Threat Intel Teams Are Consolidating for Correlation, Not Cost Cuts
The 2027 State of Threat Intelligence survey, run by Recorded Future with UserEvidence, polled 768 cybersecurity professionals at enterprises with at least 1,000 employees. Buried in the numbers is a tension that should shape how security leaders plan the next budget cycle. Teams are planning to consolidate their vendors and spend more money at the same time. That's not a contradiction. It's a signal that the goal has shifted from paying less to getting more out of the intelligence they already buy.
Cost is the weakest reason to consolidate
Seventy-nine percent of respondents plan to consolidate threat intelligence vendors within the next year. That's not surprising when the average respondent juggles five different providers. Cost is the obvious guess for the motive, but it ranked last of five drivers, at 5%.
The reasons that ranked higher tell a more useful story:
- Improving correlation across signals (34%): connecting intelligence across threat domains and internal telemetry into one view of risk.
- Bringing siloed tools together (29%): letting different risk owners act on intelligence without friction.
- Simplifying operations (22%): fewer interfaces and handoffs, so teams act sooner.
- Reducing vendor-management overhead (9%): fewer contracts to evaluate and renew.
Budgets aren't shrinking, either. Eighty-three percent plan to increase threat intelligence spending over the next 12 months. Teams are consolidating to extract better results from the spend they already have.
Jason Steer, CISO at Recorded Future, put the bar in plain terms: consolidation only counts "if that provider has the same or better outcomes." That's the right test for anyone running a vendor evaluation this year. A smaller vendor list that dulls detection is a downgrade, no matter how clean the invoice looks.
Breadth is part of the pitch, too. As programs stretch to serve more stakeholders across the enterprise, teams want coverage across the risk domains that matter to each of them. The harder requirement is that broader coverage doesn't add friction, and that the organization can see how threats in different domains actually connect. Plenty of consolidation projects will fail that second test.
Correlation is the top goal, and most teams haven't reached it
Correlation was the leading consolidation driver, yet only 25% of respondents say external intelligence and internal telemetry live in one place. The rest of the picture is messier:
- 52% are partially integrated across some tools.
- 18% rely on analysts to correlate signals manually.
- 5% keep the two in entirely separate workflows.
Poor integration with existing tools and workflows was also the most-cited pain point with current vendors. That's the real complaint behind the consolidation wave. Teams aren't frustrated by having too many feeds. They're frustrated that the feeds don't talk to the systems where decisions get made.
Putting everything into a single tool isn't the only fix, and it isn't always the right one. What tends to matter is whether external intelligence can be correlated with internal signals across the tools a team already runs, without an analyst stitching it together by hand. Collective Insights® takes that approach, pulling detections from a team's existing security tools and enriching them with Recorded Future intelligence.
For the 70% who are partially integrated or correlating manually, the practical exercise is simple. Map every point where an analyst copies a finding from one system into another. Those handoffs are where context gets lost and response slows down.
Confidence in ROI is high, but few teams measure savings
Nearly every respondent (99%) formally measures the value of their threat intelligence program, and 57% say they're very confident they can quantify its return. Look closer, though, and the confidence doesn't match the metrics. More than half of that very confident group (58%) left cost savings out of how they measure value. Across all respondents, just 39% track cost savings, and 31% check whether the actions triggered by intelligence actually worked.
The most common metrics are operational:
- Enhanced threat detection rates: 60%
- Improved response times: 59%
- Fewer incidents: 57%
Those numbers matter to the SOC. They don't translate well to a CFO or a board, which is why security programs keep struggling to justify their budgets even when they're performing well. If detection improved but nobody can say what it saved in hours, dollars, or avoided disruption, the program looks like a cost center. The Impact & Metrics dashboard in the Recorded Future platform is designed to close that gap by showing the hours and dollars saved through use of the platform.
Closed loops need documented priorities
Fixing the measurement gap means connecting intelligence to outcomes. That's what closed-loop operationalization does. Intelligence gets shared automatically with the stakeholders who need it, they act on it, and the results of that action feed back into what gets collected and shared next.
The payoff is that closed-loop programs can answer the questions executives ask: Did the intelligence drive an action? Did that action work? Answering them requires a clear definition of what intelligence is supposed to achieve, and most programs already have that starting point. Sixty-eight percent say formally documented, regularly reviewed Priority Intelligence Requirements (PIRs) govern their work.
If you're building next year's measurement plan, start with the PIRs as the benchmark. Then make closed-loop validation the mechanism for checking outcomes against them. Without that second step, PIRs stay a planning document instead of a scorecard.
Intelligence now feeds AI, not just the other way around
Ninety-eight percent of respondents use AI in their threat intelligence work. The less-discussed flow runs the other direction. Seventy-four percent say threat intelligence already feeds their organization's AI or automation initiatives.
- 44% describe that feed as established, with intelligence regularly flowing into one or more AI or automation workflows.
- 30% describe it as advanced, with structured intelligence serving as a core input across their AI and agentic systems.
- Only 4% don't use threat intelligence to support AI at all.
Comfort with autonomy is climbing fast. Across the tasks the survey assessed (summarization, alert triage, enrichment, false-positive reduction, detection rule generation, prioritization, and enforcement actions), at least 92% of respondents are comfortable giving AI some role.
That's where the earlier findings get urgent. An AI workflow inherits every weakness in the intelligence it consumes. If signals aren't correlated, the model reasons over fragments. If actions aren't validated, the system learns from outcomes nobody has checked. Teams giving AI more autonomy need the correlation and closed-loop work done first, or they're automating their blind spots.
Where programs are headed
Taken together, these findings describe programs that are still investing and still building. Their priorities are connecting intelligence across tools, proving its value in terms the business understands, and extending it into AI-driven workflows. Most teams are partway through that work, and a few are further along than the rest of the field.
The clearest takeaway for the year ahead is that consolidation should be judged by what it does for detection, response, and proof of value, not by what it removes from the budget. The teams that treat it that way will have a stronger foundation when they hand intelligence to AI systems that act on their behalf.