The short version
- Launch creative starts with what actually changed, not with what the hero image should look like.
- A hero image can only make people register that something happened. Explaining it, proving it, making it relevant and pointing to the next step are separate jobs.
- What teams miss before shipping is rarely the aspect ratio. It's stale screenshots, wrong prices, promises the product doesn't keep, and a landing page that says something else.
Before a launch, a creative team needs one plain sentence about what changed: which users can now do something that used to be harder, and what new capability lets them. Settle that, then work out the audience angles, the mix of assets, and the channel cuts. A hero image only makes people register that something happened — explaining it, proving it, and pointing to the next step are separate jobs.
Plenty of launches look well prepared from the inside. There's a hero image, an announcement post, a handful of social cuts. Then it goes live and people learn that something shipped without ever learning what it does for them.
The design is usually fine. The problem is that every asset is doing the same job.
Write one plain sentence about what changed
Before anyone opens a design file, write this out in ordinary language:
Which users can now do something that used to be harder, and what new capability lets them.
Skip "major upgrade" and "completely reimagined." Words like that tell a creative team nothing about what to emphasize, and they don't help a reader decide whether to keep reading.
If you can't get that sentence out cleanly, the product story isn't ready. Designing on top of it just makes the uncertainty better looking.
Every launch has at least five jobs to cover
| Job | What the reader wants | Usual formats |
|---|---|---|
| Register the news | What happened | Hero image, announcement headline, video thumbnail |
| Explain the change | How this differs from before | Feature breakdown, before and after, flow diagram |
| Show evidence | Why should I believe it | Real UI, step-by-step walkthrough, stated limits |
| Make it relevant | Why this matters to me | Role-based scenarios, task examples, problem lists |
| Drive action | Where to go next | Trial link, docs, demo, release notes |
Five jobs doesn't have to mean five images. One asset can carry two of them. What you want to avoid is every asset shouting "it's live" and nothing else.
The launch creative checklist
Product facts
- The feature name and the final UI are locked.
- You know whether it ships to everyone or only to some plans, regions or account types.
- Pricing, quotas, usage limits and availability dates are accurate.
- You can say what shipped and what's still on the roadmap.
- You have real UI or real output you're allowed to show publicly.
Message hierarchy
- The headline carries one change, not three.
- The subhead says who benefits.
- Evidence and limits sit where people will actually see them.
- The call to action matches what the landing page says.
- No internal codenames standing in for words customers use.
The creative set
- One core visual people will remember.
- A set of assets that explain how it works and what it's worth.
- At least one real use case.
- A short version sales, support or community folks can forward without editing it first.
- Per-channel cuts with the right ratio and the right amount of text.
Don't leave channel cuts until launch day. Start with a reusable SaaS marketing assets workflow, then check crop and safe areas against each platform's specs. For Xiaohongshu image posts, the default production baseline is 3:4 at 1080×1440.
Pre-ship check
- Screenshots come from the shipping build, not from last sprint.
- Logo, product name, prices and numbers are correct.
- Small type inside the image is still readable on a phone.
- After the vertical crop, the headline and the product itself stay clear of platform UI: captions, avatars, buttons.
- Every link opens, and the signed-in and signed-out views both look the way you expect.
- Nothing in the copy hints at something the product can't do yet.
Different audiences need more than a swapped job title
The same feature lands differently with a PM, a designer and a team lead.
A PM wants to know whether the process got shorter. A designer wants to know how much control they keep. A team lead wants to know what it does for handoffs and delivery dates. Changing "your team" to three job titles in the same headline isn't three versions. Real audience fit changes the order of the argument and the evidence behind it.
Our playbook for product and growth teams splits a launch into audiences and message angles first, then decides which versions to produce. The point isn't to make more assets. It's to stop every asset from circling the one sentence the team is most comfortable with.
Don't archive everything the day after launch
Keep three kinds of notes from every launch:
- Which creative the team actually used.
- Which versions got dropped, and why.
- Which product facts and screenshots will go stale first.
Without them, the next launch starts from an empty file again. What's worth keeping isn't only the finished art. It's the thinking that produced it.
FAQ
Is this too much work for a small team?
You don't need all of it at once. Cover the news, the explanation and the next step first, then add scenarios and evidence as channels and feedback show you what's missing. Covering every job matters more than the number of files.
Should the hero image come first or last?
Last. Settle the change sentence, the audience and the message hierarchy first. Otherwise a good-looking direction gets locked early and the product facts have to be squeezed in around it.
Does LinkBloom publish any of this for us?
No. LinkBloom handles the work before the launch: working out directions, producing the versions, and running a first quality check. Scheduling, publishing, distribution and reading the numbers stay with your team and the platforms you use.