CoinHaven

Fifteen interviews, and what people would not put in a survey

I filtered the 120 respondents down to 15 for semi-structured interviews over Zoom, scheduled across different times so I was not only hearing from one timezone or one kind of creator.

The survey gave me the shape of the problem. The interviews gave me the reason nobody had solved it, which was that the anxieties driving the market are not the ones people volunteer on a form.

• Jane — a creator priced out

  “I love how easy it is to mint NFTs on some platforms, but the high transaction fees are a real problem for me. Lower fees would make the experience much better.”

• Michael — a collector braced for loss

  “Security is my top priority. I’ve had a few scares with phishing attempts, so I need a platform that guarantees the safety of my digital assets and personal information.”

• Kim — on what actually helped

  “A positive experience for me was when I found a platform with excellent customer support. They helped me through the minting process step-by-step, which was very reassuring.”

• Emily — on why she stays

  “I enjoy platforms that offer a wide variety of high-quality NFTs. It keeps things interesting, and I feel like I’m always discovering something new and unique.”

The raw feedback, before it was tidied into anything

Four of the fifteen, quoted as said. Two named cost, one named security, one named support and one named variety — which looks like four different problems until you notice that three of the four are describing risk and only the last is describing desire.

Tap to inspect the full-resolution artifact

And what fifteen interviews collapsed into

The synthesis step: fifteen conversations reduced to three profiles — a first-time creator minting a debut piece, a seasoned artist expanding across chains, and a collector trying to buy something real. Three different jobs, and every one of them managing risk before pursuing a goal, which is not how any competitor was organised.

Tap to inspect the full-resolution artifact

Nobody tells you in a survey that they are afraid. They tell you in an interview, and then only obliquely, by describing the workaround.

Four numbers that set the brief

For many creators the incumbents were a double-edged sword: they were the giants, and they were not built for everyone. Gas fees on Ethereum made minting an expensive gamble, small creators could not get noticed among big names, and scams were constant.

But it was not only complaints. These creators had hopes — for something more affordable, more inclusive, more secure, with tools to build an audience before drop day. That was not a wish list. It was the design brief.

• 60% — lack of visibility

  Small creators could not compete with established names on platforms that ranked by volume rather than by quality. Discovery was the barrier, not the tooling.

• 75% — cost barriers

  Three quarters of respondents named high Ethereum gas fees as a significant challenge, which makes minting inaccessible to exactly the creators the product was for.

• 65% — multi-chain needs

  Two thirds wanted support for alternatives like Solana and Polygon. This is the same problem as cost, expressed as a solution people had already worked out for themselves.

• 55% — security concerns

  Scams were rampant and users wanted robust verification. More than half named it unprompted, which is high for a risk people are embarrassed to admit.

The journey, with the risk moments marked

Mapping where anxiety spikes rather than where clicks happen. The tense moments cluster before any value has been delivered, which is what made the onboarding decision later so consequential.

Tap to inspect the full-resolution artifact

The gap a four-person team could actually take

OpenSea, Rarible and Magic Eden mapped against creator pain rather than feature lists. None optimised for a first-time creator and none offered multichain minting — which is the only reason picking a fight with the category leader was defensible.

Tap to inspect the full-resolution artifact

From findings to user stories, then to a cut list

I turned the research into user stories rather than features, because a story keeps the reason attached to the request. One insight sat above all of them: simplicity was non-negotiable. Creators were overwhelmed by the existing platforms, so the product had to feel obvious from the first click.

• Then a workshop, scored rather than debated

  I facilitated an ideation session with the whole four-person team and scored every idea on user impact against technical feasibility. Scoring beat arguing: it meant a founder’s favourite and a designer’s favourite were compared on the same two axes, in public.

• And a deliberate 40% cut

  Minting, creator verification, buying, selling and marketplace filters went into the first release. A referral system for creators, search enhancements and multi-wallet swap were sized as bigger bets. Advanced creator analytics and live chat went to the next iteration. Multichain support beyond the launch set and buzz-generation tooling were explicitly not now — the hardest column to hold, and the reason the other three shipped.

• As a new creator, I want to easily mint my first NFT without worrying about high fees, so I can start selling my work.
• As a buyer, I want to see verified creators and trustworthy listings, so I feel confident in my purchases.
• As a returning creator, I want to track my sales and referrals in one place, so I can measure my success.

Impact against effort, with a Not Doing Now column

The output of the workshop. The quadrant that mattered most was the empty-looking one: naming what we were not doing, in front of the founders, is what stopped the list growing back. 40% of the initial features never made the first release.

Tap to inspect the full-resolution artifact

What shipped

Five areas carried the trust argument into the product. None of them is a badge bolted onto a finished marketplace; each is a rule about who gets seen, and what a buyer can check before spending.

• A marketplace ranked on engagement, not spend

  Connect a wallet, browse, buy, or create. Creators are ranked on engagement metrics — volume, quality, activity — so a first-time creator can surface beside an established one. Filters cover price, category, blockchain and creator verification status, and trending items plus personalised recommendations do the discovery work. This is the 60% visibility finding, answered.

• Minting as a guided sequence

  Step-by-step rather than one intimidating form. Upload audio, video or images; choose a chain — Ethereum, Solana or Polygon — to keep gas fees down; then set royalties, bidding, or a fixed price. This is the 75% cost and 65% multi-chain findings, answered together.

• Verification that changes ranking

  Creator verification exists to attack the scam problem directly, and verified creators get priority ranking. That is the difference between trust as decoration and trust as architecture: it shows up as a sorting rule, not a tick.

• Collections a buyer can assess

  A collection page gives a buyer the creator’s whole portfolio rather than one item in isolation: buy a whole collection, sort by price, popularity or type, follow the creator for future drops, share the profile.

• Referrals that pay creators for creators

  Creators earn a percentage of the revenue generated by the creators they refer, with a dashboard tracking referrals, earnings and network performance. Growth routed through the people the market already trusted, rather than through paid acquisition a four-person team could not afford.

Entry — a route in, not a wall of listings

The homepage leads with what a creator can do — explore, or create — and puts top categories beside it. For a first-time visitor who arrived braced to be scammed, the first screen has to offer an action rather than an inventory.

Discovery — where ranking becomes the trust surface

Featured projects, then top collections over 1, 7 or 30 days. This is the screen the 60% visibility finding lives or dies on: a short window lets a new creator appear, a long one entrenches the incumbents, so the window is the user’s choice rather than ours.

The creator, judged before the item

Verification tick, socials, and the four numbers a buyer needs — total items, owners, volume, floor price. Assessing the body of work before the individual piece is the sequence collectors described in interviews, so the product follows it rather than fighting it.

The item, buyer’s view

Creator and collection are named above the price, provenance sits in Details, and only two actions compete: place a bid, or buy now. Everything a buyer would use to detect a fake is above the thing that takes their money.

The same item, owner’s view

Properties become editable traits, and token ID, contract address and standard are exposed in full rather than summarised. One layout, two roles: what changes is what you can act on, not where things are.

And the loop back out

Referral earnings, tracked. Creators earn a share of revenue from creators they refer — growth routed through the people the market already trusted, which is also the cheapest distribution available to a startup with no brand equity.

Reflection

In a zero-trust market, the most important design decision is what you show before asking for anything. Every trust signal I embedded mattered more than any feature, and the onboarding result proves it: the same product, gated differently, lost or kept three quarters of its users.

The result that surprised me most was the terminology one. I had assumed minting friction was structural and would need flow changes. It was linguistic. That is a useful correction to carry, because vocabulary is cheap to fix and easy to overlook when you are hunting for architectural problems.

The prototype did a second job I had not designed it for. It became the centrepiece of the funding pitch, which is a reminder that in an early-stage company the artefact has two audiences — the user and the person deciding whether the user ever gets it.

What I would do differently: push harder for native mobile from day one. Responsive web worked, and it limited the engagement patterns available to us. The foundation is there, but that decision was made on timeline rather than on what the product needed.