How to budget roof CapEx across a multi-site portfolio

Most roof budgets start with one of two questions. How old is the building, or which one leaked last? Both are easy to explain. Neither tells you where the money will actually be needed.

A condition-based plan does. It takes more work up front, but it gives you something you can defend in a budget meeting, and something you can update instead of rebuilding every year.

Start with condition, not age

Age is a rough proxy. Two roofs installed in the same year can be in very different shape because of drainage, exposure, past repairs, and how much maintenance they've had. A roof at 12 years can be a 4 or a 2.

Condition data fixes that. If you've scored each roof on the 1 to 5 RCI scale (see What a roof condition score tells you, and what it doesn't), every building already carries an estimated remaining life and a recommended response. That is the raw material for the plan.

Age still matters as an input. It just can't be the answer by itself. For each building you want the score, the estimated remaining life, section-level notes, repair history, and warranty status where you have it. If a record is missing, mark it in the plan as a gap instead of guessing. A plan that is honest about its gaps is easier to defend than one that fills them with assumptions.

Build three horizons, not one number

A single annual figure hides too much. Three horizons are easier to manage and easier to explain.

Year 1. Buildings scored 1 or 2, plus the immediate repairs flagged on 3s. A 1 means under a year of estimated life, so those are not optional. A 2 has an estimated 1 to 3 years and lands here or in year 2 depending on severity.

Years 2 and 3. The remaining 2s and the 3s with the shortest estimated life. This is where replacement planning turns into design, scoping, and procurement, which takes time you don't want to find out you need after a roof fails.

Years 4 and 5. The 3s with more runway. Buildings scored 4 or 5 mostly stay out of the capital plan and live in your maintenance budget.

Treat the horizons as a sort order, not a schedule carved in stone. Estimated life is a range, and the plan needs a review every year as conditions change. The point of looking five years out isn't to predict the exact year a roof fails. It's to stop being surprised by the ones you could see coming.

Use ROM estimates, and keep repair and replacement apart

A Rough Order of Magnitude (ROM) estimate is a budget-level cost built from roof size, system type, and condition. It is not a bid. It's good enough to size a budget three years out, and it's honest about what you don't know yet.

TEMA builds ROM estimates per roof area on each building, so you can see which section is driving a building's number instead of getting one lump sum. Carry the estimates in the plan as budget figures, and replace them with quotes as each project gets close. When a number moves, you can explain why. That beats defending a single figure that was never firm.

Keep repair and replacement in separate lines. They are different decisions with different budgets. A repair extends the life of a roof and is often handled in the operating budget. A replacement is a capital project. Plan repairs and preventive maintenance (PM) as an annual line, and replacements as projects. How your organization treats each for accounting purposes is a question for your finance team, so confirm it with them instead of assuming.

Keep replacements from stacking up

The biggest budget risk usually isn't one expensive roof. It's several of them coming due in the same year.

Sort the portfolio by estimated remaining life, then look for pile-ups. Where several buildings cluster in one year, ask which can be repaired to buy time and which should be pulled forward. Sometimes a targeted repair buys a year or two on a 3. Sometimes it doesn't, and the plan should say which is which.

Deferring isn't free, so put the cost of deferral in the plan. A roof that stays in service another year still needs repair spend and carries leak risk. Showing that next to the deferral makes the trade visible, and it keeps the conversation from turning into "can we push this one more year?" without a number attached.

It's also worth pricing projects by region. One crew mobilizing for several nearby buildings may change the numbers compared with pricing each site alone.

Present it so finance believes it

Finance doesn't need to understand roofing. They need to see the method.

Show how each building earned its place: the score, the estimated life, the recommended action, the cost range. Show your assumptions and what would change them. Flag the buildings where you don't have condition data yet, and say what it will take to get it.

Then treat the plan as a living document. Each year, show what changed since the last version: which roofs moved up, which projects were completed, which estimates were replaced by quotes, and why the total moved. A plan that explains its own variances earns trust faster than one that never changes.

The goal isn't a perfect forecast. It's a plan that holds up when someone asks, "Why this building, and why this year?"

Next step

If you want a quick read on where your portfolio stands before you build the plan, the free Portfolio Health Snapshot takes about two minutes and needs no site visit. When you're ready for building-level condition scores, ROM costs, and a capital horizon you can budget from, that is what a Portfolio Assessment is for.