AI adoption isn’t the problem for most partner teams. It’s letting AI actually change how the work gets done.
That’s the gap that PartnerStack’s CMO Tyler Calder lays out when he talks about the future of partner management, and the broader data on AI in business backs him up.
According to McKinsey & Company’s 2026 survey research on companies’ use of AI, 89 per cent of organizations report regular AI use in at least one business function. But scaling AI use — that is, actually embedding AI into how work runs — is a different story: only 44 per cent say AI is scaling across their enterprise, up from 38 per cent last year. Push further into agentic AI, and adoption thins out even more, with roughly one in five organizations now scaling an agentic system.
Only six per cent of organizations report real financial payoff from AI — what the McKinsey & Company report calls “AI high performers” — and that number hasn’t budged since last year.
That's where Calder’s point of reference for partner teams comes in: the AI maturity curve, from AI-assisted to AI-native. In this article, we’ll explore where most partner teams actually are on the curve, and what it takes to reach the next stage.
From asking AI to operating with it
Partner teams have only been working with AI for a handful of years, but if you look back at how the technology has evolved in that time, the ask has shifted year to year.
Calder tracks it like this:
- In 2023, we asked AI to search and synthesize: “Explain this to me.”
- In 2024, we asked it to generate content: “Write this email.”
- In 2025, we asked it to make recommendations, more like a copilot: “Analyze this and tell me what I should do.”
In 2026, the ask has shifted again. Now, we’re asking AI to be more of what Calder frames as an operating layer: “Run this workflow, and bring me in when judgment is required.”
That’s the real shift — from a tool that helps you work, to a system that works without you needing to prompt it — and it’s the same shift most partner teams still haven’t made.
See more: Expert strategies for leveraging AI for partnerships.
The four stages of AI maturity
Calder maps these as four stages on an AI maturity curve, where the value compounds as your team moves up and to the right along it. Here’s each stage in more detail:
Assistant: AKA you ask, it responds
This is AI as a search bar, but with better manners. In this stage, you ask an AI tool a question or give it a task. Then it responds. Then, you take it from there.
In partnerships, this stage could look like asking AI to first-draft a partner email or summarize a contract clause that you’ll then review line by line.
You’re at this stage if AI is a tab you open, not a system running anything on its own.
Copilot: AKA it works alongside you
Here, AI starts contributing in real time, not just answering one-off prompts.
In the partnerships realm, think AI drafting a partner scorecard while you adjust the weighting as you go, or helping you build a QBR deck section by section.
You’re at this stage if AI is speeding up work that you’re still fully driving.
Workflow: AKA it runs the process, end to end
This is where AI stops waiting for a prompt and starts running a process on its own — for a defined set of cases — without a human driving every step.
Automated deal-registration validation or automated partner onboarding sequences are good examples for partner managers: once the rules are set, AI carries them out. You can see this level of AI for partnerships in PartnerStack’s AI Recruit Agent, which keeps partner sourcing running without someone prompting it each time.
You’re at this stage if a defined process runs from start to finish without you checking in.
Operating Layer: AKA it runs parts of the program, and knows when to bring you in
At this stage, AI isn’t waiting for a trigger: it’s running continuously, watching for what needs attention and only pulling you in when your judgment is required.
Intelligent lead scoring could fit well at is an example of this stage: AI continuously re-ranks leads as new signals come in, surfacing only the ones that need a human’s attention.
Few partner teams are operating here yet. It’s less about a specific tool and more about a shift in what your role becomes: from doing the work to reviewing what AI surfaces.
You might also like: AI visibility strategy: How partnerships leaders can operationalize it and own the outcome.
Stuck between Assistant and Copilot
Most partner teams today, Calder estimates, are likely still near the start of the curve. Essentially, AI is helping people work faster on individual tasks, but it’s rarely running anything on its own yet.
That gap matters because of the potential that’s on the other side of it. According to PartnerStack x Wynter’s The State of Partnerships in GTM 2026 report, almost half (49 per cent) of B2B SaaS senior leaders say they want AI to improve partner and account targeting and management.
Calder’s take: this isn’t about generating more content or reports faster. It’s about making better calls — on targeting, prioritization, account intelligence, partner selection and on where to actually spend your team’s time.
All of that requires AI working continuously across live data, not answering one prompt at a time, which is exactly the kind of maturity most teams haven’t reached yet.
So what does it actually take to move right on the AI maturity curve?
According to McKinsey & Company’s 2026 research, the organizations pulling ahead (the AI high performers from that 6 per cent) aren’t just using AI more often — they’re willing to rebuild the process around it.
Almost three-quarters of high performers have overhauled their workflows to actually fit how AI works — up sharply from just over half (55 per cent) a year ago. Everyone else is barely moving: only a quarter have made the same kind of change.
Leadership involvement tracks the same pattern. High performers are roughly twice as likely to have leaders who are genuinely engaged with AI efforts, and twice as likely to actually track whether those efforts are paying off.
Put together, the real answer to “what needs to change” isn’t necessarily a new tool, but a willingness to redesign the workflow the tool sits inside of — backed by leadership who is paying attention to what’s working.
You might also like: Partner program KPIs: The metrics you should measure and optimize.
Where do you sit?
Now that you have a framework for AI maturity, here’s how to use it on your own team. Consider:
- Is AI still just answering what you ask it? Or has it started actively working alongside you?
- When AI touches a process — onboarding, deal registration, etc. — does a human still have to trigger every step, or does it run on its own once the rules are set?
- Is anyone on your team reviewing what AI surfaces, rather than doing the work themselves?
Wherever your team lands, it’s useful to go back to a number we opened with: only six per cent of organizations have turned AI into real financial payoff so far. That’s not a reason to rush toward reaching the Operating Layer stage for its own sake — some of partnerships will always need a human’s judgment. But it is a reason to be honest about where you actually are, not where you assume you are.
Calder’s point here matters: none of this is about just making AI produce things faster. Faster output at an earlier stage of AI adoption just means more for someone to review later. The real value in AI going forward is helping your team decide what’s actually worth doing, and then handling the repeatable parts of it without you.
