Rules · dmap grant ontario
Ontario will pay up to $15,000 toward the plan you were going to pay for
Ontario's DMAP offers up to $15,000 toward a digital adoption plan for SMEs with 1 to 499 employees, and a follow-on program adds up to $50,000.
Lasse Pettersen
The Ontario Centre of Innovation runs a program called the Digital Modernization and Adoption Plan, and as published it offers up to $15,000 toward developing a digital adoption plan for an Ontario for-profit SME with 1 to 499 employees. A follow-on Technology Demonstration Program offers up to $50,000 toward implementing what that plan recommends.
If you were about to pay a consultant for AI planning work, that is the first thing to check, and most Ontario business owners I speak to have not heard of it.
What the two programs are, as published
The Digital Competence Centre at the Ontario Centre of Innovation runs a two-step path.
| DMAP | Technology Demonstration Program | |
|---|---|---|
| What it funds | Developing a digital modernization and adoption plan with a digital adoption consultant | Implementing the technologies the plan identified |
| Amount | Up to $15,000 | Up to $50,000 |
| Employees | 1 to 499 | 1 to 499 |
| Revenue floor | None stated | Minimum annual revenue in one of the last three tax years |
| Prerequisite | None | A completed DMAP |
| Availability | First come, first served, while funds last | Application windows apply |
Two cautions about that table, both of which matter more than the figures.
Verify the current status with the administrator, not with a summary. Program windows open and close. At the time this was written the Technology Demonstration Program’s published application window had a stated closing date that had passed, which means either a new window or no window, and only the administrator knows which. A grant listing site is not the administrator.
Third-party summaries disagree with each other. At least one widely circulated guide quotes the Technology Demonstration Program at $100,000 rather than the $50,000 the Ontario Centre of Innovation publishes. When a number about your money appears in two places, take the one from the body handing out the money.
The federal layer above it
In June 2026 the federal government launched a national AI strategy carrying over $3.5 billion, including a financing program delivered through BDC aimed at small and mid-sized businesses adopting AI tools and infrastructure, and a regional initiative delivered through agencies including FedDev Ontario.
Financing is not a grant. A loan to buy an automation is still a loan, and it should clear the same arithmetic as any other capital purchase: hours saved, dollars per year, payback period. That arithmetic is what an assessment produces, and what it should look like is set out in what an AI assessment should actually deliver.
Why the sequencing matters more than the amount
The most common way an Ontario business loses access to this money is not failing to qualify. It is doing the work in the wrong order.
Programs of this shape are generally built around approval preceding the eligible activity. A plan you commissioned and paid for in March is usually not retroactively eligible in June, however well it matches what the program funds. So the order to work in is:
- Confirm you meet the stated eligibility: incorporated, valid business number, permanent establishment in Ontario, headcount inside the band, and able to show you can implement and sustain what you adopt.
- Confirm the intake is open. First come first served while funds last means a program can be technically live and practically closed.
- Confirm what “eligible activity” covers and when it may start.
- Then engage somebody.
Doing steps 1 to 3 costs you an afternoon and one phone call. Skipping them costs you the difference between paying for a plan and having one paid for.
What a funded plan should contain
Whether a program pays or you do, the deliverable list is the same, and it is worth writing into the engagement rather than assuming:
- An inventory of repeated processes, gathered by watching the work rather than by survey.
- Hours per week and dollars per year against each.
- The three worth doing first, ranked, with the reason the fourth was excluded.
- Named tools with real subscription costs in Canadian dollars.
- A build estimate for each candidate.
- The candidates recommended against, and why.
- Where personal information enters each process, because that decides tool eligibility before features do. The reasoning is in what PIPEDA means for your AI tools.
A plan without the fourth and sixth items is a shopping list. A funder is unlikely to object to it and you should.
Who this is actually for
The employee band is wide, 1 to 499, but the businesses where this changes the decision cluster in a narrower range. Under about five people, the repeated work is real but frequently does not yet total enough hours to pay back a build, and a funded plan will honestly tell you so. Above roughly 200 people there is usually somebody internal already doing this work.
The band where a funded plan is most useful is roughly 10 to 150 employees: large enough that several processes are genuinely repeated, small enough that nobody has been given the job of noticing.
That is also the band where the Statistics Canada figures are least intuitive. In the second quarter of 2026, 19.9% of businesses with 1 to 4 employees reported using AI, against 27.8% of those with 100 or more. The gap between the smallest firms and the largest is under eight points, which is far narrower than most owners assume, because one person wearing four hats has more repeated work per head rather than less.
The honest caveat about consultants and grants
I do not run applications, I do not take a percentage of one, and I will not tell you an application will be approved. Nobody selling you services controls that decision, and a supplier who implies otherwise has told you something useful about how they operate.
What I will do is write the assessment so it lines up with what a plan application asks for, if you tell me on the first call that you are applying. The $999 assessment is priced under the published Canadian band of $1,500 to $5,000 partly for this reason: it is small enough to be a straightforward decision whether or not a program is paying, and the fee comes off any build you go ahead with.
If a program does pay, the sensible move is to spend the allowance on a larger scope rather than pocketing the difference. A wider process inventory finds more, and the finding is the part that is hard.
One more caution about scope. A funded plan is written to a program’s definition of digital adoption, which is broader than artificial intelligence and includes ordinary systems work. That is an advantage rather than a compromise. In a large share of Ontario manufacturers and distributors the correct recommendation is not an AI build at all. It is finishing an implementation of software the business already owns and has been paying for since 2021, with a set of spreadsheets filling the gap where a module was never turned on. A plan that can say so is more useful than one restricted to recommending the thing its author sells.
Before you do anything else
Check whether the intake is open, check the eligibility against your own numbers, and check the amount against the administrator’s own page rather than against a summary. Then decide whether you want a plan at all. If your answer is that you already know exactly what to build, skip planning and get a quote on workflow automation instead. A grant for a plan you do not need is still a project you did not need.