Marketing a DeFi product before MiCA made it urgent

I recently joined Tyler Fleischer on The Marketing Desk to talk about launching Tesseract's Dedicated Client Vaults, where I've been Head of Marketing for the past year. These are the takeaways.

Oliver Carding on The Marketing Desk, September 2026. Watch the full episode on YouTube.

We launched the vaults at the end of March 2026. Each institution gets its own vault: its own smart contract, its own address, its own positions. Nothing is pooled with other clients, so the assets, returns and risk in a vault belong to that client alone, withdrawals can only go back to the client's own wallet, and the client can see its balance in real time. For an institution that has to explain every position to its own compliance team and auditors, that separation is the point.

We built the product so institutions would be ready before MiCA, the EU's crypto regulation, finished its transition in July 2026. The gap was awareness. MiCA says little about DeFi: it covers services run by an identifiable firm, sets aside fully decentralised ones, and left the wider question of DeFi for a later review. Vaults sit somewhere in between, so plenty of buyers hadn't yet asked whether the vault they used was run in a way their compliance team would accept.

1. Start educating before the regulatory deadline

The vaults came from client feedback. When we looked at how vaults usually work, the same pattern kept coming up: deposits pooled together, and a curator in the middle choosing the strategies and moving the capital between them. On chain or not, choosing and managing strategies with other people's money is asset management, and portfolio management is a regulated service under MiCA. Running it through DeFi doesn't change that. So we built a vault per client and took that management role ourselves, through Tesseract Investment Oy, the group's MiCA-authorised entity, as a discretionary portfolio management service.

The buyer's problem already existed. An institution putting treasury or client money into DeFi needs a yield strategy that stands up in front of its own compliance team. A lot of DeFi marketing explains how a protocol works. Ours had to explain why the structure mattered, months before the deadline would make buyers ask. In January, February and March that argument landed slowly.

We launched with a small pilot group of institutions and opened direct sign-up in June through a custody partner. Interest came from crypto institutions, as planned, and from traditional finance, which we hadn't been thinking about. Authorisation opened those conversations. What buyers wanted next was detail: how the vault is set up, who holds custody, and what happens to their assets if Tesseract fails.

2. Building brand awareness through the CEO

Tesseract had been operating since 2017 but hadn't invested much in marketing; I was its first head of marketing. The vault market already had well-known names, and we were asking institutions to trust a new product from a company most of them hadn't heard of. We needed a recognisable voice before we needed a campaign, so with our PR agency, Evolve, who I'd recommend to anyone marketing in this space, we built a thought leadership programme around James Harris, our CEO.

The programme ran problem first, solution second. For months James set out the problem: vaults are a better way to manage assets on chain, but the way they're usually run, open to anyone, pooled, and managed by a curator nobody has authorised, isn't something an institution's compliance team can sign off. He made that case on podcasts and webinars, in press commentary and on panels, without naming or attacking anyone, because the aim was to be the firm that had solved it, sitting alongside what DeFi had already built.

The solution arrived as part of the go-to-market. We launched the vaults at the end of March, and the next day James gave the keynote at Vault Summit Cannes, in front of the people who build and use vaults. By then a lot of the audience had already heard the problem. The keynote gave them the answer.

3. Why I started judging PR by more than backlinks

I came to press through SEO, so I used to judge it by the attention and links it earned. Now I judge it by the story it tells about the company. In more than a decade of crypto and Web3 marketing, I've never relied on press as much as I do with institutional buyers, because it's where they already take in information. It takes more commitment than paid media. Someone has to write the content and work out where it fits in the strategy, and plenty of companies can't or won't do that, which is why it still cuts through.

Tesseract started the year with almost no press. Between mid-April and early September, the running total of placements and the number of outlets covering us both more than doubled, including repeat pieces in Bloomberg. The effect showed up before we had a product to sell: branded searches and enquiries rose off the back of the coverage in the months before launch.

It also mattered for GEO, the work of getting cited in AI answers. Generative search leans heavily on what reputable outlets say, so every citation in a credible publication was doing search work as well as PR work. When the product launched, Tesseract quickly started ranking for vault terms, on the first page of Google for searches like "crypto vaults" and "what is a crypto vault", and cited in Google's AI Overviews.

4. Remarketing to the people who engaged

As the launch got closer, we put a small paid budget behind retargeting people who had engaged with the education content, mainly on LinkedIn, where the institutional buyers we wanted were already reading. Those people went on to ask for more information, and the enquiries that followed came from ETP issuers, a large traditional asset manager, exchanges and other MiCA-authorised firms. That mix is the best evidence I have that the education reached the buyers it was written for.

5. Decide how you'll measure before you launch

Institutional DeFi sells slowly: first contact, due diligence, onboarding, then a funded vault, over months. The slow cycle wasn't the real attribution problem, though. The problem was tying a podcast, a Bloomberg quote, a LinkedIn post and a conversation at an event to the one enquiry they produced. Our form wasn't passing analytics identifiers to the CRM, and email enquiries often arrived without a source. It took until August, and tagged campaign links, to connect a tracked visit to an institutional enquiry.

Next time I'd set two scorecards before launch and agree them with leadership. Early signals: the right institutions engaging (about a quarter of our new LinkedIn followers matched the target profile), qualified enquiries and due diligence requests. Later outcomes: onboarding and funded vaults. And every enquiry, from any channel, would land in the CRM with its source.

6. At launch, switch the whole message from education to conversion

Before launch, everything we put out was built to educate and gauge demand. James framed the problem, our articles explained it, and the vaults page was long and educational, with two animations comparing how vaults work elsewhere with how ours work and a form to register interest. It was passive on purpose, because we were still testing whether the argument landed.

Launch changed the objective to selling the product and getting institutions started, and all of our messaging had to change with it: social, email, press, paid and the website. The copy moved from why vaults needed fixing to what you get, how it differs and how to start. The vaults page shows the shift most clearly. A buyer arriving from a press piece or a retargeting ad needed to see the difference in a few seconds and know what to do next, and the quickest way I've seen to do that is the comparison table on SaaS pricing pages, so I copied it and replaced the animations:

It answers the question buyers kept asking, how this differs from any other DeFi vault, and sits directly above the routes to get started. Legal's only condition was that we didn't name competitors.

7. Where AI fits in marketing compliance

At agencies, with banking clients, the back and forth over a single piece of copy never seemed to end. I didn't want that at Tesseract, and I didn't want compliance to become the reason the writing got worse, so I designed a workflow that builds the rules in from the start. In February, Legal, Compliance and I wrote the rules down as markdown files: what we can and can't say, how each product must be described, which disclaimers go where. We review them together roughly once a month. In March I built a quality gate, a Claude skill that reads a draft against those files and flags a product described against the wrong entity, a banned phrase, a missing disclaimer or a number with no source. Then someone who knows the subject checks the facts, often James or someone from the product team, and the piece goes to compliance for a full review.

It made review quicker and far less painful than anything I'd dealt with at agencies. AI is a good assistant in that process because it catches the mechanical mistakes before a person has to. A person still writes every piece, checks it and signs it off. When Tyler asked what I'd tell someone starting at a regulated crypto company, this was most of my answer: learn the regime as seriously as you learn the product, and build the system early.

Available for work

I've spent the past year taking a regulated DeFi product to institutions: the positioning, the press, the website and the compliance process behind all of it. I'm open to new roles, either full time or on a contract basis. If you're launching a DeFi product or marketing a regulated crypto business, whether you're hiring in-house or looking for a DeFi marketing agency, get in touch. There's more on the work I do on Web3 and crypto marketing.

The full episode is on YouTube: The Marketing Desk with Oliver Carding.