Your intent data platform says an account is surging, but your SDR finds no project, no budget, and no reason to talk. Across North American SaaS teams, that gap burns rep capacity and makes marketing look disconnected from the buying process.
The fix is a signal-to-action operating model: match each signal to the decision it can actually support, then expire it before it becomes false confidence.
Key takeaways
- Treat intent data as evidence of research, not proof of a funded project or permission to contact someone.
- Separate account fit, signal strength, buying stage, and contact eligibility before routing work to sales.
- Give signals an expiration rule. An old research spike should not keep an account permanently “hot.”
- Route accounts differently based on customer status, opportunity stage, territory, and the number of independent signals.
- Measure whether signal-driven actions improve sales acceptance and opportunity progression, not just list size.
- Design US and Canadian outreach rules separately. CAN-SPAM and CASL are not interchangeable.
Why North American intent data programs produce false urgency
The most expensive mistake is collapsing three different statements into one: someone researched a topic, the company may have a problem, and a specific person wants a sales conversation.
Only the first statement may be supported by the data.
A US software company researching cloud security could be evaluating vendors, preparing an internal presentation, or helping a customer. A Canadian systems integrator may consume the same material because its consultants need to understand the category. Neither behavior automatically creates a sales opportunity.
Account identity is not buying authority
Third-party intent commonly arrives at the account level. Depending on the provider and collection method, account matching can be affected by remote work, shared networks, corporate structures, and other identity limitations.
That matters when a vendor maps research to a parent company while the relevant budget sits inside a subsidiary.
Before a rep acts, ask:
- Which entity does the signal describe?
- Is that entity in our ideal customer profile?
- Is the topic specific to the problem we solve?
- Do we know a relevant contact, or only a company name?
- What evidence would justify the next step?
A 6sense or ZoomInfo signal can help prioritize investigation. It cannot establish buying authority on its own.
Dark social makes the picture incomplete
Buyers discuss vendors in private Slack communities, peer messages, partner calls, and conversations around SaaStr or Dreamforce. Much of that activity never enters your attribution system.
The practical implication is not to invent visibility. It is to accept that observed intent is a partial view.
No recorded signal does not mean no demand. A recorded signal does not mean sales readiness.
Keep a route for high-fit accounts with weak observable intent. Otherwise, your targeting model will favor companies whose research happens to be measurable, rather than companies most likely to buy.
Build a signal-to-action framework, not another score
A single intent score is convenient for dashboards and dangerous for decisions. It can blend excellent fit with stale research, or poor fit with intense activity.
Use a four-part framework: Fit, Evidence, Freshness, Action, or FEFA. Keep the components visible in the CRM instead of burying them inside a composite number.
Fit: would we pursue this account without the signal?
Start with firmographics, use case, technical environment, and commercial constraints.
For a hypothetical $60,000 annual contract, a 40-person company may be outside your viable segment unless it has an unusually strong use case. A large enterprise may also be a poor fit if your product cannot meet its security requirements.
Intent should reorder eligible accounts. It should not erase your ICP.
Review fit when the business changes. A move from SMB to midmarket, a new integration, or a change in pricing can make last quarter’s account model actively misleading.
Evidence: what does the behavior actually establish?
Group signals by what they support:
- Category research: broad interest in a problem or market.
- Solution evaluation: comparison content, integration research, or product-specific questions.
- Commercial engagement: a direct pricing question, demo request, or procurement discussion.
- Customer change: activity that may indicate expansion, implementation friction, or competitive evaluation.
These are evidence categories, not a universal ladder. A comparison-page visit could come from a competitor. A customer researching alternatives could be doing routine market monitoring.
Confidence improves when independent evidence converges. Two vendors reporting activity from the same underlying publisher network may not represent two independent signals. Ask providers about overlap.
Freshness: when should the evidence stop influencing priority?
Use different clocks for different behaviors.
As illustrative starting rules, a direct demo request could require same-business-day handling, a cluster of evaluation activity could trigger review within three business days, and a broad topic surge could remain in a monitoring queue for two weeks. These are operating choices to test, not industry benchmarks.
Expiration should reduce priority unless new evidence appears. It should not delete the historical record.
A long enterprise sales cycle does not make every signal durable. The research project may last months; the usefulness of a particular outreach trigger may last days.
Action: what changes because we know this?
Every signal category needs a permitted action.
Broad research might change an account’s advertising audience or content sequence. Multiple evaluation signals from a high-fit account might create an SDR research task. A direct hand-raise should follow your inbound response process.
If a signal changes nothing, stop paying to move it between systems.
Route North American accounts by context, not heat
The same signal should produce different actions depending on the account’s relationship with your business. A prospect, an open opportunity, and a customer are not three versions of the same outbound lead.
Use a routing hierarchy
A workable hierarchy is:
- Open opportunity: notify the account executive and attach the evidence.
- Current customer: send the signal to the account manager or customer success owner.
- Named target account: route to the assigned account team.
- Unassigned ICP account: create a research task in the appropriate territory.
- Outside ICP or restricted: suppress sales activation and retain only permitted analysis.
Apply this hierarchy before assigning urgency. Otherwise, marketing creates an SDR task against an account already negotiating with an AE.
Salesforce and HubSpot workflows can support the process, but they need clean ownership and lifecycle fields. Automation cannot compensate for two teams claiming the same account.
Match the message to the evidence
A weak message says, “We noticed your company researching data warehouses.”
That may feel intrusive, and the rep may not know whether it is accurate at the individual level.
A better message introduces a relevant business hypothesis: “When analytics teams consolidate warehouses, migration sequencing often becomes harder than vendor selection. Is that part of your remit?”
The second approach still needs genuine account relevance. It simply avoids pretending that account-level observation proves personal behavior.
For teams building intent-based outreach workflows, the core deliverable should be this connection between evidence, account context, and a defensible reason to engage.
Respect buying calendars without inventing urgency
North American technology buyers often face quarter-end approval pressure, but “Q4” does not always mean October through December. A US public software company may have a fiscal year that differs from the calendar year, and Canadian buyers have their own budget cycles.
Store known fiscal timing as account context. Do not infer an available budget from a research spike near December.
For enterprise deals, security review, legal, procurement, and stakeholder alignment can outlast the initial research burst. Intent can tell an account team where to investigate; it cannot compress those processes by itself.
Make the operating model work in your CRM
The implementation should be boring enough to survive staff turnover.
Start with a narrow segment, one signal source, and a small set of actions. Adding more feeds before sales trusts the first one usually creates a larger rejection queue.
Define a minimum signal record
Each activated signal should carry:
- Account identifier and matched entity.
- Source, topic or behavior, and observation date.
- Fit classification and evidence category.
- Expiration date and suppression status.
- Assigned owner, recommended action, and disposition.
Keep observation time separate from ingestion time. A signal imported this morning may describe behavior from last week.
Require readable evidence. “Score increased to 87” is not enough for a rep deciding whether to spend 20 minutes researching an account.
Pilot around capacity
Consider an illustrative pilot with 200 ICP accounts, two SDRs, and a six-week operating window.
If each rep can thoroughly investigate eight new accounts per week alongside existing responsibilities, the workflow has capacity for 16 account reviews weekly. Sending 80 “hot” accounts does not create more capacity. It creates selective neglect.
Cap the queue. Prioritize the strongest eligible evidence, then monitor what expires before review.
Six weeks can reveal routing defects, response patterns, and workload problems. It may not establish a revenue outcome when the normal sales cycle runs several months.
Fix the denominator before judging results
Do not label every account surge an MQL. That changes the meaning of your funnel and makes historical comparisons unreliable.
Suppose, purely as an example, 100 person-level MQLs produce 20 SQLs. That is a 20% MQL-to-SQL conversion rate. Adding 100 account alerts to the MQL denominator without changing the SQL definition would push the reported rate to 10%, even though the original cohort performed identically.
Track account signals separately from person-level lifecycle stages.
Useful operating measures include signal-to-review time, sales acceptance, rejection reasons, conversation quality, and opportunity progression. Examine conversion by segment and evidence type, using consistent definitions and comparable observation windows.
Most importantly, require rejection feedback. “Wrong entity,” “existing opportunity,” and “no relevant use case” each point to a different fix.
Put privacy and governance inside activation
Compliance is not a footer attached after the audience is built. It determines which records can enter which workflow.
Intent data does not create consent. It also does not remove obligations attached to the underlying personal information or the outreach channel.
Separate US and Canadian requirements
For US commercial email, CAN-SPAM applies to B2B messages as well as consumer messages. Requirements include accurate sender information, nondeceptive subject lines, a valid physical postal address, an appropriate opt-out mechanism, and honoring opt-out requests within 10 business days.
For businesses subject to the CCPA, as amended by the CPRA, relevant obligations can include notice, consumer rights, and opt-outs of sale or sharing. California business-contact information is not categorically exempt simply because the context is B2B.
In Canada, CASL generally requires consent for commercial electronic messages, along with identification and unsubscribe requirements, subject to specific exceptions. Do not assume that a publicly listed business email address or a B2B label automatically permits outreach.
Have counsel approve the channel-specific rules, especially for purchased contacts and cross-border campaigns.
Audit providers and internal access
Ask vendors to explain collection sources, permitted uses, identity methods, retention, and how applicable privacy choices propagate through the data supply chain. Review contracts and documentation rather than accepting “compliant data” as a complete answer.
Internally, maintain suppression lists, restrict access, and retain only what the workflow needs.
Reps should receive enough context to act responsibly, not a sprawling record of inferred browsing behavior. The goal is better judgment, not surveillance theater.
FAQ
What is the difference between intent data and a lead?
Intent data is evidence that a person or account may be researching a topic or solution. A lead is a record managed through your marketing or sales process; it may or may not contain meaningful intent. Account-level research should not automatically become a person-level qualified lead.
How quickly should sales act on intent data?
Direct hand-raises deserve faster handling than broad research signals. Set response rules by evidence type, source latency, staffing, and buyer expectations. Measure whether faster action improves outcomes rather than assuming every alert requires an immediate call.
Is first-party intent better than third-party intent?
First-party behavior usually gives you clearer context about interactions with your own properties, but it covers only part of the buying journey. Third-party data can reveal research outside your channels, often with less individual-level certainty. Use each for decisions its evidence can support.
Can intent data identify an entire buying committee?
Not reliably on its own. Account-level activity does not establish which individuals control budget, evaluate security, or approve procurement. Build committee coverage through direct conversations, CRM history, role research, and confirmed stakeholder relationships.
What should we ask 6sense or ZoomInfo before buying?
Ask about source coverage, account matching, update frequency, topic definitions, underlying data overlap, and permitted activation uses for the specific product. Request a sample against your own ICP, including difficult subsidiary structures and Canadian accounts. Test whether the output changes a useful decision before expanding deployment.
The bottom line
Intent data earns its place when it changes the right action for the right account. Keep fit separate from activity, make evidence readable, expire stale signals, and protect ownership. A smaller queue that sales trusts beats a dashboard full of unexplained urgency.
If your signals are multiplying faster than useful sales conversations, talk to the Tech Talks Media team about building a signal-to-action workflow around your ICP, sales capacity, and US and Canadian activation requirements.
