The Next Scarcity in B2B Is Permission

The B2B data industry is getting remarkably good at making contact information abundant. It is much less clear that this is making customers easier to win.

A revenue team can assemble a target market, enrich the contacts, generate copy, and automate follow-up with increasingly little manual work. Clay’s integration with lemlist, for example, can move lead data and personalized copy directly into email campaigns.[1] The technical distance between identifying a person and putting a message in their inbox keeps shrinking.

The distance between sending that message and earning a conversation is another matter.

That distinction should reshape how we think about the future of B2B data. My view is that undifferentiated contact data will face growing pressure on its utility, even as its supply expands. The advantage will move toward understanding a specific business, recognizing a relevant moment, and earning permission to participate in its decision.

Cheap data can obscure expensive failure. A campaign may look efficient when measured by cost per record or cost per thousand sends, while consuming sales time, damaging sender reputation, and producing very few willing buyers. Adding mailboxes makes the activity easier to expand. It leaves the underlying reason to respond untouched.

One benchmark illustrates the gap. Belkins’ 2026 study reported a 0.45% reply rate across approximately 7.5 million cold emails sent in its own client campaigns during 2025. That is one provider’s dataset, rather than a universal market average, and a reply is still a long way from a qualified opportunity.[2]

There is a collective problem here. When one team increases its sending volume, it may gain a temporary advantage. When every team does it, each buyer receives more messages while retaining the same number of hours in the day. The sender’s marginal cost falls; the recipient absorbs more of the burden of sorting, ignoring, and blocking.

My concern is that the industry is treating this as an infrastructure problem because infrastructure is easier to buy than relevance is to create.

Privacy law is adding another constraint, and 2027 brings concrete changes. California’s SB 923 takes effect on January 1, extending deletion rights to non-exempt personal information obtained from third parties. California’s privacy regulator also says AB 883 will require data brokers to process DROP deletion requests at least every 30 days beginning that same day.[3][4]

Oklahoma and Louisiana also bring new consumer privacy laws into effect on January 1, 2027. Both exclude people acting in a commercial or employment context from their definitions of a consumer.[5][6] These laws should not be presented as a universal B2B opt-in mandate. California’s former B2B exemption, meanwhile, expired at the end of 2022.[7] Geography, purpose, and the specific data involved matter.

The strategic implication I draw is that acquiring information creates an ongoing responsibility. A record can require deletion, suppression, or a change in how it is used. Its availability in a purchased file cannot establish every permission a subsequent campaign might need.

Legal scrutiny of major platforms reinforces the need for that distinction. In October 2024, the Court of Justice of the European Union ruled that Meta could not aggregate and process personal data for targeted advertising without limits on time or distinctions between types of data.[8] Separately, LinkedIn’s 2025 resolution of its lawsuit against Proxycurl illustrates how legal disputes can reach the supply of professional data itself.[9] These are different legal contexts, but both challenge the assumption that accessible information is an unrestricted commercial resource.

For revenue leaders, I think the response begins with a more useful relationship between TAM, intent, and consent.

TAM establishes which companies could become customers. Intent helps prioritize where a relevant need may be emerging. Useful content gives those buyers a reason to engage. Consent records the choice a person actually makes about that engagement and what can follow.

Each step adds something the previous step cannot supply.

A company entering a new market, hiring a security leader, opening locations, or changing its technology stack may be revealing a problem worth understanding. Those signals can support a buying hypothesis. They do not prove an active budget, identify every stakeholder, or establish that anyone wants to hear from your company.

The work is to connect the observable change to a credible business consequence, then offer something that helps the buyer think or act.

Imagine an HR software company identifying an employer that has expanded into three new states and begun hiring payroll specialists. That combination could point to growing operational complexity. A useful next move might be a practical briefing on the payroll challenges of that expansion, a relevant benchmark, or an invitation to a discussion with peers facing the same transition.

The prospect should receive useful insight before being asked to make a larger commitment. They can choose to request an assessment, subscribe to a clearly described series, or speak with a specialist. Each choice should be represented accurately. Downloading a briefing does not establish willingness to take a sales call.

This also requires care with the word “consent.” An opportunity to opt out and affirmative opt-in are different choices. Permission needs to be specific enough to reflect who may contact the person, about what, and through which channel, with a clear way to withdraw it. It cannot be treated as a transferable property attached permanently to an email address.

That is a much more demanding data model than a spreadsheet of names and titles. It connects account research, the evidence behind a buying hypothesis, the content offered, the person’s expressed interests, and the boundaries of the relationship.

It also changes the meaning of personalization. Referencing a recent post or congratulating someone on a promotion may show that a system found information about them. Demonstrating an understanding of the decision they face gives them a reason to care.

This is account-based marketing taken seriously: company by company, problem by problem, and person by person. Teams can reuse expertise, research methods, and strong content. The relevance of an offer still has to survive contact with each account’s circumstances. AI can make that research faster; someone remains accountable for whether the resulting message is useful.

I believe this creates a structural challenge for providers whose economics depend on selling access to the same prebuilt database. A standardized record can describe a person’s job. It cannot, by itself, establish why that person would welcome a conversation with a particular seller today. Adding a generic intent score leaves much of that work unresolved.

The opportunity is in bespoke data and custom insight developed alongside the business using them. That may bring a managed data partner, an agency, researchers, and the customer’s own marketing and sales teams into the same process. They need to understand the solution, the market, the buying problem, and the value they can offer before asking for attention.

This is the direction I believe the data industry should take, and the case for a bespoke approach at LeadGenius. The value of data grows when it helps a team make a better decision about where to invest effort and how to earn the next conversation.

The measurement should follow. Cost per record and cost per send tell us about inputs. Revenue leaders should also examine the cost of a willing, qualified conversation, the opportunities those conversations create, and the preferences honored along the way.

We may be entering an era in which contact data is abundant and permission is scarce. The companies that create useful insight around a real buying problem will have something worth inviting a prospect into.

That is where I would place the next bet in B2B data. LinkedIn version B2B data is getting cheaper. A willing buyer’s attention is getting harder to earn.

I think that gap will define the next chapter of this industry.

When a campaign underperforms, it is tempting to add contacts, mailboxes, and follow-ups. All three are easier to scale than a compelling reason to respond.

But the buyer still has the same number of hours in the day.

2027 adds another reason to rethink the model. California is expanding deletion rights to non-exempt personal information acquired from third parties and shortening the DROP processing cycle for data brokers to 30 days.[3][4]

The rules vary by jurisdiction. They do not amount to a blanket ban on B2B outreach. They do make ongoing control over data harder to ignore.

Here is the model I would build around:

TAM tells you who could buy.

Intent helps you identify where a relevant need may be emerging.

Useful content gives someone a reason to engage.

Consent records what that person actually agrees to next.

A hiring spike or international expansion can justify research. A practical briefing, benchmark, or assessment can earn interest. A clear invitation lets the buyer choose the next step.

And an opt-out is different from affirmative opt-in. Both choices need to be respected.

This takes custom insight, a clear point of view, and an understanding of each account’s circumstances. It is ABM with the work left in.

That is why I believe bespoke data has the stronger future. A prebuilt database can supply a contact. The value comes from understanding the business well enough to earn a conversation.

At LeadGenius, that is the direction I believe our industry should pursue.

What would change in your GTM strategy if the primary metric became cost per willing, qualified conversation?