it? If the product could be cut from the edit and the day would run the same, you have made a mood film. Two guards. Anchor it to one specific moment rather than a montage of pleasant ones, and let the day keep the parts that are still annoying — a day in which everything is easy is an advertisement for a life the reader does not have, while a day in which one thing got shorter is one they can picture having by Thursday.

Publishes the process while it is still running. It is the only angle here in a different tense: the rest tell a finished story, this one tells an unfinished one, and the unfinished part is what creates the investment. A reader who has watched you decide something has a small stake in what you decide next, and a stake is what brings somebody back to a second impression voluntarily.

That is what the Cold–Warm stage is describing — the only such row in the library. It works on strangers, and its actual payoff arrives on the third and fifth exposure, which makes it the family’s best answer to the frequency problem: the same account can run it repeatedly without the audience learning the shape, because the content is genuinely new each time. It is also the cheapest angle here to supply, since the material is whatever you were doing anyway. A decision, a number, a thing that broke.

Two failure modes, and the taxonomy names the first: it exposes unfinished work. Once a roadmap has appeared in an ad it is a promise with a date attached, and the reader who remembers it is the one you least want to disappoint — share decisions and outcomes, never intentions. The second is more common and less discussed: build-in-public is fascinating to other builders and frequently irrelevant to customers. If your buyer is not a founder, filter hard for the parts that touch what they receive. Nobody’s purchase has ever been decided by your refactor. And the form itself promises a next one, so a single build-in-public ad from a company that then goes quiet is worse than none at all.

A letter: second person, one signature, and none of the compression that makes an ad look like an ad. The form is itself the signal. Nobody writes a letter to a stranger, so being sent one implies a relationship already exists — which is exactly why it is Warm–Hot, and exactly why it is reliably wasted on cold traffic, where the reader has no relationship to be reminded of.

What it is uniquely good at is the message that would sound wrong compressed into ad copy: explaining a price change, asking somebody who drifted away to come back, saying thank you at a milestone, or making the case for a decision that cannot survive being a headline. Length is permitted here in a way it is nowhere else in the library, because the form sets the expectation before the first line does. There is one mechanical rule and it is absolute: from one person, to one person. A letter with a “we” in every sentence is a press release with a signature at the bottom.

Less scalable if the brand is not founder-led, says the taxonomy, and there is a second scaling limit worth knowing: an audience can only be sent this form a few times before it becomes a template, at which point it stops implying anything at all. Reserve it for moments that genuinely warrant it. On Meta specifically, mind the mechanics — a long letter needs a first line that survives truncation and a reader warm enough to tap “see more”, so run it to customers or a retargeting audience and let that opening sentence carry the entire thing.

The rarest angle in the library, and the only one whose subject is a live injury. Attention is enormous precisely because it violates what a reader believes paid placement is for — nobody buys distribution to say they were wrong — and that violation is the mechanism. It is also the only angle here that can make things materially worse, which is why it deserves more caution than anything else on this page.

It works when three things are true and fails if any one is missing: the audience already knows or is about to, the fix is finished rather than planned, and the apology is unconditional. “We’re sorry if” and “we’re sorry you felt” are the two constructions that convert an apology into an aggravation, and both survive into more drafts than anybody would like. The order that works is what happened, what it cost the customer, what has changed, and what you are doing for the people affected — and that last part is what makes it an apology rather than a statement.

Risky if the issue is serious or the apology feels tactical, and both halves need reading. Genuinely serious problems belong in email and support, where the affected people actually are; paid distribution is for the case where the story has already outrun your owned channels. Tactical is the subtler failure — an apology with an offer stapled to the bottom is a promotion, and readers price it as one instantly. The honest version is close to commercially worthless in the short term, and that is exactly why it buys what it buys. If you are reaching for it because the numbers were soft this week, don’t.

Takes the price objection head-on and answers it with the reason rather than the justification. It is the only place in the library where a founder’s voice is the mechanism of a price argument, and that is not decoration: the same defence from a company is a policy, and from a person it is an explanation. A policy can only be accepted or refused; an explanation can be evaluated, which is the difference between a reader arguing with you and a reader doing arithmetic.

It works because the buyer’s real question is never “why is this more” but “what is the money doing”, and most premium pricing has a perfectly legible answer that nobody bothers to say out loud. The Mid–Hot stage is exactly right and worth respecting: this ad is for somebody who has already decided they want the category and is now choosing inside it. Run it cold and you have introduced price before value, which is the oldest way to lose a sale in Offer & Economics (issue #7) and works no better here for being told in a nicer voice.

Must be backed by real differences, says the taxonomy, and the constraint is sharper than it looks: the difference has to be one the buyer would pay for, not one you happen to be proud of. A cost you incur that they do not value is a confession of inefficiency dressed as a virtue. There is also a comparison choice specific to this category that decides whether the ad works at all — the honest reference point is usually not the cheapest tool on the shelf but the freelancer or the agency, where the real price includes the four-week wait and the round of resizes. Compare with what the reader is doing now, not with the lowest number in the market.

The mirror of the row above, and the harder of the two to run well. The reader arrives with a default explanation for a low price already loaded — it must be worse — and unless the ad replaces that explanation with a better one, your own price becomes the strongest piece of evidence against you. This is the one angle in the library where saying nothing would have been safer than saying it badly.

The word carrying the whole thing is structural. Cheaper because we are generous is a discount, and a discount implies the price was wrong before and might be wrong again. Cheaper because of how the business is built is a fact the reader can check against what they already know about the alternative: no account managers, no per-project pricing, no billable hours between having the idea and having the ad. The more mechanical you make it the stronger it gets, and the more it flatters you the weaker — name the cost your competitor carries and you don’t, then let the reader finish the arithmetic.

Can accidentally lower quality perception, and there are two guards. Never use the word cheap: the sentence is about a structure, and the price is its consequence rather than its subject. And pair the claim with a quality proof inside the same ad, because the reader is running one silent question the entire time and leaving it unanswered is how a good price argument still loses. Then decide whether this is your angle at all before committing to it — a brand that argues on price gets chosen on price, and there is always somebody prepared to go lower next quarter.