Creative brief template for B2B SaaS

The creative brief with three columns a subscription business needs: which slice of the ideal customer profile the asset speaks to, which objection it answers, and what proof it carries that survives being forwarded to somebody who never saw the ad.

Everything else is the general creative brief, unchanged. One row per asset, a hook written as what happens, one message with a proof beside it, the specs up front and an honest must-avoid.

Write it after the campaign brief has settled the segment and the offer, and before anybody designs a report cover. Most creative in this category is a document rather than a film, and documents are briefed badly because everybody assumes they know what a report looks like.

Use it again when the product changes. An asset that demonstrates a screen that no longer exists is worse than no asset.

The defining fact about creative in this category is that the person who sees the ad is often not the person who decides. A champion reads it, believes it, and forwards it to a budget holder who has no context and a sceptic who has heard three similar claims this quarter.

That trip is where most assets die, and it is why the forwardable proof column exists. Something in the asset, or in what it leads to, has to work for a reader with no relationship to you: a number with a source attached, a document that stands on its own, a calculator that uses the reader's own figures, a demonstration of the thing working.

What does not survive the forward is anything that depends on trust already being present. Adjectives, category language, claims about being leading or trusted, and testimonials from companies the new reader has not heard of all evaporate in transit. Briefing the proof explicitly is what keeps an asset useful after it leaves the feed.

The ICP segment column changes the language and the stakes. The objection answered column changes the hook, and it is the more useful of the two to write down.

There are only a handful of reasons a qualified buyer has not acted. It costs too much. It is risky to change. It is too much effort to implement. It is not the right time. Something is already in place. Each of those produces a completely different first line, a different proof and a different call to action, and an asset that tries to cover them all produces a generic opening that addresses nobody.

So one objection per asset, named in the row. A campaign that needs to answer four objections needs four assets, which is a more honest plan than one asset that gestures at all of them.

Most creative here is not film. It is a report, a benchmark, a checklist, a calculator, a one-page comparison — and those get briefed carelessly because everybody assumes they know what a report looks like.

They should get rows like anything else. The cover is a hook. The first page is a message. The length is a specification, and it is a real one: a forty-page document nobody finishes is worse than a six-page one that gets forwarded. The promise made on the landing page is a claim that the document has to keep.

Write those into the row and the output changes. Briefed properly, this category's documents become its best assets, because a genuinely useful document is the rarest thing in a buyer's inbox.

A large share of assets here show the product: a screen, a workflow, a result. Those assets have a hidden expiry date, because the product changes.

Write into the row exactly what is being demonstrated — which screen, which workflow, which version — so that when the product changes, the affected assets can be listed rather than discovered by somebody noticing an old interface in a live ad.

It is also worth writing whether the demonstration is real or a mock-up, and if a mock-up, that it is labelled as one. Showing something that does not exist is the fastest way to lose the sceptic the champion forwarded it to.

This category writes too much. The instinct is understandable: the product is complex, the buyer is informed, and there is a lot to say.

The fix is a specification rather than a plea. Put a word limit in the specs column and make the hook carry the weight. A static with twelve words and a clear proof outperforms one with sixty, not because attention is short but because sixty words in a feed placement renders as an unreadable block and is scrolled past.

Where the argument genuinely needs length, that length belongs on the landing page or in the document, which is what the asset is there to send people to.

The recurring ones: no stock photography of people in meetings, which signals that there was nothing real to show. No claim about time or money saved that the linked calculator cannot reproduce. No customer name without written permission. No competitor comparison that has not been checked against the competitor's current page. No screenshot containing real customer data, ever.

The last one deserves its own line in every brief, because it is the mistake that cannot be undone by pulling the ad.

One more decision belongs in the brief rather than in the edit: which format each objection deserves.

A cost objection is usually answered fastest by a number, which wants a static or a plain text card rather than a film. A risk objection wants evidence and detail, which usually means a document or a customer account. An effort objection wants a demonstration, which is the one case where video earns its production cost. A timing objection is usually answered by an offer rather than by creative at all.

Briefing format against objection stops the default from taking over, and the default in this category is to make a video because video is what everybody makes. A short video that gestures at a cost argument is worse than a static that states the number, and it costs a great deal more.

Seven checks. One row per asset, each naming one ICP segment and one objection. Every asset carries proof that survives being forwarded to somebody with no context. Demonstrations record exactly what they show, so they can be found when the product changes. The specs carry a text limit. Must avoid names the stock imagery, the unverifiable claims and the customer data that are out. And every due date leaves room for a review and a re-edit before the day the asset is needed.