The signals that matter put a date on a vendor decision. The main ones are a network upgrade or testnet shutdown, a new chain deployment and a deposit jump. A governance post or hire that names your function also counts. Posts and listings without a date rarely lead to a purchase.

Which signals put a date on the decision?

Six signal types do, and each points to a different buyer:

  • A network upgrade forces client and tooling updates. Watch core developer blogs. Contact staking operators, node teams and block builders.
  • A testnet shutdown moves test infrastructure to a new network. Watch foundation blogs. Contact infrastructure leads.
  • A new chain deployment forces new RPC, oracle, indexing and bridge choices. Watch governance forums. Contact the CTO or head of engineering.
  • A deposit or volume jump adds load and value at risk. Watch DefiLlama TVL. Contact engineering or the risk lead.
  • A governance post that names a function needs a vote. Contact the proposal author, delegates and risk service providers.

The sixth is an infrastructure engineering hire, which opens a build-or-buy choice. Watch careers pages, and contact the CTO or hiring manager.

How do network upgrades create demand?

An upgrade gives every operator on the network the same deadline. Ethereum's Glamsterdam upgrade was scheduled to activate on the Sepolia testnet on 6 October 2026. Hoodi and mainnet dates were not yet set (Ethereum Foundation, 28 September 2026).

Glamsterdam includes enshrined proposer-builder separation (ePBS, EIP-7732), so the protocol handles the payment from builder to proposer. The Foundation says this reduces reliance on trusted middleware. It told staking operators to review new ePBS duties and tooling changes.

So every Ethereum staking operator has a dated reason to review its MEV, relay, builder or validator software.

Testnet shutdowns work the same way at smaller scale. On 1 September 2025, the Foundation said most Holešky validators would shut down two weeks after Fusaka finalised there. Operators had to move to Hoodi, and app developers to Sepolia (Ethereum Foundation, September 2025).

Write when the date is announced. Teams test changes on the testnet fork first, so the vendor choice comes before that fork.

Why does a deployment to a new chain matter?

Each deployment repeats the vendor decision. A lending protocol live on four chains chose RPC, oracle and indexing providers four times. It chooses again on the fifth.

Governance forums show these decisions early. A deployment proposal often names the planned oracle, bridge and data providers. If it names an incumbent in your category, the slot is taken. If it names none, or asks for comments, the slot is open and the author is your contact.

What does a jump in deposits tell an infrastructure seller?

More deposits mean more data requests, more liquidations and more value behind each oracle update. Our catalyst monitor logged 6,069 buying signals from 26 July to 7 October 2026. They included 566 TVL moves and 40 new DeFi listings (our data, October 2026).

Treat a rise as a reason for a capacity or risk review. Write to engineering or risk with a figure from your product at that size. If you sell on price, a sustained fall also counts, because a protocol that earns less reviews its vendor bills.

A new listing alone says only that the protocol exists. Pair it with a deployment date or a governance post before you write.

What does an infrastructure engineering hire mean?

A job post for a node, site reliability or MEV engineer means the team plans to run that function itself. That opens a build-or-buy window, which closes once the hire starts.

In general B2B outbound we drop engineering roles, as covered in which buying signals predict a B2B purchase. For an infrastructure seller the same post is a direct signal. Write to the hiring manager with the loaded cost of the hire and the time to production, set against your price and setup time.

How do you rank the signals you collect?

Score each signal before anyone writes:

  1. Match the signal to exactly one company, and confirm the domain. Skip it if the match fails.
  2. Add 2 points for a date in the next 90 days. Add 2 points if it names your function.
  3. Add 1 point if you can name the decision owner.
  4. Subtract 1 point if a proposal or the chain's docs name an incumbent in your category.
  5. Write within a week to signals that score 4 or more. Log the rest until a second signal arrives.

Step 1 does the most work. We caught two namesake errors in our own matching, where a signal matched a different firm with the same name (our data, October 2026). An email about an upgrade the reader never planned costs more than a skipped signal.

What is the decision rule?

Act on a signal when it names one company, a date and your function. Write to the owner of that function, and open with the dated fact. Send before the testnet fork, the deployment vote or the hire's start date.

To see the protocols in your category with a live signal, start with five named buyers, free. Each comes with the dated signal behind it.