What 60 crypto threads taught me about explaining hard products
Hard products get simpler when you start from the reader’s actual situation, keep one idea per post, and test every explanation on someone outside your field.
If your reader has to learn three new words before your second sentence, you have already lost them.
Two kinds of writing taught me that. At CoinDCX I wrote more than 60 educational threads on crypto and Web3 basics. Before that, at Axelar, I wrote 40+ research reports on interchain infrastructure. The reports taught me how the systems work. The threads taught me how little of that machinery a reader needs before they can make a decision.
At HeimLabs I switched between the two all the time, writing for builders who want the machinery and for users who want one answer: will this solve my problem? The audience changes, but the editing questions stay the same: what does this person need to decide, and what can I leave out?
Start from the reader’s situation
The natural first draft starts with a definition. “A cross-chain bridge is a protocol that lets assets move between blockchains.” That sentence is accurate, but it doesn’t answer the question that brought most readers there.
What a newcomer actually has in their head is a situation: “My USDC is on one network and the app I want to use is on another. What do I do?” Open with that, and the definition can come later, once the reader cares. You are explaining the same bridge, starting from the moment someone needs one.
At Jaro Education, selling higher-education programs, the calls were consultative, so most of each one was listening. Those calls taught me to listen for the worry behind the question. “How long is the program?” can really mean “Can I do this while keeping my job?” The calls went better when I answered the worry first. Explainer content works the same way. Before you answer, ask what decision the reader is trying to make.
One idea per post
A thread feels like it has room for everything, which is exactly the trap. Put wallets, seed phrases and gas fees in one thread and every tweet introduces a new idea before the last one has landed.
My rule: if a post needs a second new idea to make sense, the second idea gets its own thread, and I link to it. So “how to set up your first wallet” becomes three pieces. One covers what a wallet is and which kind to pick. One covers the seed phrase and why nobody should ever ask you for it. One covers why your first transaction costs a fee. Each answers one question fully, without asking the reader to understand the other two first.
Borrow a picture the reader already owns
Analogies do a lot of work in crypto writing. Gas fees are a bit like surge pricing on a ride app: when more people are trying to get their transactions processed, the price goes up. Most readers already understand surge pricing, so they get the idea in one line.
An analogy also needs a line saying where it breaks. On many networks, if your transaction fails you still pay for the work the network did trying to run it. No ride app charges you for a trip that never happened. Leave that out and the analogy teaches something wrong, and the reader finds out with their own money. So I write the picture, then one line on where it stops being true.
Make every claim something you can picture
Harry Dry’s copywriting test at Marketing Examples asks whether you can visualize a line and whether it can be proven false. When I edit an explainer, I ask two questions: can the reader picture this, and could someone check it?
“Faster and cheaper cross-chain transfers” fails both questions. “Move USDC from one chain to another in one step, and see the fee before you confirm” passes, because you can picture the screen and anyone could check if it’s true. When I can’t turn a vague line into something concrete, it usually means I don’t understand the product well enough yet. Then I go back to the product and engineering team with questions, not back to the thesaurus.
Write the long version first, then cut
The Axelar reports were long because the subject needed it: how messages move between chains, who verifies them, and what happens when something fails. Once I had mapped who has to be trusted and where things can break, I knew what a short post could safely leave out and what it couldn’t.
I still start complex launches with a private working document: how the product works, where it can fail, and which conditions change the answer for a user. The public post takes one slice of it. Starting from the full version lets me simplify without quietly dropping a risk or a step that matters.
Test it on someone outside your bubble
Before posting, send the draft to someone outside crypto and ask them to explain the product back to you in their own words. If they can’t, the draft goes back. It is the quickest way to find the jargon you stopped noticing months ago.
After posting, I look past impressions. Impressions tell you a thread was shown to people. Saves, replies and follow-up questions are imperfect signals, but they say much more about whether it was useful or confusing. I keep those numbers in a simple sheet next to each topic, so over time I can see which kinds of explanation people kept and which they scrolled past.
Your wording also gets quoted outside the page you wrote it on. Search summaries and AI answers often reuse the wording they find in product pages, docs and public explainers, so a reader may meet your explanation before they ever visit your website. If every description of your product is full of jargon, that jargon is what gets repeated.
Before you publish your next explainer, give it to someone outside your field and ask three things: what does the product do, who is it for, and what could go wrong? Every answer they miss points to a sentence you still need to rewrite.
