After the AI answer gap was named and debated, a reasonable conclusion formed in many technology organizations: the solution is to rely less on AI and more on proper research. Get back to the analyst reports, the proprietary data, the sourced and defensible intelligence that organizations like IDC have been producing for decades.
It’s a reasonable conclusion. It’s also incomplete.
The problem isn’t that research lacks quality. IDC’s body of work — six decades of proprietary technology intelligence — remains among the most rigorously sourced in the industry. The problem is what happens to research between publication and the moment a decision actually needs it.
Research has a delivery problem
Consider how research actually travels inside most organizations. A report lands in an inbox. Someone reads it, extracts the relevant sections, and summarizes them in a slide deck or a document. That document gets shared with some people but not others. Six months later, someone else in the organization needs the same answer — and re-researches it from scratch because they don’t know the summary exists, or because the original report has been superseded, or because the person who read it has moved to a different team.
This cycle repeats constantly.
IDC’s 2025 Knowledge Management Solutions research, based on a survey of 717 IT and business decision makers in North America, found that nearly half of organizations operate with immature, ad-hoc knowledge management processes — meaning the infrastructure for capturing, sharing, and applying what an organization knows is, for most companies, not functioning as designed. Only one-third of organizations using knowledge management solutions report being satisfied with their effectiveness.
Research that lives in a portal, a PDF, or a folder on someone’s desktop isn’t failing because it’s wrong. It’s failing because the infrastructure to get it to the right person at the right moment doesn’t exist.
The timing problem is as serious as the accuracy problem
There is a timing dimension to research quality that rarely gets discussed. A perfectly accurate market sizing report published in March may be exactly what a team needs for their June board presentation. But if the team building that presentation doesn’t know the report exists, or can’t access it without opening a portal they haven’t used in three months, the report’s accuracy is irrelevant.
The same dynamic plays out in competitive intelligence, vendor evaluation, and technology strategy. The decisions that get made poorly aren’t always made poorly because the underlying research was absent or wrong. They’re made poorly because the research arrived too late, reached the wrong person, or required enough effort to retrieve that the team used whatever they had to hand instead.
“Where it used to take weeks to draw conclusions from hundreds of reports, I can now do that in minutes.” — Mark Terranova, Director, Worldwide Analyst Relations, Kyndryl
The time cost of finding and applying knowledge is measurable and material. In IDC’s 2025 Knowledge Management Solutions research, “reduced time to problem or issue resolution” ranked as the top KPI organizations use to measure the value of their knowledge management investments — ahead of employee satisfaction, customer experience, and cost reduction. The fact that speed-to-answer is the primary value metric for KM confirms that the current model is too slow, and that organizations know it.
Silos make the delivery problem structural
Most enterprise technology organizations aren’t operating with a single, coherent research infrastructure. They’re operating with several: an analyst subscription here, a market intelligence tool there, a set of PDFs that a team member compiled during a project twelve months ago, and a growing number of AI tools that team members have started using because they’re faster than the alternatives, even if they’re less reliable.
IDC’s 2025 Knowledge Management Solutions research identified “numerous unconnected silos of data, unable to collaborate on knowledge” as the top process challenge across nearly every industry surveyed, from financial services to manufacturing to professional services. The technology challenge that ranked first: other systems not integrating well or sharing knowledge bidirectionally. The issue isn’t that knowledge doesn’t exist. It’s that it can’t move.
The fragmentation isn’t just an inefficiency. It’s a risk. When team members default to whatever tool is fastest and most accessible, the quality of the underlying research stops being the deciding factor. Convenience becomes the deciding factor. And convenience, left to its own devices, tends to favor speed over defensibility.
What a better delivery model looks like
Closing the research delivery gap requires rethinking where intelligence lives, not just what intelligence is available. Three things distinguish organizations that close this gap from those that don’t.
First, research needs to be accessible where work happens. Not in a separate system that requires a context switch to access, but embedded in the tools and workflows where decisions are actually being made.
Second, intelligence needs to surface proactively, not just reactively. The most useful research isn’t the research someone finds after they realize they need it. It’s the research that arrives before the question has been fully formed, informed by what the team is working on and what the rest of the industry is paying attention to.
Third, the research base itself has to be trustworthy enough that speed doesn’t come at the cost of defensibility. Proprietary data, cited outputs, and reasoning that can be traced — not summarized from the public internet.
Why this matters now
The AI answer gap created urgency around the accuracy problem. The delivery problem has been present far longer, and it compounds the accuracy problem in ways that are easy to underestimate. An organization with access to excellent research and a broken delivery model will consistently underperform one with a coherent delivery infrastructure — regardless of how good the underlying research is.
IDC’s 2025 Knowledge Management Solutions research made the point directly: knowledge management falters when reduced to passive storage, rather than fostering active discovery, sharing, and application across the organization. The same principle applies to research infrastructure. Having it isn’t enough. It has to move.
IDC Quanta is IDC’s answer to that architecture question. If you want to see what it looks like in practice, the link below takes you there.
