Roofing
Size a roof-coatings market from building counts, not guesswork
Count buildings, validate a sample, and carry eligible roof area into a territory worksheet. Separate the roof stock from annual owner decisions and the work your crew can complete.
Direct answer
Size a roof-coatings market by counting distinct buildings in a defined territory, measuring their roof areas, and validating a sample for coating suitability. Expand the sample results within comparable roof groups to estimate total addressable market (TAM), then apply your service limits to estimate serviceable available market (SAM). Model owner timing, sales conversion, and crew capacity separately. The result is a conditional planning estimate with an evidence trail; a visible roof is not a guaranteed buyer.
On this page
What does a roof-coatings market estimate measure?Set the boundary and scope before countingHow do you count buildings without counting them twice?Which area and age records belong in the model?What makes a roof eligible for the coating scope?How do you build a useful validation sample?Worked validation sample: 100 buildingsHow do the counts become TAM and SAM?Worked money example: stock value and annual demandHow much of the annual pool can your crew complete?How sensitive is the estimate to weak inputs?What to documentBuild the next research batchCommon mistakesMarket-sizing checklistStandards and referencesUnits, terms, and definitionsMethodologyOpen the related tool for the calculation, checklist, or packet.
FieldOS by AnvilfieldOffline-first FieldOS for job photos, proof packets, property memory, calculations, teams, billing, and owner-ready reports.What does a roof-coatings market estimate measure?
A useful estimate tells you how much roof your business could investigate, what portion might support your coating scope, and how much work could reach production during a stated period. Start with the building register. Keep the records that explain every subtraction and multiplier. Another estimator should be able to rebuild your result without asking what you meant by a good roof.
Use three units throughout: buildings for research workload, roof square feet for physical scope, and dollars for a defined installed package. Accounts are a separate count. Count the twelve buildings controlled by one owner separately, and identify the roof systems within each building. A list of twelve addresses does not establish twelve owners or twelve coating projects.
This guide uses TAM for the estimated technically eligible roof stock inside your chosen boundary. SAM is the portion your business can serve under its stated operating limits. Both are stock measures here. Label any annual demand calculation separately, with its own timing input. That prevents a once-through inventory value from appearing in a one-year sales budget.
Set the boundary and scope before counting
Write the territory polygon or county list, building uses, minimum roof size, permitted systems, evidence cutoff, and planning horizon at the top of the workbook. This example uses November 4, 2026 as its reference date and models the following twelve months. Specify how boundary properties, campuses, new construction, and mixed-use buildings will be treated.
Define the offer just as carefully. Restoration of existing membrane roofs, metal roof maintenance coatings, and recoating a documented existing system can require different preparation, testing, labor, and sales conversations. Keep those groups separate until your estimator has a reason to combine them. Set a minimum practical job size from your own setup and travel costs.
Keep a rejection ledger. A property outside your chosen scope stays in the research history with its exclusion reason. This makes a changed boundary or expanded crew capability easy to recalculate. Otherwise, the next person sees a smaller list without knowing whether buildings disappeared, were duplicated, or failed a technical review.
How do you count buildings without counting them twice?
Assign a stable building ID, then link parcel IDs, street addresses, owner accounts, and roof-section IDs beneath it. Resolve multiple tenant addresses and repeated data deliveries before totaling buildings. Keep separate structures on a campus separate. For mixed roofs, measure nonoverlapping sections and retain their material and evidence states individually.
Cross-check a sample of included buildings against a second local inventory, and check excluded areas for omissions. Inspect a few boundary blocks directly on the map. Keep demolished structures, additions, incomplete coverage, and address conflicts in an exception tab. An eligibility sample cannot repair a building inventory that systematically misses an industrial corridor.
National building statistics provide context, not an address list. The U.S. Energy Information Administration's 2018 Commercial Buildings Energy Consumption Survey reports building counts and commercial floorspace [1]. Keep that historical survey year visible. Those totals do not establish the current roof inventory of your service territory. Obtain local building and roof records before assigning coating demand.
| Worksheet field | Entry and counting rule |
|---|---|
| Building and account IDs | One physical building; owner account linked separately |
| Parcel and address matches | Preserve alternate addresses without creating extra buildings |
| Roof-section ID and boundary | Nonoverlapping area, tied to the building |
| Roof area and basis | Plan or measured surface area, source, exclusions, and date |
| Covering and existing coating | Separate fields; documented, observed, conflicting, or unknown |
| Technical outcome | Eligible for stated scope, unsuitable, or unresolved; record basis |
| Commercial outcome | Approving role, owner timing, next action, and evidence date |
| Operating fit | Travel, job size, access, system capability, and assigned reviewer |
Which area and age records belong in the model?
Use roof area, with its measurement basis stated. Floor area can span several stories; it cannot stand in for the exposed roof. A footprint can support preliminary roof plan area after you resolve courtyards, additions, canopies, and shared boundaries. Actual coating quantities need the surface dimensions, slopes, vertical details, and exclusions in the eventual takeoff.
A roof installation record belongs to a section, not automatically to the whole building. Store installation, recover, coating, repair, permit issue, and completion dates separately. An old building may carry a recent roof, and a recent coating may cover an older assembly. Use age bands to organize research without assigning a remaining service life.
Seattle's project-research guidance illustrates why record dates need checking: historical scan upload dates differ from the dates inside the documents, older addresses can differ, and some plans require a request [2]. Check your own jurisdiction's coverage and completion evidence. A portal search with no result is an unresolved history, not proof that a roof has never been replaced.
What makes a roof eligible for the coating scope?
Write the assessment questions before selecting the sample. Have the qualified reviewer establish the assembly, repair needs, moisture condition, substrate soundness, adhesion, drainage, and compatibility with the proposed system. Obtain authorization for access and any tests or openings, including their closure. Keep missing technical evidence unresolved until the assessment supplies it.
The Roof Coatings Manufacturers Association directs preparation and coating quantities to the manufacturer and project requirements, identifies limitations for gravel and ballast, and calls for leaks to be repaired before coating [3]. Its guidance supports an investigation, not automatic eligibility based on material class. No universal wet-area percentage or roof-age cutoff is used in this worksheet.
Gaco's metal-roof application specification MR-S42-26-15, revised May 2026, requires sound panels and structural elements and recommends adhesion testing before bidding [4]. Those are requirements for that system. Record the actual acceptance basis for each proposed system rather than copying one manufacturer's conditions onto every roof in the territory.
Color, patches, and rust-colored staining can help choose questions for the visit. They do not establish concealed moisture, corrosion depth, material chemistry, or owner intent. Likewise, ENERGY STAR explains that cool-roof energy benefits depend on conditions such as climate and insulation [5]. Do not turn a reflective surface or a modeled energy saving into an annual purchase probability.
How do you build a useful validation sample?
Freeze the cleaned inventory, then group buildings by characteristics likely to change the assessment: roof class, size band, location, and known history. These groups are strata. Select buildings randomly within each stratum, saving the inventory version, selection procedure, random seed, and selected IDs. Existing customers and the easiest roofs to access deserve their own labels.
Schedule the selected buildings through an owner-authorized assessment. Track selected, contacted, authorized, assessed, unsuitable, eligible, and unresolved counts separately. If an owner refuses access, retain that selected record. A replacement visit may help operations, but silently substituting a friendly owner changes what the sample represents.
Define eligibility against the same coating scope for every assessed roof in a group. Record the repair allowance and unresolved tests. Have a second reviewer check ambiguous classifications and a few accepted records. The purpose is to expose classification errors before they multiply across hundreds of buildings.
For area estimates, retain eligible square feet for every selected building, including zero eligible area on confirmed unsuitable roofs. Within each stratum, multiply building count by the sample's mean eligible area per selected building when outcomes are complete. This handles partial roofs and the link between size and suitability. Do not multiply a count pass rate by an unrelated average building size.
Worked validation sample: 100 buildings
Anvilfield research models a typical regional building inventory. The case uses a cleaned inventory of 1,000 buildings in three adjacent counties, with one in-scope roof section per building and no overlapping areas. Every building within a row has the same net horizontal roof area shown below. That simplifying input makes count rates and area rates identical within each row.
A random sample within each row selects exactly ten percent of its buildings: 60 membrane, 25 metal, and 15 unresolved at desk review. All 100 receive completed authorized assessments against the defined scope, with no refusals, missing results, or pending tests. Unknown desk materials are resolved during assessment. The original desk group remains attached to each sampled record.
36 membrane roofs, 10 metal roofs, and 3 initially unresolved roofs qualify in full after the allowed minor repairs. All other sampled roofs are confirmed unsuitable for this package. No sampled roof qualifies only in part. These modeled outcomes produce estimated eligibility rates of 60, 40, and 20 percent respectively.
Multiply each inventory count by its own sample rate. The implied eligible building counts are 360, 100, and 30. Multiply those by the row areas to obtain 7,200,000, 1,600,000, and 360,000 square feet. Their sum is 490 estimated eligible buildings and 9,160,000 square feet. Unsampled roofs remain unverified; the estimates do not release individual properties for coating.
| Inventory and validation sample | Estimated eligible scope |
|---|---|
| Membrane: 600 buildings; 20,000 sq ft each. Sample: 36 / 60 eligible = 60%. | 360 buildings x 20,000 sq ft = 7,200,000 sq ft. |
| Metal: 250 buildings; 16,000 sq ft each. Sample: 10 / 25 eligible = 40%. | 100 buildings x 16,000 sq ft = 1,600,000 sq ft. |
| Unresolved at desk review: 150 buildings; 12,000 sq ft each. Sample: 3 / 15 eligible = 20%. | 30 buildings x 12,000 sq ft = 360,000 sq ft. |
| Total: 1,000 buildings; 17,800,000 sq ft. Sample: 49 / 100 eligible. | 490 estimated eligible buildings; 9,160,000 sq ft. |
How do the counts become TAM and SAM?
The example's technical TAM is the 9,160,000 eligible square feet just calculated. Now operating limits retain 75 percent of eligible membrane buildings, 60 percent of eligible metal buildings, and 40 percent of eligible buildings from the initially unresolved group.
Those filters leave 270 membrane buildings, 60 metal buildings, and 12 from the initially unresolved group: 342 serviceable building equivalents. Their areas are 5,400,000, 960,000, and 144,000 square feet, totaling 6,504,000. That is SAM stock. Preserve rejected buildings and the limiting reason so a second crew or changed service boundary can be modeled without rebuilding the inventory.
For real portfolios, apply operating fit to each assessed record before estimating serviceable area. Technical suitability and travel economics can be related. The simplified percentages here use the stated retention within each row and its fixed roof size. Do not treat the worksheet's convenient arithmetic as evidence that all filters are independent.
TAM area = sum of (stratum building count x mean eligible sample area per building)SAM area = sum of estimated eligible area retained by the stated operating limitsWorked money example: stock value and annual demand
The case uses an installed selling value of $4.50 per net horizontal roof square foot for this typical package. It includes cleaning, specified primer and coating, ordinary detail work, routine access, agreed minor repairs, and the contractor's overhead and margin within that selling price. Every eligible example roof fits that package. Exclude major deck or insulation replacement, drainage redesign, tear-off, hazardous-material work, sales tax, and financing.
Actual production estimates require system-specific quantities and local costs. Use the same scope and price basis for all dollar comparisons here. There is no inflation, discounting, price escalation, repeat coating, or second contract on the same area during the twelve-month example.
At that price, TAM stock has a once-through installed value of $41,220,000, and SAM stock has a value of $29,268,000. Neither is annual revenue. Exactly 20 percent of serviceable area becomes owner-authorized coating work available to contractors during the horizon, spread proportionally across the three groups. Timing and funding remain inputs until owners confirm them.
Annual available scope is therefore 6,504,000 x 0.20 = 1,300,800 square feet, worth $5,853,600. It represents 68.4 expected building equivalents under the fixed-size inputs. The fraction is an expectation for budgeting; actual projects are whole jobs with their own scopes and dates.
Your business competes for that entire annual pool and wins 25 percent of its area at the price. That produces 325,200 square feet and $1,463,400 in potential sales before checking capacity. This is installed contract revenue, not coating-product shipments, contribution, or profit. Acquisition costs and direct job costs have not been calculated, so the result cannot establish a campaign return.
| Stage and area calculation | Installed value at modeled $4.50 |
|---|---|
| Technical TAM stock: 9,160,000 sq ft. | $41,220,000 once through. |
| Serviceable SAM stock: 6,504,000 sq ft. | $29,268,000 once through. |
| Annual available scope: SAM x 20% = 1,300,800 sq ft. | $5,853,600 during horizon. |
| Potential won scope: annual scope x 25% = 325,200 sq ft. | $1,463,400 before capacity limit. |
How much of the annual pool can your crew complete?
One coating crew has 80 productive crew days remaining after existing commitments and allowances for weather, travel, and downtime. Its completed-system output averages 3,000 net square feet per productive crew day, including preparation, detailing, coating passes, and closeout.
Capacity is 80 x 3,000 = 240,000 square feet. Materials, access, inspections, cash, and owner schedules allow that area to be completed and billed inside the horizon. Potential won work exceeds that capacity by 85,200 square feet. Limit the production-backed sales plan to 240,000 x $4.50 = $1,080,000 under these inputs.
This limit still needs a job schedule. Averages do not fit an oversized building into a short weather window or resolve simultaneous owner shutdowns. Keep overflow as separately scheduled backlog, decline it, or price an explicit subcontract plan. Do not increase the win rate to solve a production constraint. Track completion and billing separately if their timing differs.
How sensitive is the estimate to weak inputs?
Start with the input that changes the business decision. Keeping the example's SAM, price, win rate, and crew capacity fixed, annual owner release of 5 percent produces 325,200 available square feet and 81,300 potentially won square feet. At 10 percent release, those amounts double. At 20 percent, the crew becomes the limiting factor. None of these release rates is a measured local buying rate.
Sampling uncertainty deserves a separate check. Using the NIST Wilson method [6], 36 eligible roofs out of 60 gives an approximate 95 percent binomial interval of 47.4 to 71.4 percent. The interval uses z = 1.96 and omits finite-population correction. It applies to that group's count eligibility under the model, not to the whole territory's area or revenue.
That interval does not include inventory omissions, classification errors, nonresponse, price uncertainty, or owner timing. Keep those limitations beside the sample results. A wider technical sample will not answer a funding question. Use the next research budget to resolve the variable that changes whether you hire, buy material, or open another territory.
When outcomes are incomplete, show bounds before extrapolating. If 10 of 60 selected buildings remain unresolved and 30 are eligible, the selected sample has confirmed eligibility of 30/60 and a maximum of 40/60 if all unresolved roofs qualify. Those are outcome bounds, not confidence limits. The complete-case rate of 30/50 describes assessed buildings only; report why the others were not assessed.
| Annual release and available value | Potential win and capacity limit |
|---|---|
| 5% of SAM: $1,463,400 available installed value. | 81,300 sq ft potentially won; $365,850 capacity-limited sales. |
| 10% of SAM: $2,926,800 available installed value. | 162,600 sq ft potentially won; $731,700 capacity-limited sales. |
| 20% of SAM: $5,853,600 available installed value. | 325,200 sq ft potentially won; $1,080,000 capacity-limited sales. |
What to document
Release four connected sheets: the building register, the selected validation records, the input ledger, and the calculation summary. Give each rate a numerator, denominator, date, source, responsible person, and evidence state. Separate assessed outcomes from management inputs. Save the inventory version so the next review can distinguish new buildings from corrected classifications.
A strong note reads: membrane group, 600 buildings; 60 randomly selected and assessed; 36 eligible for the stated package; 20,000 square feet each modeled for this typical case; annual owner timing unverified. A weak note reads: 600 old roofs, sixty percent good, twenty percent buying. The first note tells the manager exactly which result exists and which input still needs work.
Hand sales a building ID, matched roof section, verified contact role, permission status, owner decision, and next action. Keep the research estimate linked to that record. Update the model after rejected surveys as well as wins. Review material, size, travel, and owner-timing groups separately before ordering a larger outreach batch.
Build the next research batch
Start with a bounded territory and a written selection request. PorchRocket's roof-coating list page supports roof age band, membrane or metal, building size, business type, visible signs, and area criteria [7]. Keep those match reasons in your register. Ask what coverage the delivered selection represents before treating its count as a complete territory inventory.
Use the file to organize records requests, owner calls, and authorized assessments. Compare the returned mix with the groups in your worksheet. A filtered prospect batch can support outreach without representing every building in the boundary. Preserve that distinction when expanding a sample or presenting a market total. Owners on a matching list have not necessarily requested work.
Use the related Anvilfield guides for the next task: roof measurement for area, inspection for condition records, restoration qualification for survey booking, and coating yield for material takeoff. The older-roof and prospect-list guides support research and handoff. Keep product quantities and individual proposals tied to their own verified scopes.
Common mistakes
Using floor area, tenant addresses, or parcel counts as coating scope changes the denominator before the first percentage is applied. Reconcile physical buildings and roof sections first. Another common error is applying an overall pass rate to every roof class, even though the sample contains a different mix of metal, membrane, and unresolved roofs.
Dividing roof stock by a unverified warranty term invents an annual replacement cycle. Counting restoration and replacement on the same area doubles alternative scopes. Treating a survey invitation as funded work inflates the sales pool. Give each transition its own evidence and keep alternative scopes mutually exclusive within the planning horizon.
Avoid decimals that imply more certainty than the inputs support. Preserve exact arithmetic inside the workbook, then round planning quantities sensibly and show the scenario range. Do not translate the territory estimate directly into a material purchase. The approved system, field quantities, access, and schedule must support that order.
Market-sizing checklist
Want this checklist to run itself on every job — with photo proof and a signed record crews can hand the customer? That's FieldOS.
Standards and references
Sources reviewed November 4, 2026. EIA's figures describe 2018, not a current local census. Seattle's archive guidance covers its jurisdiction. Gaco's May 2026 specification applies to its named system. The association and ENERGY STAR pages support the technical points cited above; confirm local code adoption and the proposed system's requirements for each project.
The sampling design, sample outcomes, operating filters, prices, owner timing, win rate, and capacity are disclosed worksheet inputs. Confirm the proposed system's current documents and the project's requirements before releasing an individual scope. The linked guidance is not a blanket approval for any listed roof.
EIA: 2018 CBECS principal building activitiesSeattle: Research a Project, Permit, or PropertyRoof Coatings Manufacturers Association: FAQsGaco: MR-S42-26-15, revised May 2026ENERGY STAR: Cool RoofsNIST: Confidence intervals for proportionsPorchRocket: roof-coating prospect criteriaUnits, terms, and definitions
A roofing square is 100 square feet. Keep roof plan area, actual coated surface, and product-order quantities distinct. TAM and SAM in this worksheet describe estimated eligible and serviceable roof stock. Annual available scope adds an explicit owner-timing input. Potential won scope adds a sales input; the production plan also needs available crew capacity.
A stratum is a defined group used for sampling and estimation. An unresolved outcome is missing evidence, not a confirmed rejection. An expected building equivalent is a budgeting quantity that can be fractional; an actual contract requires a particular building, agreed scope, and authorization. Installed contract value includes the stated package and does not equal manufacturer revenue or contractor profit.
Methodology
Counts, prices, work rates, schedules and outcomes in typical cases are inputs to the Anvilfield research model. Calculations use the stated scope, formulas, cost inclusions and exclusions. Cited public records and technical documents support their attributed facts. The regional inventory, sample results, eligibility factors and production rate are model inputs; the stated filters are applied in sequence. Territory counts, field-sample outcomes, owner budgets, coating quotes and production rates are model inputs rather than measured field results.
FAQ
How do you estimate a local roof-coating market?
Count distinct buildings, measure their roof sections, and validate suitability within comparable groups. Estimate eligible area, apply your operating limits, then model annual owner timing, sales conversion, and crew capacity separately. Retain a source and denominator for each rate.
What is the difference between TAM and SAM for roof coatings?
TAM is the estimated technically eligible roof stock within the stated boundary in this worksheet; SAM is the portion your business can serve. Neither is annual sales. State the planning period and owner-timing inputs when converting stock into annual demand.
Can commercial building floorspace be used as roof area?
No. Floor area can include multiple stories and does not establish exposed roof area. Map roof sections, label plan versus surface measurements, and verify the dimensions and details before pricing a coating scope.
Is a 100-building sample enough to size every territory?
No. The required sample depends on roof diversity, precision, access, and the decision at stake. The research model uses a 100-building sample. Report outcomes by group and keep refusals and unresolved assessments visible.
How should unassessed roofs be treated in the sample?
Keep them unresolved with the reason and follow-up action. Show confirmed and possible outcome bounds rather than treating missing roofs as passed or failed. A rate among assessed roofs does not automatically represent every selected building.
Can roof age or aerial images establish coating suitability?
No. They can organize research and assessment questions. Obtain roof-specific records and an authorized investigation against the proposed system's requirements before classifying the individual roof as suitable. Keep technical eligibility separate from the owner's willingness to buy.
Is $4.50 per square foot a verified coating price?
No. Obtain local, system-specific estimates for preparation, repairs, details, access, materials, labor, and project requirements before using a price in a business commitment.
Can a purchased building list establish the whole market?
Only if its population coverage and selection method support that use. A filtered prospect batch alone does not establish a territory census, technical eligibility, or buying intent. Reconcile coverage and validate selected buildings before expanding its results.
People also ask
Sources checked
- EIA: 2018 CBECS principal building activities Checked November 4, 2026 with curl, HTTP 200. Historical building counts and commercial floorspace; not local roof area or current buying demand.
- Seattle: Research a Project, Permit, or Property Checked November 4, 2026 with curl, HTTP 200. Municipal archive guidance on upload dates, historical addresses, and plan requests.
- Roof Coatings Manufacturers Association: FAQs Checked November 4, 2026 with curl, HTTP 200. Used for preparation, substrate limitations, manufacturer-specific quantities, and leak repairs.
- Gaco: MR-S42-26-15 metal-roof application specification May 2026 revision; checked November 4, 2026 with curl, HTTP 200. System-specific soundness and pre-bid adhesion-testing guidance; not universal coating eligibility.
- ENERGY STAR: Cool Roofs Checked November 4, 2026 with curl, HTTP 200. Climate and insulation context for energy benefits; savings and buying probability are modeled separately from this page.
- NIST: Confidence intervals for proportions Checked November 4, 2026 with curl, HTTP 200. Wilson interval method; the article's 36-of-60 calculation uses z = 1.96 without finite-population correction.
- PorchRocket: roof-coating prospect criteria Checked November 4, 2026 with curl, HTTP 200. Live roof-class and territory selection criteria; matching properties are not verified coating buyers.