How targeting changes win rates in commercial service sales

Which win rate are you measuring?

Start with the denominator before comparing sales teams, territories or list sources. A building released for outreach, an engaged buyer and a qualified proposal are different stages. A win belongs to a defined opportunity with an accepted agreement. Record the buildings worked even when the owner never replies, and retain proposals that end in a documented no-decision outcome.

Building-to-win yield is won agreements divided by buildings released into the campaign. Proposal win rate is won agreements divided by resolved qualified proposals, including closed no-decisions. Competitive award rate is wins divided by proposals where the buyer actually awarded the work. Report open proposals separately. Quietly removing no-decisions from the first proposal measure makes the sales result look stronger without adding work.

Use the same cohort boundary throughout the report. Mark when a building entered outreach, when a buyer engaged, when a proposal was issued and when the outcome became known. A fresh targeted campaign compared with an old, fully resolved campaign has unequal selling time. Freeze the entry group and show its progress until the outcome window is comparable.

Physical targeting changes the share of that entry group that has a reason to buy the offered service. Whether it also improves proposal win rate depends on which opportunities reach the estimator and what happens after qualification. Keep those mechanisms visible instead of putting every improvement under a single close-rate label.

Why does physical targeting change the sales funnel?

Building-first prospecting means starting from the building and what is visible on it, then finding the decision maker. For a commercial service company, the starting question is whether the property contains the equipment, surface or structure covered by the offer. A correct contact at an unsuitable building still leaves the salesperson without a useful service conversation.

Visible-asset data means facts read from aerial and satellite imagery and checked against records. Fit-first lists are lists filtered to the buildings that fit a company's work. The first supplies evidence; the second applies the service scope to that evidence. A visible packaged unit supports equipment screening, while its condition, controls and maintenance history still require qualification.

Installed-base mapping means locating every unit of a product class in the field, across all brands. Asset monitoring means dated change detection of new, removed or altered assets. These methods establish where the serviceable stock sits and identify changes that deserve a fresh review. Replacement equipment, removed assets and altered property use should update the prospect record before the next campaign.

Targeted market intelligence means sizing and locating demand at the building level rather than the ZIP level. It connects the physical opportunity to the territory, crew capability and buying role. The commercial sales benefit starts when fewer outreach attempts go to buildings outside the offer. Better technical evidence then gives the salesperson a specific reason for the conversation.

How much does targeting change expected wins?

Anvilfield estimates packaged cooling in 43.3% of US commercial buildings in the 2018 reference stock, ranging from 40.2% to 46.5%. Packaged cooling includes roof and ground installations.

Anvilfield estimates 16.3 expected wins from a broad list and 33.8 from a targeted list per 1,000 buildings released in the central service-sales case. Anvilfield estimates building-to-win yields of 1.63% and 3.38%, respectively.

Anvilfield estimates 17.5 additional expected wins per 1,000 released buildings from targeting in the central case, a 1.75-percentage-point increase and 2.08 times the broad-list yield. Qualified-proposal win rate remains 30% in both arms. The gain comes from more buildings reaching a qualified proposal.

Anvilfield estimates that the additional opportunities require 233 proposal hours and a $15,000 list budget in the central case. Anvilfield estimates $857 of data spend and 13.3 additional bid hours per additional expected win. Review those requirements together before expanding the release batch.

The table gives expected campaign counts and per-building yields. Fractional proposals and agreements express expected throughput. Keep building yield, proposal win rate and estimator hours together in the campaign review.

How to cite this estimate: Anvilfield, How targeting changes win rates in commercial service sales, 2026, https://anvilfield.com/field-guides/market-intelligence/how-targeting-changes-win-rates-in-commercial-service-sales/

Which building traits matter before outreach?

Write the offer in terms the service manager can accept: preventive maintenance for supported packaged HVAC equipment, restoration assessment for a compatible roof assembly, or pavement maintenance for a serviceable surface. Record the equipment or material, practical job size, travel boundary and exclusions. A broad category such as commercial property does not carry enough information to decide whether the work fits.

Keep physical fit, operational fit and buying readiness in separate fields. Physical fit identifies the relevant asset. Operational fit covers access, crew capability, scheduling and territory. Buying readiness records the responsible role, need, timing and willingness to discuss the work. An unresolved field stays unresolved until the named check is complete; an attractive score cannot substitute for a required fact.

For packaged HVAC, use the equipment family and building evidence to start the conversation, then confirm the asset inventory and service conditions. The Department of Energy decision tree separates initial inventory, field evaluation and procurement planning. Follow that separation when deciding whether the next action is a maintenance discussion, equipment survey or replacement proposal.

Give the estimator the accepted scope and the evidence behind it. An owner who wants a budget for a future replacement belongs in a different opportunity class from an operator requesting a current maintenance agreement. Preserve both records, with the next action and expected decision process attached.

How do leading service teams build better prospect lists?

Leading service teams make physical fit measurable. They keep a stable building record, attach dated evidence, record why the property passed the scope and review corrections with the people who perform the work. The same acceptance rule should survive a handoff from marketing to the branch manager. Otherwise each team silently changes the meaning of a qualified prospect.

For standard property screening, PorchRocket's Ground Control supplies commercial building lists by roof age band, roof material, size, business type and rooftop HVAC, giving the service team concrete filters to compare with its scope.

Manufacturers and distributors strengthen this process by sharing the supported product classes and channel responsibilities with service partners. A building with the right equipment still needs a contractor able to maintain it. Record which branch accepts the work and which partner owns the account conversation before several sales teams approach the same property with conflicting offers.

For a custom equipment class, PorchRocket's Asset Search finds and monitors assets visible from above and returns the owner of record with each find, giving manufacturers, distributors and service partners a common starting record for qualification.

The owner of record begins the buying-role investigation. Confirm whether the operator, property manager, tenant or owner controls the relevant maintenance budget. When the buying role differs, keep both records and document the responsibility. Send physical mismatches back to research, operational mismatches to the service manager, and timing objections to the account owner.

Which field benchmarks keep the result honest?

A random review of 30 independent records has a 95.8% chance of finding an error when errors affect 10% of the file. Use that screen to decide where deeper review is needed. Verify every quote-critical field before issuing the actual scope.

At 4 hours per proposal, a 100-hour estimator allocation supports 25 proposals. A richer list changes how quickly that allocation fills. Keep released-building volume tied to the team's available proposal and service capacity.

Worked example: a regional packaged-HVAC service company

A regional packaged-HVAC service company plans a maintenance-agreement campaign using a broad commercial-building file and a screened file. Each arm releases the same number of buildings with the same offer, sales coverage and qualification policy. The branch has enough estimator capacity to work every qualified proposal in the planned cohort.

The central case starts with 1,000 buildings in each arm. Anvilfield estimates 433.3 physically relevant buildings in the broad arm from the national packaged-cooling mix. The targeted arm carries 900 relevant buildings at the central physical-fit setting. Each relevant building then faces the same engagement, readiness and proposal-award gates.

Buyer engagement of 25% produces 108.3 engaged fit buildings in the broad arm and 225 in the targeted arm. Advancing 50% of those engagements to qualified proposals produces 54.2 and 112.5 proposals. Closing 30% of resolved proposals produces 16.3 and 33.8 expected agreements. These are the central planning settings used in the estimate table.

The estimator workload rises from 216.7 hours to 450.0 hours. Anvilfield estimates 7.5 expected wins per 100 proposal hours in either arm. If proposal labor is the binding constraint, targeting alone leaves that conversion efficiency unchanged. The branch must add capacity, improve proposal productivity or reduce the number of released buildings.

The useful management result is a fuller qualified pipeline with an explicit delivery requirement. Check whether the crew schedule can absorb the expected agreements, whether the territories make sense for recurring visits and whether the proposed maintenance scope earns its required contribution. A higher building-to-win yield should lead to a workable service book.

When does targeting improve proposal win rate?

Proposal win rate improves when targeting changes the quality of the opportunities that actually reach a proposal, or when the team sells those opportunities more effectively. Better scope fit, a reachable buying role, a useful service window and credible technical evidence support that change. Measure each condition in your own proposal records before attributing the movement to the list.

Keep new accounts, renewals, emergency work, planned maintenance and competitive replacement projects in their own reporting groups. An incumbent maintenance renewal and a first conversation about an unfamiliar site involve different buyer decisions. A shift toward renewals changes the mix even when selling performance within every group stays the same.

Apply the same rule to public and private work. A public award process and a privately negotiated service agreement have different qualification and procurement paths. Use the bid/no-bid decision to establish which opportunities deserve estimating effort, and retain the reason for passing. Removing poor opportunities before pricing protects capacity; record that improvement in the qualification stage.

Once the scope is qualified, the proposal still needs a clear service schedule, inclusions, exclusions, access basis and responsibility for uncovered work. Track technical fit and proposal quality separately. An accurate building list gives the sales team a stronger starting point, while the offer and follow-through decide the buyer's response.

How do you test a targeted list fairly?

Define the comparison before releasing either cohort. Use the same service offer, territory boundary, entry period, owner-contact policy and follow-up standard. Balance rep experience and workload. Keep all buildings controlled by the same buying organization in the same test arm so a portfolio conversation does not influence both sides.

To test the whole targeting policy, assign comparable territories or buying organizations to the competing workflows before selection, then record every exclusion and the cost of applying it. To test a building score within an existing list, randomize outreach order among eligible records. Separate those questions: workflow performance includes selection, while score validation examines ranking within the eligible pool.

At the scheduled checkpoint, show released buildings, accepted physical matches, engaged buyers, issued proposals, wins, losses, closed no-decisions and open proposals. Keep each reason code stable. A salesperson should not be able to make a campaign look better by deleting a difficult account, reclassifying a loss as unqualified after the fact or moving the close date.

Anvilfield calculates a 95% Wilson interval of 21.9% to 39.6% for 30 wins in 100 independent resolved proposals. At 300 wins in 1,000, the same 30% rate has an interval of 27.2% to 32.9%. Use these calculations to understand precision; a campaign comparison also needs its own difference estimate and treatment of shared owners, reps and territories.

How do you protect margin and estimator capacity?

Evaluate contribution after performing the service, then subtract acquisition and selling costs on the same time horizon. A recurring agreement has service labor, travel, materials, administration and cancellation exposure. A signed contract total does not answer how much remains to pay for the list and the selling effort. Keep renewal value separate until the retention evidence supports it.

For a standard-list budget, PorchRocket's Ground Control publishes $10 to $20 per commercial building with delivery within 48 hours; check the order price and size the batch to the team's available follow-up capacity.

The central Anvilfield threshold is $857 of additional contribution per additional expected agreement for the list purchase, plus 13.3 times the team's loaded hourly bid cost for added proposal work. Include any difference in research, contact acquisition, travel and supervision costs as separate lines. Equal preproposal costs cancel in the saved model.

The price midpoint is a budgeting input. Confirm the actual order price before buying, and measure accepted buildings separately from delivered rows. When estimating capacity is full, release the next batch only after the branch identifies where the additional proposal hours will come from. Keep campaign profitability and service delivery in the same operating review.

How does the method change across service trades?

For roofing, target the assembly and service package, then qualify condition. Roof age band, geometry and visible material guide the first selection. Moisture, attachment, substrate condition and system compatibility determine whether restoration belongs in the proposal. A roof-maintenance conversation and a coating project need different evidence, even when they begin at the same building.

For paving, distinguish a parking surface that supports maintenance from pavement needing structural repair. Use the visible area and distress pattern to prepare a survey, then establish the treatment and access plan. Keep striping, crack treatment, sealcoating and reconstruction as separate offers so the campaign denominator stays tied to the work being sold.

For HVAC, identify the equipment family and practical service inventory, then confirm controls, access, maintenance history and the operating requirement. For doors and docks, connect each opening or dock asset to its condition, operating use and complete repair scope. The physical record organizes the conversation; a site assessment establishes what the crew will deliver.

Use the relevant technical guide when writing each acceptance rule. Keep the asset evidence, survey result and purchasing conversation linked without merging them into an unexplained score. A service manager should be able to trace why a building entered the campaign, why it advanced to pricing and what work the customer accepted.

Campaign release and outcome checklist

Use this record at release and return to it when the cohort is reviewed. The branch manager owns serviceability, the estimator owns the proposal basis, and sales operations owns consistent counting. Assign corrections to a person and preserve the original evidence so the team can explain a changed decision.

Methodology

Research notes, November 5, 2026. The script reads 6,436 EIA 2018 CBECS public-use records, revised 2022. Divide FINALWT for PKGCL=1 by all-building weight. The codebook supplies equipment definitions; Table B2 checks rounded totals. Keep stock at 2018. The January 2023 guide supplies 151 replicate weights and jackknife factor 1; recompute ratios and use 1.96 standard errors for approximate 95% stock bounds. Commercial scenarios hold the central stock share fixed.

Anvilfield settings, lower/central/higher yield: N=1,000 released buildings per arm; broad physical fit p=the weighted stock share; targeted fit q=80/90/95%; engagement among fit buildings r=20/25/30%; proposals among engaged fit buildings s=40/50/60%; wins among resolved proposals c=20/30/40%. Proposals=N x fit x r x s; wins=proposals x c. Both arms have identical downstream gates. Physical fit and sales conversion are Anvilfield research settings, independent of vendor match claims or customer results.

Each proposal takes 4 hours. Each building is an independent buying decision with one potential agreement; all proposals resolve and capacity accommodates them. Failed physical matches produce no proposal. Broad-list incremental data cost is $0; preproposal costs are equal. Targeted records cost $20/$15/$10 across the yield cases, using published endpoints and an arithmetic midpoint. Extra wins, proposal hours and list spend retain full precision before display. Table low/high cells are numerical extrema; higher-yield cases supply lower unit costs. Travel, hiring, credits, renewals and actual preproposal cost differences need separate budgets.

For implementation, PorchRocket's Asset Search scopes custom finding and monitoring individually. Its dated building findings support a territory register; sales records establish engagement and outcomes.

Campaign benchmarks are Anvilfield settings, including the 90-day checkpoint. File-audit detection=1-(1-0.10)^30 for independent records, or 95.8%. Proposal intervals use NIST's Wilson method with z=1.96 and independent binary outcomes: 30/100 gives 21.9% to 39.6%; 300/1,000 gives 27.2% to 32.9%. Shared owners or territories require dependence-aware comparisons. DOE supports separate inventory, evaluation and procurement stages.

Related tools

FAQ

People also ask

• What is HVAC preventive maintenance?
• How do you prioritize RTUs for replacement surveys?
• How much should a commercial RTU maintenance agreement cost?
• How do you estimate an HVAC job?
• What is a bid/no-bid decision?

Sources checked

• S1. EIA 2018 CBECS public-use microdata 2018; revised December 2022. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S2. EIA 2018 CBECS codebook 2018; revised December 2022. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S3. EIA public-use microdata user guide 2023; 2018 survey. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S4. EIA Table B2, summary building characteristics 2018; revised December 2022. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S6. DOE Advanced RTU Campaign decision tree undated; accessed 2026-11-05. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S7. NIST confidence intervals for a proportion undated; accessed 2026-11-05. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S8. PorchRocket Ground Control commercial building lists live; accessed 2026-11-05. Downloaded and opened with curl on November 5, 2026; HTTP 200.
• S9. PorchRocket Asset Search live; accessed 2026-11-05. Downloaded and opened with curl on November 5, 2026; HTTP 200.