An ADU is usually worth it when your lot permits one without a costly variance, the rent covers the financing, and you plan to hold long enough to recover the soft costs. It is usually not worth it on a constrained lot, on a short ownership horizon, or where servicing upgrades dominate the budget. The arithmetic decides it, not the idea.
- Three conditions: permitted without a variance, rent covers financing, long enough hold
- Three situations where it clearly isn’t worth it — and they’re the sections competitors omit
- Value uplift and rental income are two different returns and are routinely confused
- The unit must be legal for either return to fully materialise
- Servicing capacity is the cost that most often turns a good project into a bad one
This question deserves a genuinely two-sided answer, because the honest answer is “often, but not always” — and knowing which situation you’re in is worth more than any amount of encouragement.
What follows is the method, not a promise. We don’t publish rent figures or percentage uplifts, for a reason explained below.
The short answer
Worth it when: your zoning permits the unit as-of-right, your lot can physically carry it after setbacks and parking, servicing is already close, the achievable rent comfortably covers the increased carrying cost, and you’ll hold the property long enough to recover the design and approval costs.
Not worth it when: the lot needs a variance you might not get, servicing has to be trenched or upgraded at the street, you expect to sell within a few years, or an overlay — heritage, floodplain, protected trees — constrains the build.
The encouraging structural fact: the permission half of this is no longer usually the obstacle. Ontario permits three units as-of-right on most serviced residential lots, British Columbia three to four (six near frequent transit), and Edmonton up to eight on a large enough lot. That’s a real change from five years ago, and it moves the decision from “will they let me?” to “do the numbers work?”
How to work out your own number
Four inputs. All four are knowable before you commit to anything.
1. Total project cost.
- Construction cost, quoted
- Soft costs — drawings, survey, engineering, arborist report, permits
- Servicing — the wildcard, and the one to establish first
- Contingency. A budget without one isn’t a budget
2. What you can borrow, and what it costs to carry.
CMHC’s refinance product for secondary suites, available since 15 January 2025, allows borrowing up to 90% of the as-improved value on properties with up to four units, with as-improved value capped below $2 million — subject to owner or close-relative occupancy, funds going to construction rather than equity take-out, and CMHC approval before construction starts.
The number you need is the increase in your monthly payment, not the loan size.
3. Realistic net rent.
Start with gross market rent for a unit of that type and size in your neighbourhood. CMHC publishes the authoritative Canadian source in its Rental Market Report, broken down by centre and unit type — look up your own market rather than relying on a national average, because the spread between Canadian cities is enormous.
Then subtract, honestly:
- Vacancy. Even in a tight market, assume some.
- Maintenance and repairs.
- Any incremental insurance.
- Tax on the rental income, which is real and frequently omitted.
- Ongoing municipal fees. In Surrey, a registered suite attracts secondary suite utility and service fees on your annual property taxes. Check whether your municipality does something similar.
What’s left is net rent. That’s the number that meets the carrying cost.
4. Your hold period.
Soft costs are paid once and recovered over years. A three-year hold recovers far less of them than a fifteen-year hold.
The test: does net rent cover the increased carrying cost, and does the value uplift justify the capital after the hold period you actually expect?
When it’s clearly worth it
Your basement already has the height. Ceiling height is the one expensive fix — 1.95 m minimum in Alberta and Ontario basements, 2 m in BC. If you clear it, the remaining code work (egress, separation, alarms) is ordinary construction, and a basement suite is the cheapest ADU there is.
Servicing is already close. For a rear-yard build this is the difference between a normal project and an expensive one.
Your zoning permits it as-of-right. No variance means no discretion, no hearing, and no risk of refusal.
You’re housing a relative and may rent later. The unit does a job now and becomes an asset later — and if the occupant is a senior or an adult eligible for the disability tax credit, the Multigenerational Home Renovation Tax Credit may cover part of the cost, on up to $50,000 of qualifying expenditures.
You’re holding long term. The longer the hold, the more the arithmetic favours building.
When it isn’t
The section most articles skip. All four of these are real, and recognising yourself in one saves money.
1. A constrained lot needing a variance. If your proposal misses a zoning standard, you’re into a variance process that adds time, cost and genuine uncertainty. In Ontario the Committee of Adjustment test requires the variance to be minor, desirable for the appropriate development of the land, and to conform to the intent of both the official plan and the zoning bylaw — a real test, not a formality. Designing within the standards is worth money; relying on a variance is a risk.
Parking and amenity space are the usual culprits. Calgary requires a parking stall for the tenant in addition to the property’s own requirement, with published minimum dimensions — and street parking and tandem stalls behind other required stalls don’t count. Calgary also requires outdoor amenity space of at least 7.5 m² with no side under 1.5 m. Neither is something a builder can solve.
2. Servicing upgrades dominating the budget. Where water, sewer or electrical capacity has to be upgraded — particularly where that means work in the street — the cost can exceed the construction itself. This is the constraint homeowners consider last and it kills more projects quietly than anything else. Establish it first.
3. A short ownership horizon. Construction is paid upfront and recovered over years of rent. Under about five years, you’ll typically pay the full cost and capture a fraction. If you might move, that changes the answer.
4. Overlays. A heritage designation adds an approval step and constrains exterior changes. A protected tree in the rear yard can move the building, shrink it, or rule it out — routine rather than exceptional in mature Toronto and Vancouver neighbourhoods. Floodplain restricts new residential outright. Calgary’s Airport Vicinity Protection Area can restrict new units and isn’t obvious from the street.
A fifth, less discussed: you don’t want to be a landlord. That’s a legitimate reason, and it doesn’t need justifying with arithmetic.
Value uplift vs rental income
These are two different returns and conflating them is the most common error in this analysis.
Rental income is cash flow while you hold. It’s what covers the borrowing. It arrives monthly, it’s taxable, and it’s reduced by vacancy and maintenance.
Value uplift is what the property is worth when you sell. It’s realised once, at the end, and it doesn’t help you carry the loan in the meantime.
They’re related but not the same, and a project can succeed on one and fail on the other. A suite that rents well in a market where buyers don’t pay for suites produces good cash flow and modest uplift. A laneway house in a city where the unit can eventually be stratified may produce moderate rent and substantial uplift.
On the size of the uplift, we’ll be straight with you: we don’t publish a percentage, because we haven’t found a Canadian source robust enough to stand behind. Plenty of pages quote American figures or unattributed percentages. What we can tell you is the mechanism — a legal second unit adds value through recognised income and a wider buyer pool, an unpermitted one adds much less and can reduce value — and that mechanism is covered properly in our guide on whether an ADU adds value.
The one condition on both returns: the unit must be legal. Lenders generally won’t count rental income from an unpermitted unit toward qualifying a future buyer, insurers may treat an undisclosed unit as compromising a claim, and a buyer’s lawyer will find it. That’s the strongest practical argument for permitting properly, and it’s financial rather than moral.
How to get a real assessment
The four inputs above are all establishable before you spend meaningful money:
- Zoning — your municipality’s mapping tool, free
- Lot capacity — draw your lot, apply setbacks, subtract parking and amenity requirements, see what’s left
- Servicing — the one that needs a professional, and the one to do early
- Rent — CMHC’s Rental Market Report for your centre, adjusted honestly for a below-grade or detached unit
Do those four and you’ll have a defensible answer rather than a hopeful one. Our financing page covers the borrowing side, and our your options page covers the alternatives if the answer is no.
Get the numbers for your own property
The general answer to this question is “it depends.” The specific answer for your address is knowable — and it comes down to what your lot permits, what the build would cost, and what the unit would realistically earn.
HouseLyft’s free property assessment establishes the property side of that. Our results page covers the outcomes we work toward. Request your free report.
This guide explains project economics in general terms and is not financial advice. It does not account for your circumstances, and no figure here is a projection — confirm rents with current market data, financing terms with a mortgage professional, and what your lot permits with your municipality before committing to a project.
Checked by Lee Yousaf, Founder