Build vs buy: should we build our own tokenization stack?

Build only if tokenization is your product and you will staff smart-contract, security, and compliance engineering for years. Otherwise buy: the compliance, standards, custody, and maintenance are the real cost, not the first contract. Most should buy.

The question that decides it is not “can we build a token contract?” Almost any competent team can. It is “will we own the compliance, security, and maintenance of a securities-grade system for as long as the asset is outstanding?” That is a decade-long commitment, and it is where build decisions usually go wrong. The buy side of the ledger is concrete: an all-in tokenization runs about $10K–$90K on a platform whose flat fee includes the offering documents, versus $50K–$720K when counsel drafts them separately (third-party ranges researched July 2026 for the tokenization cost index) – while the standards a build must track moved again as recently as May 2026, when ERC-7943 reached Final status as Ethereum’s RWA standard (GlobeNewswire).

Why the first contract is not the decision

Minting a token is the cheap, visible part. The expensive, permanent parts are the ones a demo does not show:

• Compliance at the transfer layer – enforcing eligibility, lock-ups, and jurisdiction rules on every transfer, and keeping that correct as rules change.
• Custody and key management – integrations with qualified custodians, and the operational discipline that keeps keys safe for years.
• The register and its continuity – being the enforceable record of ownership, and staying enforceable if your team moves on.
• Standards upkeep – the token standards (ERC-3643, ERC-7943) and the chains they run on evolve; someone has to track and adopt.
• Repeated audits – not one audit at launch, but re-audits after every material change.

Build means owning all of that, forever. Buy means paying someone whose full-time job it is.

When build is genuinely right

It happens, and pretending otherwise is dishonest:

• Tokenization is your product. If you are building a platform others will use, the stack is your differentiator – build it, and staff it accordingly.
• You already run the team. You have standing smart-contract, security, and compliance engineering, and the appetite to keep them on this for years.
• Control or sovereignty requires it. A regulatory, data-residency, or infrastructure mandate that no vendor can meet – for example a bespoke permissioned or central-bank ledger.

If that is you, build – and budget for audits and maintenance, not just the launch sprint.

When buy is right

• Tokenization is infrastructure under your real business – you are a fund, an issuer, or a bank shipping a product, not a tokenization vendor.
• You want the compliance and standards maintained by someone accountable for them, so your team ships the raise instead of a security-engineering roadmap.
• You value time-to-first-raise and do not want to carry a specialist team for a system that is not your core product.

The failure mode to avoid

The common mistake is underestimating the recurring cost. A team ships v1, then cannot fund the re-audits, standard migrations, custody integrations, and the “who maintains this in year six?” answer. A half-maintained securities-token stack is worse than not building one – it carries real assets on code no one is watching.

What this means for your decision

Separate two questions you are tempted to merge: can we build it (usually yes) and should we own it for a decade (usually no, unless it is our product). If you buy, buy on the durable criteria – non-custodial architecture, open standards, documented continuity, and a published flat fee (Stobox’s runs $1,499–$6,999 per Raisable window and $1,248 to issue on-chain, never a percentage of the raise) – not on the demo. If you build, commit to the maintenance, not just the launch.

Gene Deyev’s take

In seven-plus years of building this, I have almost never seen building be the mistake. The mistake is building and then under-maintaining it. Anyone can ship a token contract; almost no one budgets for the re-audits, the standard migrations, and the plain question of who owns this in year six. Buy unless tokenization is your product – and whether you build or buy, insist on non-custodial architecture, open standards, and an ownership record that outlives the platform. Those are the things that are painful to bolt on later.

– Gene Deyev, Founder & CEO, Stobox. Author of the Stobox Tokenization Framework and the STV3 protocol; ERC-7943 backer.

Related questions

• How to choose a tokenization platform – the buy-side due-diligence checklist.
• What if the platform disappears? – the continuity question that build-vs-buy is really about.
• What an institutional smart-contract audit covers – the recurring cost people forget to budget (sibling draft T2-23).

Last updated: 2026-09-12.

Reviewed and maintained by Stobox. Last updated September 12, 2026. Educational reference, not legal advice.