How I launch a product when nobody is listening yet
When a new Web3 or AI product has no audience, borrow other people’s attention and plan the launch as two weeks of small moments instead of one big post.
Most launch plans assume someone is waiting for the announcement. For the products I usually work on, nobody is.
At HeimLabs I worked on go-to-market for Web3 and AI products, and many of them began with a working product, an almost empty social account, and a pitch that takes a full paragraph to explain. The usual move is to write a big launch thread, post it, and ask the team to repost. That gets you one busy afternoon and then quiet. So before I write the announcement, I make a distribution list: community moderators, ecosystem accounts, newsletter writers, and the builders our future users already follow.
Start with other people’s audiences
I build it by tracing where potential users already ask questions, compare tools and take recommendations. For a Web3 product that is usually a few Discord and Telegram groups, some research accounts on X, and the community of the chain the product is built on. For an AI tool it might be a subreddit, a niche newsletter, or a handful of builders who post their workflows every week.
I saw this from the creator side first. At Axelar I was in the research collective and posted about the ecosystem, and at CoinDCX I wrote educational threads for the community. Each week I had to decide what was worth explaining to the people who followed me. I was most likely to cover a project when I could try a demo, see what was actually new, and answer one question for my readers: why switch? A logo and a launch date gave me nothing to write.
So I start with a one-page brief for the people I hope will share the product. It has the product in one sentence, the one thing it does that wasn’t possible before, a link they can try today, and the three questions their followers will probably ask, with answers.
Speak each community’s language
Each community worries about something different, and generic launch copy answers none of those worries. Take a tool that moves stablecoins between chains. I would describe it differently to a DeFi trader and to an online seller.
For a DeFi Telegram group: “You can move USDC from Arbitrum to Base in one step. Fees are shown before you sign, and here is what happens if the transfer gets stuck.”
For a group of small online sellers: “If a customer pays you in USDC on a different network than your wallet uses, this moves it over without you learning what a bridge is.”
The facts don’t change. The first version answers a trader’s worry about cost and risk. The second answers a seller’s worry about being confused. Writing a version for each group takes longer than pasting the same thread everywhere. I know it worked when people reply asking whether their chain is supported or whether it works with their wallet. That is a question from someone thinking about using it.
Plan two weeks of moments
It is tempting to treat launch day as the finish line. I treat it as the first reason to talk about the product. Elena Verna has described how Lovable ships and announces constantly, with engineers posting about their own smaller features, and MKT1 has made the case for keeping momentum after day one.
When the brand account is new, the first announcement brings little traffic. It works better as a reference point: something concrete that founders, partners and community members can link to. So I plan two to three weeks of smaller posts and events:
- Day 0: the announcement and a short demo video, posted from the founder’s account as well as the brand’s.
- Days 1 to 3: one technical post from engineering on how the hardest part works. I help them shape it, and it stays in their words.
- Week 1: a step-by-step walkthrough of one real use case, with a screenshot of each screen.
- Week 2: what the first users asked for and what changed because of it.
- Every day: someone from the team answers each question in the community threads on the day it is asked.
By week two the announcement is old news. A shipped request or a real workflow gives partners a fresh reason to mention the product again.
Let the team do the talking
A new brand account with no history reads like an ad. A founder explaining why they built something reads like a person. In early launches I spend most of my writing time helping the team with their own posts, and the brand account mostly reposts them.
At ScribbleDAO I was a core contributor for five months and built the influencer and copywriting frameworks for its campaigns. A DAO has no marketing team in the usual sense, so the members have to start the discussions themselves. It changed how I run launches: I want the people building the product to be the ones explaining it. Every early post ends with something a reader can do or answer: a use case to try, a question about their setup, or a request for the feature they want next. That gives readers an easy way in, and it gives the founder real replies to answer instead of the brand account talking to itself.
Decide what “worked” means before you post
The last part of my launch plan is written before anything goes live. I pick the one action that shows real interest, like a first transaction from a newly connected wallet, or the first output someone saves from an AI tool. Impressions don’t count.
Then I make sure that action is tracked, with links tagged per channel and per partner. The next morning I compare that action by source in Google Analytics and with a simple SQL query on product events. If the walkthrough brought people who completed the action and a partner post brought only views, the walkthrough gets the next round of effort, and that partner gets a better brief.
If you are launching to no audience, start your plan with a list of other people’s audiences and one honest sentence about why their followers should care. Write that sentence before you write anything else.
