Standardize the tenants first. Then make AI measurable.
Nerdio's 2026 State of the MSP Business Report (Oct 20, 2026) finds 32% of MSPs use AI internally but only 12% can show measurable gains. What an MSP should standardize, govern and measure before it sells AI.

Published October 21, 2026 · ~6 min read
What Nerdio published
On October 20, 2026, Nerdio released its 2026 State of the MSP Business Report. The headline finding is blunt: 32% of MSPs have adopted AI tools for their own operations, another 46% are piloting or planning adoption within 12 months, and just 12% can point to measurable productivity gains.
The gap shows up on the revenue side too. Only 14% of MSPs operate a revenue-generating AI service line, and 17% say customers are asking about Microsoft Copilot or other AI tools while they are not yet ready to respond.
Channel Insider covered the report the same week, and the framing from Nerdio's Will Ominsky, VP and GM of its MSP business, is the useful part: "AI didn't create the divide in the MSP industry. It's making the one that was already there easier to see."
What the survey measured
The report is based on a survey of 153 MSP decision-makers who manage Microsoft 365 and Azure across their customer bases. Thirdside ran it through an independent research panel from June 22 to July 7, 2026. Sixty-three percent of respondents identify as primarily or pure-play MSPs, and the sample leans toward owners, C-suite and service-delivery leaders.
Nerdio sorts MSPs into three operating models: native Microsoft tools and manual processes, community and open-source tools, and commercial multi-tenant management platforms. It calls operating model the sharpest predictor of most outcomes in the dataset, ahead of firm size, headcount and tool count. Keep in mind that Nerdio sells a commercial management platform, so that conclusion favors its own category.
Nerdio's numbers, and what they are
Every figure is Nerdio's own, from a survey of 153 respondents. None is a Consultiply measurement.
| Claim | What was measured | Status |
|---|---|---|
| MSPs that adopted AI for their own operations | 32% | Vendor claim |
| MSPs piloting or planning AI within 12 months | 46% | Vendor claim |
| MSPs with measurable productivity gains from AI | 12% | Vendor claim |
| MSPs with a revenue-generating AI service line | 14% | Vendor claim |
| MSPs whose customers ask about AI before they are ready | 17% | Vendor claim |
| MSPs applying a documented AI governance framework across every AI customer | 14% | Vendor claim |
| MSPs enforcing one documented security and configuration standard across every tenant | 20% | Vendor claim |
| MSPs that call configuration drift at least an occasional issue | 71% | Vendor claim |
| MSPs that need more than a business day for an audit-ready Microsoft 365 assessment, or cannot produce one | 41% | Vendor claim |
| MSPs that lost technicians to burnout in the past 12 months | 29% | Vendor claim |
Why this matters for an MSP building an AI practice
Most MSPs are not short on AI tools. The report suggests they are short on the operating basics that make AI repeatable: one standard per tenant, a governance framework that is actually applied, and evidence on demand.
Those basics are also what a client's AI rollout depends on. If an MSP cannot say which tenants match its security baseline, it cannot say where Copilot or an agent is safe to switch on. If it takes days to produce an audit-ready assessment, it will struggle to answer an insurer or a client board asking how AI is controlled.
The report also points at time. Sixty percent of MSPs say Microsoft management complexity is growing faster than their ability to scale, and 29% lost technicians to burnout in the past 12 months. Technicians who are buried in manual work have little room to learn AI well enough to deliver it.
Four foundations to set before you sell AI
One baseline per tenant.
Write a single security and configuration standard and check every customer tenant against it. Track drift as a number you review monthly, not as an occasional surprise.
Evidence on demand.
Be able to produce an audit-ready Microsoft 365 assessment the same day it is asked for. Treat a slow answer as a gap to close before any AI project starts.
A governance framework you apply everywhere.
Decide who may use which AI tools, with what data, and who approves new ones. Apply the same framework to every customer using Copilot or other AI, then record exceptions.
Fluency on the desk.
Give technicians daily, supervised use of AI and Copilot on their own work first, so the people delivering AI to clients know where it helps and where it fails.
A 30-day plan to make one result measurable
Pick one internal workflow.
Choose something your team repeats every day, such as ticket triage notes or onboarding checklists.
Record a baseline.
Before you change anything, write down time per task, rework and escalations for two weeks.
Run AI with a human in the loop.
A technician reviews every output. Log what was accepted, edited or rejected.
Compare and decide.
At day 30, compare against the baseline. Keep, change or drop the workflow based on your own numbers, not a vendor's.
Package what works.
Once one workflow shows a result you can stand behind, write it up as a scoped service for clients, with the governance framework attached.
Measured internal results are the strongest proof an MSP can bring to a client conversation about AI. They are also the start of an AI service line that is more than resale.
Read with care
Vendor survey, 153 respondents.
These are vendor survey results from 153 respondents, fielded June 22 to July 7, 2026. They describe MSPs that manage Microsoft 365 and Azure, not the whole channel.
Nerdio sells a platform.
Nerdio sells a commercial multi-tenant management platform, and its operating-model comparisons favor that category. Nerdio says its segment comparisons are statistically significant at the 95% confidence level unless otherwise noted.
Gains are self-assessed.
"Measurable productivity gains" is the respondents' own assessment. The press release does not define how gains were measured.
Press release, not the full report.
We read the press release, not the full report. Check the full report before quoting a figure that is not listed here.
This guide is Consultiply's independent reading of Nerdio's published survey results. No partnership with Nerdio is implied, and it is not a recommendation to buy any product.
Sources
32% of Managed Service Providers Have Adopted AI, but Only 12% Can Show It's Paying Off (Nerdio, press release, October 20, 2026). Only 12% of MSPs Report Measurable AI Gains, Nerdio Finds (Channel Insider, October 2026).
Reviewed, not run by us.
Consultiply's independent reading of Nerdio's published survey results; every figure is Nerdio's own. Nerdio, 2026 State of the MSP Business Report (press release, Oct 20, 2026) →