GTIA just released a research report called "AI Adoption in the Channel: What ITSPs Are Actually Doing, What's Working, and Where to Start." It draws on 19 interviews with MSPs, vendors, distributors, investors, and end customers conducted in the first quarter of 2026.
The report covers everything from 20-person shops to systems integrators with $3 billion in revenue. Most of that does not apply to you. So here is what stood out for a 10 to 20 person MSP, and what you can do with it.
The gap is not adoption. It is operating.
GTIA's numbers tell a simple story. 97 percent of ITSPs use AI internally. Only 20 percent have written policies and governance around that use. Only 25 percent qualify as "AI Driven," which GTIA defines as having formal adoption, governance, and a clear revenue strategy.
Another 47 percent land in the "AI Invested" bucket. They deploy tools but lack the structure to scale. The last 28 percent are still exploring or actively resisting.
So almost everyone uses AI. Almost nobody runs a business around it. That distance between using and operating is the whole problem.
Nobody handed you a roadmap
One MSP operator in the study put it plainly. He wanted someone to stand up and say, here is a logical path for an MSP over the next three years. That never happened. Another operator said the same thing a different way: plenty of training exists on how the technology works, but nobody teaches how to build a business around it.
That gap is not laziness. GTIA calls it rational risk management. You cannot afford to fail publicly in front of clients. Waiting feels safer than guessing wrong.
The catch is that waiting costs you too. GTIA cites the OpenText 2025 Global Managed Security Survey, which found MSP readiness to guide clients on AI dropped from 90 percent to roughly 50 percent in a single year. Confidence fell while the stakes went up. Meanwhile one major vendor reported that partners embracing an AI agenda grow two and a half times faster than partners who do not.
The speed is different this time
Cloud took about a decade to reach mainstream adoption in the channel. AI got to near-universal internal use among MSPs in roughly two years.
That compression matters for a small shop. With cloud you had time to watch what your peers did and copy the winners. You do not have that runway now. Client expectations get set by whoever gets there first, and re-entry gets harder the longer you sit out.
What the firms getting results do differently
GTIA found six repeatable behaviors across the interviews. Four of them work at your size.
Start internal, then sell external
Deploy AI on your own operations before you sell it. You build proof points, you learn the failure modes, and you can speak from experience instead of from a vendor deck.
The numbers back the sequencing. GTIA reports only 11 percent of ITSPs generate more than 25 percent of revenue from AI today. Internal adoption is nearly universal. External monetization is not. The firms that skipped the internal step have a harder time selling because they cannot answer basic questions about what breaks.
Measure before you market
Track before-and-after metrics on the workflows where you deploy AI. Ticket resolution time. First-call resolution rate. Technician utilization. CSAT.
One consultant in the study measured 30 to 60 percent efficiency gains and 30 to 35 percent cost reduction against pre-AI baselines. Those numbers become your sales material. An M&A advisor quoted in the report said buyers now ask for operational and financial metrics, not the claim of AI.
Multiple interviewees flagged the opposite behavior as a credibility risk: MSPs who change their website to say "AI MSP" with nothing behind it. Clients figure that out.
Systematize before you automate
This one gets skipped the most. Technicians spend 20 to 30 percent of their day hunting for SOPs, scripts, and client details. That is unbillable time. AI documentation and runbook generation cut that fast, and one MSP in the study reported documentation time dropped 60 to 70 percent.
But GTIA is direct about the sequence. Firms that deploy AI on messy, undocumented workflows perform worse than their own baselines. AI speeds up whatever process you already have. If the process is ad hoc, you get faster chaos.
Use platforms as accelerators, not as your strategy
Every major RMM and PSA vendor built AI into their core products during 2024 and 2025. ConnectWise acquired zofiQ in January 2026 and brought agentic AI to service desk automation. Kaseya added workflow generation to VSA 10. NinjaOne launched Patch Intelligence AI.
Your competitors buy the same tools. A venture investor in the study called vendor platforms a connector, not a platform, meaning the tool is not the differentiator. Your differentiation sits above it, in domain knowledge and client relationships.
The pricing problem heading your way
Per-device and per-user pricing assumes labor scales with endpoints. AI breaks that assumption.
GTIA describes the efficiency paradox with a Chicago-area MSP that grew from 3,000 to 10,000 endpoints with the same headcount. Revenue tripled. Labor stayed flat. That works right up until the client does the math and realizes they pay the same per-endpoint rate for a fraction of the labor.
GTIA cites a Gartner prediction that 60 percent of large IT services contracts will include AI clawback clauses by 2027, forcing providers to share efficiency gains with clients, and Forrester research putting 45 percent of services decision makers on track to expand performance based pricing.
You do not need to rebuild your pricing model this quarter. Pick one AI service and price it differently. Per assessment. Per project. Per agent deployed. One advisory firm in the study charges per assessment output rather than per subscription and reports the assessment generates 8 to 12 times its fee in follow-on implementation revenue.
Pick a vertical and go deep
This is the strongest data in the report.
Vertically focused MSPs achieve 30 percent higher profit margins, 38 percent higher client retention, 22 percent shorter sales cycles, and 42 percent higher average contract values. Specialized providers command 10 to 20 percent premium pricing over generalists. Healthcare-focused MSPs lead at 31 percent annual growth.
Then the gap. GTIA cites data showing only 23 percent of MSPs have implemented vertical specialization, while 66 percent say it matters. Everyone agrees it works. Few do it.
AI makes the vertical advantage bigger, not smaller. Generic AI capability commoditizes fast. Industry-specific compliance knowledge does not. An end customer in financial services said it clearly: he needs partners who understand SEC, FINRA, FinCEN, and OCC requirements, and general AI competency without regulatory depth is insufficient and potentially dangerous.
If you serve healthcare, legal, financial services, or insurance, compliance is not the thing slowing you down. It is the thing keeping generalists out of your accounts.
You will reskill your team, not replace it
The report is honest about the workforce piece. AI does not eliminate MSP jobs. It changes them.
Roles shrinking: L1 help desk, manual documentation, data entry. GTIA cites projections that AI triage handles 50 percent of support tickets by 2027, and partners in the research reported endpoint-to-technician ratios jumping from 190:1 to 357:1.
Roles growing: AI governance, security operations, vCIO and strategic advisory, automation engineering.
At your size you will not hire AI specialists. CompTIA data in the report shows 52 percent of channel companies already struggle to find candidates with the right skills, and AI-related skills demand rose more than 100 percent year over year. You will train the people you have. GTIA's recommendation for small shops is one hour per week of structured AI learning for every team member.
One more warning worth repeating. AI boosts senior staff productivity by 30 to 60 percent and quietly suppresses junior development. If AI handles the work your junior techs used to learn on, they never build judgment. Nobody in the study had solved this.
Where to start
GTIA's guidance for early-stage firms comes down to five moves you can start this month.
- Write a basic AI governance policy. Not 50 pages. Which tools you approve, when you disclose AI use to clients, and how an employee proposes a new use case. The NIST AI Risk Management Framework gives you a credible foundation.
- Run a data readiness assessment on your own environment first. Data quality, not AI implementation, is the bottleneck. GTIA cites research showing more than 90 percent of MSPs name data quality as their biggest roadblock.
- Pick one internal workflow where AI cuts manual effort. Ticket triage, documentation, or billing. Measure it before and after. One, not three. A shop your size that starts three finishes none.
- Train your client-facing staff on AI conversation starters. One VAR in the study did exactly this and shifted client relationships from reactive support toward strategic guidance before they had anything to sell.
- Identify which verticals in your client base carry the highest AI opportunity, usually the regulated ones.
Ninety days from now you want a documented policy, one internal pilot with real before and after numbers, and at least two client conversations about AI governance.
The takeaway
GTIA ends the report with a line worth keeping. The firms that lead the next phase of the channel are not the ones with the most AI tools. They are the ones turning AI into operational discipline, vertical depth, and trusted relationships, and doing it on Monday morning instead of in the next strategic plan.
Talk it through
You do not need a three year strategy. You need one pilot, one policy, and one vertical. Picking which vertical, and what it does to your pricing, is the conversation worth having.
Everything above is a template. Run it yourself this week and never talk to me. What I will not do is hand you a system and tell you your shop is wrong for not matching it.
Not sure this is your actual constraint? Take the MSP Owner Reality Check. Five questions, nine minutes, and it names the two or three things quietly capping your growth. https://themsphero.com/resources/msp-owner-reality-check-assessment
If you already know what is broken, book a 30 minute fit call at https://letschat.themsphero.com
Mike Kolb The MSP Hero
Source: GTIA, "AI Adoption in the Channel: What ITSPs Are Actually Doing, What's Working, and Where to Start" (2026). The research was conducted by AchieveUnite on behalf of GTIA and draws on 19 semi-structured interviews across the channel ecosystem. GTIA notes the findings are directional rather than statistically representative. Read the full report at gtia.org.