Here’s a scenario I hear from resource managers constantly, in one form or another:
The pipeline looks promising. Sales is telling you there’s meaningful work coming. But meaningful isn’t the same as certain, and every time you’ve moved early on the basis of pipeline, something shifted. So you wait. You staff reactively. The work comes in. You scramble.
And then someone in the executive suite asks why you weren’t ready.
This is the cruelest part of the reactive resourcing trap: the people inside it aren’t there because they’re bad at their jobs. They’re there because they never had the intelligence required to make a credible, defensible case for moving sooner. The problem isn’t judgment. It’s information.
That problem was already difficult to solve. Now the variables are multiplying.
AI is Making the Case Harder, Not Easier
Building the business case for hiring ahead of demand has always required connecting unreliable dots. What does the pipeline actually signal about future capacity needs? What skills do we need, and do we have them? Who will be overloaded in six months, and who will have room? How confident are we, really?
You are essentially being asked to produce a recipe for what the workforce should look like six, twelve, or eighteen months from now, using ingredients that refuse to sit still.
The first is demand. No, not the optimistic number sales puts in the CRM in February. Demand-adjusted, probability-weighted pipeline insight that tells you with enough fidelity to actually plan against. The gap between “we have opportunities in the system” and “here’s what we can credibly expect to be staffing in Q3” is where most workforce planning falls apart. Closing that gap requires bringing together pipeline data and resource data in a way that produces a signal you’d actually stake a hiring recommendation on.
The second is human capacity: Who will be available, and what can they actually do? That sounds like the easier, more predictable half of the equation. It most certainly is not. Availability changes as projects expand, contract, accelerate, stall, and collide. Skills inventories begin decaying almost as soon as they are completed. Valuable experience accumulates inside project histories, delivery artifacts, and people’s heads without ever finding its way back into a searchable profile.
The third is an honest accounting of agent capacity. Not a vague gesture toward AI doing “some of the work,” but a genuine model of where agents are already contributing to delivery, what tasks they’re handling, how that translates to effective capacity, and where the limits are. The organizations making smart hiring decisions right now are the ones that have stopped treating AI capacity as an asterisk and started treating it as a variable that belongs in the model.
Resource managers have always been asked to compare an uncertain view of future demand with an incomplete view of the workforce available to meet it. And now there is a third ingredient to reckon with: agentic capacity.
Not a vague gesture toward AI doing “some of the work,” but a genuine model of where agents are already contributing to delivery, what tasks they’re handling, how that translates to effective capacity, and where the limits are. What work will AI agents perform? How consistently can they perform it? How much human time will they actually save? Where will they augment someone’s expertise, and where will they genuinely reduce the amount of human capacity required?
Those questions now sit underneath another one that resource managers should expect to hear in nearly every hiring conversation:
“Wait, don’t we need fewer people here if agents are going to do this work?”
The executives asking that question aren’t wrong to ask. According to Kantata’s State of the Professional Services Industry research, 89% of PS leaders say future revenue growth will depend more on scaling AI than scaling headcount. That belief is shaping how leadership evaluates every workforce investment.
Resource managers who can’t speak to the agent question aren’t just missing a piece of the puzzle. They’re missing the piece leadership is most focused on.
Assembling the Ideal Team Isn’t the Same as Building It
A lot of the conversation about AI in resource management focuses on making it faster and easier to assemble the ideal team. That is an important and valuable use case. AI can analyze more variables, identify relevant experience that a resource manager might not have known about, and surface stronger matches far faster than a manual search.
But even the most intelligent staffing system cannot put the right person with the right skills on a project if you never hired (or developed) that person in the first place.
That’s where the gap between what organizations are doing with AI today and what they ultimately expect from Intelligent Resource Management becomes revealing. In the Resource Management Institute’s Q3 2026 research, 85% of respondents from externally focused services organizations identified reducing manual effort as a primary driver for adopting AI in resource management. By comparison, 35% selected enhancing strategic workforce planning, and 35% selected improving forecast accuracy.
That near-term focus makes a lot of sense. Resource management contains no shortage of manual work worth eliminating. If AI can turn what used to be a 40-hour skills-inventory exercise into a far faster (and potentially much richer) analysis, that is meaningful progress.
But accelerating the work resource managers already do is not the same as giving them the foresight to shape what the workforce should become. Encouragingly, that more strategic ambition is exactly where RM leaders believe Intelligent Resource Management is headed. When the RMI asked what outcome would define Intelligent RM, the most common answers were predictive and scenario-based planning, at 32%, and AI-assisted strategic workforce planning, at 29%. Only 7% selected fully automated scheduling and allocation.
That’s a really meaningful distinction. The destination isn’t just a machine that makes staffing assignments faster. It’s a resource management function capable of seeing further ahead, testing more possible futures, and helping the business make better decisions before its options narrow.
This Takes More Than a Time-Saving Agent
Making that vision real will require more than deploying an agent to automate pieces of today’s staffing process. It requires agentic business intelligence: a connected layer of intelligence that can investigate what is happening across the business, bring the relevant evidence together, test possible responses, and help turn the result into action.
The difference matters. A time-saving agent might find qualified people more quickly. Agentic business intelligence should help you determine whether the qualified people you will need six months from now exist at all.
It should be able to look across pipeline, project, resource, skills, and financial data to identify which demand signals are credible and what they imply for future capacity. It should uncover experience hidden in project histories rather than relying entirely on self-reported skills profiles. And it should bring agent usage, quality, cost, and outcomes into the same capacity model as human work.
Most importantly, it should let a resource manager interrogate the assumptions beneath a workforce plan rather than simply accept a forecast.
Imagine being able to model an expected increase in demand for a particular service and ask:
- Which skills will become constrained first?
- Can targeted development close the gap in time?
- What portion of the work could agents credibly absorb?
- What happens to utilization, margin, and delivery risk under different assumptions?
- At what point does hiring become the least risky—or most profitable—decision?
Answering those questions once would be useful. Being able to continue the conversation—investigate why the model reached an answer, generate a dashboard to make the case, test an alternative scenario, and research the implications without assembling a small fleet of analysts—is where the value becomes transformative.
The goal isn’t to prove that every anticipated gap requires another hire. Nor is it to accept that every promised AI efficiency eliminates one. The goal is to give leadership a clear view of the tradeoffs and enough confidence to act while there is still time to hire, develop, contract, automate, or redesign the work.
Because by the time a project needs a particular combination of expertise, it may already be too late to create it.
The Intelligence Gap Is the Strategy Gap
What I keep coming back to is that the resource management challenges that feel operational are almost always intelligence problems underneath.
- Why do firms hire reactively? Because the pipeline signal isn’t trusted enough to move earlier.
- Why don’t they know what skills they’ll need? Because skills data is stale and incomplete.
- Why can’t they account for agent capacity? Because they’ve never had a way to model it systematically.
The solution isn’t better instincts. It’s better information, surfaced faster, in a form that actually supports a decision. The resource managers who are going to lead their organizations through this next period aren’t necessarily the ones with the best gut feel for staffing. They’re the ones who build the intelligence infrastructure to make that gut feel testable, defensible, and scalable — with pipeline forecasting they would stake a recommendation on, skills intelligence grounded in actual experience, scenario modeling that makes the tradeoffs visible, and an honest accounting of where agents do and do not change the capacity equation.
You’re not going to be able to build the 2028 workforce on 2018 intelligence. The variables are different. The stakes are higher. And the executives asking the questions aren’t going to accept “we don’t have visibility into that” as an answer.
Kantata will tackle this challenge head-on on September 24 in the RMI Connect session Beyond AI Staffing: Harnessing Agentic BI to Build the Case for Where Your Workforce Goes Next. Melissa Korzun, Kantata’s Vice President of Industry Solutions, will demonstrate how agentic business intelligence can turn natural-language conversation into the forecasting, skills intelligence, dashboards, scenario modeling, and deep research resource managers need to build a stronger case for what their workforce needs to look like.