How to announce a product launch
Launch posts are the most written and least read announcement there is, because almost every one of them describes the building rather than the using. The reader is being asked to change what they do, which is a much larger request than any other announcement on this list makes.
It is the only one that asks the reader to change a habit
Every other announcement here asks for attention. This one asks for adoption, and adoption is a behaviour change with a switching cost attached. That single difference should reorganise the whole post. A reader deciding whether to try something needs to know what they will be able to do afterwards that they cannot do now, whether their existing work survives, and what it costs them to find out. None of those three questions is answered by how long the team worked on it or how proud anyone is. The launch post that fails is not badly written, it is aimed at the wrong decision: it argues that the thing exists when the reader is deciding whether to move.
The rule for this one
Open with the sentence that begins "You can now". If you cannot finish that sentence, you have shipped a feature and not a launch.
What the category publishes, and what the rule produces.
The first draft is a composite written to match the register the ranking pages for this query actually publish. It quotes no company and no real post. The second obeys the rule above and nothing else, so anything that changes between them is the rule doing it.
The version everyone posts
We are excited to announce the launch of our new platform. This innovative solution is designed to help businesses of all sizes streamline their workflows and unlock their potential. After months of hard work from our incredible team, we cannot wait for you to see what we have built.
The version that obeys the rule
You can now schedule a month of posts in one sitting instead of one evening a week. The calendar view went live this morning. It is on every plan, including the free one, and nothing you have already scheduled moved.
It describes the object rather than the change. Nothing in it tells a reader what they will be able to do on Tuesday that they could not do on Monday, so there is nothing to act on and no reason to keep reading past the first clause.
8 of 10 measured dimensions moved.
Both drafts were read by the engine behind our free brand voice generator. It is arithmetic over the text rather than a judgement, which is the reason it is used here: every figure below can be checked by counting the draft printed above it.
| Dimension | Before | After | Moved |
|---|---|---|---|
| Sentences | 3 | 3 | No |
| Words | 49 | 40 | Yes |
| Average sentence, in words | 16.3 | 13.3 | Yes |
| Longest sentence, in words | 20 | 17 | Yes |
| Contractions per 100 words | 0 | 0 | No |
| Speaks to the reader | No | Yes | Yes |
| First person (we, our, us) | 5 | 0 | Yes |
| Second person (you, your) | 1 | 2 | Yes |
| Vocabulary richness, out of 100 | 84 | 85 | Yes |
| Contains a stock phrase | Yes | No | Yes |
Read across all fifteen moments the pattern is sharper than any single page: the hub has the corpus numbers.
What earns the post
- The specific thing that is now possible, in the reader's words rather than the feature name.
- What happens to work they have already done. Silence here reads as a migration.
- Where it sits on pricing, said plainly, because the first suspicion is that this is the upsell.
When not to post it at all
Skip it when the change is invisible to anybody outside the team. An internal refactor announced as a launch teaches your audience that your launches are not worth reading, which is expensive and takes months to undo.