The economics of owning your enrichment contracts

Your enrichment either comes bundled inside a platform subscription, or it runs on data contracts you hold and pay for directly. Which one you are on was usually settled by the platform you bought rather than by a comparison anyone ran.

The bundled version is better for the vendor’s margin. The one you hold yourself is usually better for your cost per meeting. Neither of those facts appears on an invoice, which is where most of the confusion in this category comes from.

This is the math side of that question. A RevOps lead should be able to take the formula below, put their own contract numbers into it, and reach whichever conclusion the numbers support, including the conclusion that a bundled seat is fine for them.

What you are buying when data comes bundled

A bundled data seat is a resale arrangement. The vendor holds contracts with one or more upstream data providers, buys records in volume, and includes some allowance of enrichment inside your subscription price.

Three things follow from that structure, and none of them are hidden or dishonest. They are just consequences of how the arrangement works.

The price you see is blended. You are quoted a subscription, and enrichment sits inside it. There is no line item that says what a record cost, so there is no way to compare that cost against buying the same record directly.

The waterfall order is chosen by the vendor. When a record fails to enrich at the first provider, it falls through to a second, then a third. Each fall-through costs the vendor money. A vendor paying for those calls has a reason to stop the waterfall early, and a buyer paying per usable record has a reason to let it run. Those incentives point in opposite directions.

The data lives inside the subscription. Records enriched under a bundled seat generally do not follow you out of the door in usable form. Whether that matters depends entirely on how much of your target account list you have enriched by the time you consider switching.

Cost per usable record, and why the denominator moves

Here is the whole argument in one line:

Cost per usable record = annual data spend / (records enriched x share that survive your ICP filter)

The price per record is the part vendors compete on. The share that survives your filter is the part that decides the outcome, and it varies far more.

A record is usable when it has a work email that is deliverable, a title specific enough to route, a company that matches your ICP, and a region you sell into. A record with a personal email address and a title of “Manager” cannot be routed to a rep, and it costs you the same as a record that can.

That is why two vendors quoting the same per-record price can differ by a factor of two on the number that matters. Coverage is not uniform. It is thick in North America and Western Europe and thin almost everywhere else, and it degrades fastest exactly where event lists are most interesting, which is at regional summits outside the two markets most data vendors optimise for.

The three costs a bundled seat hides

None of these appear on an invoice, which is why they are worth writing down before a renewal conversation.

The blend. You cannot benchmark a price you cannot see. If enrichment is inside a subscription, you have no way to know whether you are paying above or below what the same volume would cost on a direct contract, and no way to argue about it at renewal. Owning the contract is what gives you the numbers to argue with.

Coverage you cannot inspect. A bundled vendor reports coverage as a single number, if they report it at all. Your business does not experience coverage as a single number. It experiences it as “we sourced four hundred attendees for the APAC summit and could contact ninety of them”. Coverage measured per region is a different report from coverage measured overall, and the second one hides the failure that costs you meetings.

Exit cost. Enriched records under a bundled seat are usually a feature of the subscription. Cancel the subscription and the enrichment goes with it. That turns what looked like a tooling decision into a data decision, and data decisions are much harder to reverse. The switching cost is measured in the months of enrichment you would be abandoning rather than in migration effort.

Running the comparison on your own numbers

This takes an afternoon. The rates below are published ones, so you can check them before you use them. Apollo lists its own overage credits at $0.20 a record. Luminik meters primary-path managed enrichment at $0.10 a record, half that list price, and does not meter gap-fill at all inside the tier allowance: 3,000 records a year on Starter, 12,000 on Growth, 50,000 on Scale. Put your own negotiated rates in place of the list rate, because a signed contract usually beats one.

Take a team enriching 40,000 records a year across their event program.

The point of the table is the third row. If you cannot fill it in, you are comparing invoices rather than prices. The first task is to get the denominator measured, before any vendor decision.

To measure it, take one recent event list. Count how many records went into enrichment, how many came back with a deliverable work email and a routable title, and how many of those matched your ICP. Do that per region rather than in aggregate. Two hours of work gives you a real denominator.

Coverage is regional, and one vendor is never enough

The most reliable finding across event programs is that no single data provider covers everywhere you want to sell. Coverage in the United States and Western Europe is good enough that the question rarely comes up. Coverage in APAC, the Gulf, Latin America and Eastern Europe is thinner, and different providers are thin in different places.

That argues for a waterfall with more than one provider in it, which means holding more than one contract. When you own the contracts you get to order the waterfall by hit rate in the region you are enriching, rather than by whatever order costs the vendor least.

It also means the honest version of the coverage question has a region in it. “What is your coverage” is unanswerable. “What is your coverage for director-level risk and compliance titles at financial institutions in Singapore, Malaysia and Indonesia” is a question with a number behind it, and it is the number that decides whether an APAC event is worth sponsoring. What five event programs taught us about event outbound covers what happens when that number is worse than assumed.

The same logic applies to your model keys

Bring your own key is the same argument pointed at AI features rather than data.

When a product calls a model through your own provider account, four things follow. Token spend appears on a bill you can read, so the cost of a feature is inspectable instead of blended. Your data sits inside a contract your legal team signed, which is a far shorter conversation in a security review than explaining a vendor’s subprocessor list. You control the model version, so a provider changing a default does not silently change your output quality. And when a cheaper or better model arrives, you can move to it without waiting for a vendor roadmap.

The counterargument is real: your own key means your own rate limits, your own quota management, and your own bill to watch. For a small team that is friction with no benefit. For a team with a security review, a finance partner who asks about AI spend, or a compliance requirement about where inference runs, it is often what gets the purchase cleared.

When bundled is the right answer

Owning contracts is not free. It costs procurement time, at least one minimum commitment, and somebody’s attention to manage credits across vendors. There is a size below which none of that is worth it.

Bundled is the better call when you run one or two events a year, when your enrichment volume sits below any vendor’s minimum commitment, when you sell into one region where coverage is uniformly good, and when the hours to manage a second contract are not available. In that situation the blended price is genuinely competitive and the flexibility you are giving up is flexibility you would not use.

The math flips at the point where coverage gaps start costing you meetings, where enrichment volume clears a minimum commitment on its own, or where somebody asks a question about data cost that you cannot answer. That threshold tends to arrive earlier than planned for, usually the first time a regional event returns a list that largely fails to enrich.

If you want the positioning argument rather than the arithmetic, why we refused to own your enrichment explains the choice we made and why. The mechanics of how sourced attendees get enriched before an event are in pre-event intelligence, and the vendor connections themselves are listed on integrations.

Frequently asked questions

What does BYOV mean in sales and marketing software?

BYOV stands for bring your own vendor. The platform runs on the data contracts you already hold rather than reselling data through its own licence. You keep the commercial relationship with Apollo, Clay, ZoomInfo or whoever else you use, you see the per-vendor price you negotiated, and the data stays yours if you change platforms. The alternative is a bundled seat where enrichment is included at a blended price you cannot break apart.

How do I calculate cost per enriched record?

Divide your annual data spend by the number of records you enriched, then divide again by the share of those records that survived your ICP filter and were worth contacting. The second division is the one people skip, and it is usually where two vendors that look identically priced turn out to differ by a factor of two or more.

Is bundled enrichment ever cheaper than owning your own contracts?

Yes, for a team below the volume where a standalone data contract makes sense. If you run one or two events a year and enrich a few thousand records, a bundled seat avoids a minimum commitment you would not otherwise hit, and the blended price is genuinely competitive. The math flips when regional coverage gaps start costing you meetings, because that cost never appears on the invoice.

What is BYOK and why does it matter for AI features?

BYOK means bring your own key: the product calls the AI model through your own provider account rather than through the vendor’s. You see the token spend on your own bill, you control which model version runs, your data stays inside a contract you signed, and you can move to a cheaper or better model without waiting for the vendor’s roadmap. It is the same argument as BYOV applied to inference instead of data.

What questions should I ask a vendor about bundled data pricing?

Ask what the per-record cost is when broken out of the bundle, what the coverage rate is in each region you sell into, which upstream providers are in the waterfall and in what order, what happens to the enriched records if you leave, and whether unused credits expire with the contract. A vendor who cannot answer the coverage question by region is telling you they have not measured it.