Product launch

What a launch has to do that a campaign does not

A campaign persuades people to buy something they already understand. A launch has to teach first and persuade second, and the teaching is the part that gets skipped, because the team has understood the product for months and can no longer hear how it sounds cold.

The one rule: settle the positioning sentence before any copy exists. Everything downstream inherits from it. The announcement email, the ad, the objection handling, and the number you will judge the launch by are all restatements of one sentence. Rewriting that sentence in week three means rewriting all of them.

The sentence has a shape. For who, who situation, product is the category that single differentiating benefit, unlike the alternative they use today. The last blank decides whether you have a launch at all, and the alternative is usually not a competitor. It is a habit, a spreadsheet, or doing nothing.

How to think about the sequence

• Proof is a phase, not a paragraph. A launch with no checkable proof is an announcement. A number, a demonstration, a named customer, a guarantee: something a sceptic could go and verify. Adjectives are not proof, and a launch that leans on them spends its budget teaching people to distrust the claim.
• Objections belong in the customer's words. Write the sentence someone says to themselves right before they do not buy. Marketing's paraphrase of that sentence is always more flattering and always less useful.
• Aim narrow first. A launch pointed at the whole list reaches nobody. Start with the group for whom this is obviously, immediately for them, and widen on evidence rather than on hope.
• Build the schedule around one moment. Teaser for curiosity, explainer to teach, announcement to sell, then objection handling and social proof once real buyers exist. A last-call send only when there is a real deadline; an invented one is borrowed from the next launch.
• Write the kill criteria before launch day. What result at T+2 would make you change the plan rather than send the next email anyway. Once a launch is running, nobody has the standing to invent that threshold.

Failure modes worth naming: announcing before the thing can actually be bought; opening the announcement with excitement rather than explanation, which loses everyone who has never heard of this; treating "new" as the benefit, when new is only interesting to people who already know you; and moving the success number after seeing the result.

Numbers to hold it against

Launch benchmarks are softer than campaign benchmarks, because a launch is a one-off against an audience with no history of the product. Prefer your own comparable launches and hold published ranges loosely.

• An announcement to a well-chosen segment usually opens above your list average, often five to fifteen points above, because the segment was picked for affinity. If it does not, the subject line was about you rather than about them.
• Most launch email revenue lands within 48 hours of the announcement. Design the T+2 and T+7 sends to work the people who did not act, rather than to repeat the announcement to the people who did.
• A first audience of roughly five to twenty percent of the list is the usual shape: narrow enough to be genuinely relevant, large enough to produce a readable result.
• On paid social, cold ecommerce link click-through commonly sits between 0.8 and 1.8 percent, and CPMs move by season far more than by creative. Never read a November CPM against a February one.
• An ad set needs roughly 50 optimisation events a week to leave the learning phase. A launch that splits a small budget across six ad sets learns nothing about any of them. Consolidate first, split later.
• If the click-to-purchase rate on launch traffic is under half your site's normal rate, the problem is the page, the price, or the positioning, not the creative. Fix that before buying more clicks.

What has to be true before a card asks for approval

• The announcement email opens with what the thing IS, in one sentence, above the fold. Then the single differentiating benefit from the positioning card, then one proof point, then one call to action.
• Every claim in an email or an ad traces to a proof point card on this board. If it does not, the claim is either unproven or unwritten.
• The ad and the landing page say the same thing, in the same words where they can. The gap between them is where launch traffic goes to bounce.
• The creative works muted and without the caption. The first frame carries the message on its own.
• Price, availability, shipping dates, and any guarantee are checked against reality, not against the plan from six weeks ago.
• No image placeholder and no unresolved bracket survives into awaiting_approval.

The seed cards are the first instance, not the ceiling

Eight seed cards show the shape. A real launch is denser, and the Proof phase is where thinness costs the most.

A healthy Position phase carries one positioning card and one audience card per wave. Proof carries four to eight cards: each significant objection paired with the proof that answers it, and the objections you cannot answer written down anyway, because those are the launch risks. Sequence holds one plan and the notes it needs. Create holds one card per timed touch in the plan, plus the variants each paid touch needs, which is usually four to eight emails and three to six creatives. Measure holds one result card per number the positioning card named.

Two objection cards on a launch board means the launch has two objections written down. It does not mean the launch has two.

The layout is one arrangement of many

This board runs by production step, with Launch as a schedule that renders the timed cards rather than holding its own. That fits a single launch moment. A staged rollout by market, a phased release by tier, or a launch running retail and paid in parallel often reads better arranged by wave or by channel. Rearrange the phases; the sequencing logic and the proof discipline survive it.

When this is the wrong playbook

A restock, a new colourway, a price change, or a seasonal promotion is a campaign rather than a launch: the audience already understands the thing, so the teaching phase is dead weight and the campaign playbook will get there faster. Use this board for a genuinely new product, a major version, or a category your audience has never bought from you. And if the product will not ship for six months, this board will build a plan that goes stale before it is used. The honest move then is to run Position and Proof now and open the Sequence phase when the date is real.