How Do I Get My Field Reps to Sell Service Agreements?

This vs. the common alternatives

Most service organizations reach for one of four levers when agreement attach stalls, and only one of them holds up past the first quarter.

The pep-talk lever. A manager stands up at Monday's meeting, says agreements are the priority now, and asks the team to pitch every job. Attach rate ticks up for two or three weeks and then decays back to baseline, because nothing structural changed. The rep's day is still measured by revenue booked and jobs closed, and the agreement is still the optional last thirty seconds of a call that is already running long. Verbal priority without a measured line is the single most common failure mode in field organizations, and it is expensive precisely because it feels free — no comp redesign, no system change, no CRM work. The cost shows up as a manager who now believes "our reps just won't sell agreements," which is a diagnosis of the incentive, not the people.

The spiff lever. A flat bonus per agreement sold — fifty dollars, a hundred, a gift card, a trip. Spiffs work, and they work fast, which is why they are so seductive. The problem is that they work only while running. A spiff is an event, not a system: attach rate spikes during the contest window and collapses when it ends, and every subsequent spiff has to be larger to produce the same lift because the team has learned that agreements are a special promotion rather than part of the job. Worse, a flat per-unit spiff rewards volume over quality. Reps chase the cheapest, shortest agreement they can write because every unit pays the same, and twelve months later the renewal book is full of one-year minimum-tier contracts that churn.

The mandate lever. Every quote must include an agreement line; every work order requires an attach disposition before it closes. Mandates are underrated — they fix the "the rep never mentioned it" problem, which is genuinely the largest single leak — but they only govern the offer, not the sale. A rep required to present an agreement can present it in a way engineered to get a no: mumbling the price, framing it as an upsell, presenting it after the customer has already reached for the door. Mandates raise the presentation rate, which is a real and measurable gain, and they generate the data you need. They do not by themselves raise close rate on the presentation.

The weighted scorecard. You enumerate every outcome a complete rep produces — often eight or nine lines — assign each a weight set with leadership, score every rep 1-to-5 on each line, and roll it to a composite: the sum of weight × level across all KPIs. A rep who is a level 5 on new-equipment revenue and a level 1 on service-agreement attach and renewals lands a mediocre composite, visibly, on a published matrix. The gap becomes impossible to hide and impossible to argue with, because the weights were published before the scores were.

The reason the matrix beats the other three is that it survives contact with a rep who is optimizing. Reps are rational: they do what is measured and paid. A pep talk isn't measured. A spiff is measured but temporary. A mandate is measured but only at the offer stage. The composite is measured continuously, covers the whole job, and — critically — is re-weightable. When leadership decides recurring revenue matters more this year than unit volume, you change the weights and the entire field team re-aims within a pay period, without a reorg, a new tool, or a new speech.

The practical answer for most teams is not one lever but a stack: mandate the presentation so the offer always happens, weight the scorecard so the composite reflects the whole job, and put the spiff on the *quality* dimension — multi-year terms, premium tiers, renewals — rather than raw unit count. That combination fixes the leak at the offer stage, the measurement at the outcome stage, and the mix at the quality stage.

How to choose between them

The right lever depends on where your leak actually is, and most teams guess wrong about that. Before redesigning comp, pull three numbers for the trailing ninety days: the percentage of eligible jobs where an agreement was *presented*, the close rate on those presentations, and the renewal rate on the existing book. Each points at a different fix.

If presentation rate is under roughly 40%, your problem is the offer, not the pitch — the agreement is simply not coming up on most jobs. A mandate plus a required disposition field on the work order fixes more of that gap than any comp change, and it fixes it in days rather than a pay cycle. If presentation rate is healthy but close rate on presentations is low, that is a skill and framing problem: the rep raises it, the customer says no, and the rep moves on. That responds to call structure, objection handling, and a scorecard line that scores the outcome. If both look fine but renewal rate is sagging, the sale is happening and the service delivery is not — no incentive redesign on the sales side will fix a book churning because nobody performed the maintenance visits the contract promised.

There is a second axis: where you want the teeth. Visibility tools — scorecards on TVs, leaderboards, recognition feeds, mobile push — change behavior through social pressure and are cheap to run, but they plateau with reps who don't care about the leaderboard. Compensation tools change behavior through money and don't plateau, but they take a pay cycle to land and carry real design risk if you get the accelerators wrong. Most field organizations under about fifteen reps get more per dollar from visibility plus a hand-built matrix; above that, the administrative cost of running multi-component comp by spreadsheet exceeds the cost of a comp platform, and you should buy one.

A third consideration is trade context. In HVAC, plumbing, and electrical, the agreement is sold at the point of service by a technician who did not think of themselves as a salesperson when they were hired, so the scorecard has to be paired with a genuinely short, memorable pitch and a price the tech can quote without calling in. In capital equipment and industrial service, the agreement is negotiated alongside a large purchase by someone who is unmistakably a rep, and the constraint is that the agreement margin is a rounding error next to the equipment margin — so the weight on the agreement line has to be disproportionate to its revenue contribution or it will never compete for attention. In facilities and commercial service, the agreement often *is* the product, and the real KPI is not attach but term length and scope creep.

The choice rule that holds across all three: buy the layer where your teeth are weakest, and build the matrix yourself regardless of what you buy. Nobody ships a weighted scorecard that matches your business out of the box. The weights are the strategy, and the strategy is yours.

Costs, timelines, and expected impact

Budget the work in three buckets — design time, tooling, and comp exposure — and set expectations on a realistic curve, because a scorecard that is judged at week two will be killed before it works.

Design time. Building the initial matrix is a two-to-four hour working session with a sales leader, a service manager, and whoever owns RevOps or the CRM. You are producing three things: the KPI list, the weights, and the definition of levels 1 through 5 on each line. The level definitions are the part teams skip and the part that determines whether the matrix survives its first dispute. "Level 3 on attach" must mean a specific number — for example, attach on 30–39% of eligible jobs — not "meets expectations." Expect a second session two weeks later to fix the two or three lines you defined badly the first time. Plan for that revision instead of treating it as failure.

Tooling. The honest range is zero to substantial. A spreadsheet costs nothing but your time and carries a real risk of going stale — a matrix nobody updates is worse than no matrix, because it teaches reps the whole exercise is theater. Purpose-built scorecard and gamification platforms commonly run in the low-to-mid tens of dollars per user per month, quoted rather than listed at the higher end. Incentive-compensation platforms are almost always custom-quoted and are priced for organizations where comp complexity, audit, and accurate payout at scale are the pain. Field-service operations platforms in the trades carry the heaviest price tag but replace the whole operational stack and, importantly, are already recording attach and membership conversion inside the work-order flow — meaning the data for your matrix arrives clean and same-day rather than by manual export. If you're already running one, pull attach from it before you buy anything new.

Comp exposure. This is the number that determines whether any of it works. If the composite governs a token slice of variable pay, reps will price the tradeoff correctly and ignore it. The practical threshold is that a meaningful share of variable compensation — commonly discussed in the 30–50% range — has to move with the composite before behavior changes. Below that, the rational move for a rep is to keep selling equipment and eat the scorecard hit. Above it, attaching the agreement is the shortest path to their own paycheck, and the coaching conversation changes from persuasion to arithmetic.

Timeline. Week one is design and publication. Weeks two through four are when reps start asking about their scores — the leading indicator that the matrix is real to them is unprompted questions about how a specific line is calculated. Behavior shift typically becomes visible in the data in the first four to six weeks, with attach rate moving before renewal rate for the obvious reason that renewals lag the sale by the contract term. Full adoption, where the composite is simply how the team talks about performance, is a one-to-three month arc. Renewal-rate improvement won't be legible for a full contract cycle, so don't put it on the ninety-day scoreboard.

Expected impact. Be careful here, because this is where teams invent numbers. What you can say with confidence is directional: presentation rate responds fastest and most reliably because it is nearly a pure compliance metric. Close rate on presentations responds to coaching and script work over a longer arc. Mix — term length, tier — responds to how you weight those lines specifically, and will not improve on its own just because total attach improved. Renewal rate is downstream of service delivery more than sales, so improvements there tend to trace back to fulfillment and customer-success work rather than the scorecard.

The second-order effects worth budgeting for. A rising agreement book changes your capacity math: every agreement sold is a scheduled maintenance obligation, and a team that doubles its book without adding technician hours will miss visits, which torches renewal rate about a year later. Model the labor before you incentivize the sale. It also changes cash flow and valuation — recurring revenue smooths seasonality in trades where equipment sales are brutally weather-dependent, and a service book is generally the most valuable asset on the balance sheet at exit. Those two facts are usually the reason leadership wants the shift, and saying them out loud to the field makes the scorecard read as strategy rather than as a new way to withhold bonus money.

Implementation and handoff details

The rollout is where most matrices die, almost always for one of three reasons: the data feeding it is manual, the reps found out about the weights after they were scored, or nobody owned it after launch.

Fix the data source first. A scorecard is only as trustworthy as its inputs. Before you publish anything, confirm that every KPI on the matrix can be pulled without a human retyping it — equipment revenue, agreement attach, renewal dates, contract term, tier, and parts should all come out of the CRM or field-service platform on a schedule. If a line requires a rep to text in their own numbers, either automate it or cut it. One self-reported line poisons the credibility of the entire composite, because the first rep who disputes a score will point at that line, and they'll be right.

Publish the weights before the first score. Announce the matrix, the weights, and the level definitions in a meeting where reps can argue about them, and then actually change one or two things based on that argument. The concession is cheap and it converts the matrix from something done *to* the field into something built *with* it. Publishing after scoring is the single fastest way to make the whole program read as a pay cut in disguise.

Run a shadow period. Score for two to four weeks with no comp consequence. Reps see their composite, dispute what looks wrong, and you find the three lines that are calculated in a way nobody anticipated. Then go live on pay. Skipping the shadow period saves a month and costs you the first quarter.

Assign an owner. Someone in RevOps or sales ops owns the refresh cadence, the dispute queue, and the quarterly weight review. Without a named owner the matrix is stale inside sixty days.

The handoff between sales and service is its own project. An agreement sold is a promise the service organization has to keep, and the two sides usually don't share a system of record for it. Define, in writing, what happens at the moment of sale: who schedules the first maintenance visit, what the agreement entitles the customer to in plain language, where the terms live so a dispatcher can see them, and who calls the customer at renewal. In the trades this is often the difference between a book that renews above 80% and one that renews in the fifties. Score the renewal line on the rep who owns the account, but investigate every non-renewal for a fulfillment cause before you score it against them — otherwise you're punishing reps for a dispatch failure and they will stop trusting the matrix entirely.

Coach the gap, not the score. The composite is a diagnostic, not a verdict. In a one-on-one, the useful move is to find the line with the largest weight × gap-to-next-level and work only on that — a rep at level 2 on a heavily weighted attach line has more available upside than the same rep going from 4 to 5 on a lightly weighted one. That arithmetic makes coaching concrete and short, and it's the mechanism that turns a scoreboard into a development plan.

Watch for the predictable gaming. Every scorecard gets gamed; the job is to make the gaming productive. Weighting raw agreement count invites cheap one-year minimum-tier deals, so weight term and tier alongside count. Weighting attach rate alone invites reps to mark marginal jobs ineligible, so define eligibility in the system rather than by rep judgment. Weighting renewals invites pre-emptive discounting at renewal time, so score renewal *value*, not just renewal count. None of this means the matrix is broken — it means reps are doing exactly what you asked. Adjust the ask.

Extend the pattern once it works. The same weighted-composite machinery applies to any recurring outcome a field organization wants: extended warranties, membership programs, consumables replenishment, monitoring subscriptions, and inspection contracts all behave identically under the matrix. So do adjacent functions — a service dispatcher can be scored on first-time-fix and schedule density, a call center on booking rate and agreement leads passed. The reason it generalizes is that the mechanism is not about agreements at all. It's about making the full definition of the job visible and paid, so that the thing you say matters is the thing that shows up in someone's check.

Related questions

What attach rate should we target?

Set the first target off your own trailing baseline rather than an industry figure, since eligibility definitions vary wildly between organizations. A common approach is targeting a relative lift — measurable improvement over your ninety-day baseline — for the first two quarters, then resetting once the definition of an eligible job has stabilized.

Should technicians or dedicated salespeople sell agreements?

Technicians close at the moment of trust, right after fixing something, which is the strongest position anyone will have. Dedicated sellers close better on complex, multi-site contracts. Most trades operations get more volume from technicians and use a specialist only for commercial accounts above a size threshold.

How do we stop reps from discounting agreements to close them?

Score and pay on agreement *value* and term, not unit count, and set a floor price the rep cannot go below without a manager approval that is logged. Discount authority delegated without a logged approval always drifts down to the floor.

Does this work for a team of three reps?

Yes, and it's cheaper — a spreadsheet with weights and a published composite works fine at that size. The failure mode at small scale isn't tooling, it's the owner forgetting to update it. Put the refresh on a recurring calendar block.

What if the CRM doesn't track agreement renewals?

Fix that before building the matrix. Renewal date, term, and tier are three fields; adding them is a small RevOps task compared to running a scorecard on numbers nobody can verify. Until they exist, score only the lines you can pull cleanly.

FAQ

What if my field reps ignore the scorecard and just sell equipment?

Then the scorecard isn't wired to anything that matters to them. A matrix with no comp consequence is a report, and reps correctly treat reports as optional. Move a meaningful share of variable pay onto the composite — commonly discussed in the 30–50% range — so a low agreement score visibly reduces the check. Reps aren't ignoring the scorecard; they're reading the incentive accurately.

How many KPIs belong on the matrix?

Most teams land between six and ten lines. Typical entries include new-equipment revenue, agreement attach rate, renewal rate, average contract term, tier mix, parts and consumables, quote-response speed, and territory activity. Fewer than six and the matrix misses parts of the job; more than ten and no rep can hold it in their head, which defeats the purpose of publishing it.

Can I change the weights after the matrix is live?

Yes — that's the main advantage over a fixed comp plan. If renewals become the priority, raise that weight and announce it before the period starts. Reps re-aim within a pay cycle because the matrix is public and their income depends on it. The rule is that weights change between periods, never retroactively inside one.

My reps don't trust the scoring. How do I fix that?

Publish the weights and the numeric level definitions before anyone is scored, run a shadow period with no pay impact, automate every input so nothing is self-reported, and hold a standing dispute window where a rep can challenge a line and get a written answer. Trust in a scorecard comes from the process being auditable, not from the manager being persuasive.

How fast should we expect results?

Presentation rate moves within weeks because it's close to a compliance metric. Close rate on presentations follows over a longer arc as coaching lands. Renewal rate can't meaningfully move until a full contract cycle has turned, so keep it off the short-term scoreboard. A reasonable check-in is four to six weeks for behavior signals and one to three months for adoption.

Does this only work for service agreements?

No. The same weighted composite drives any recurring outcome — extended warranties, memberships, monitoring subscriptions, inspection contracts, consumables replenishment. You change the KPI list and the weights; the machinery is identical. It also works on adjacent roles, like scoring dispatchers on first-time-fix or a call center on agreement leads generated.

Sources

• https://hbr.org/2012/07/motivating-salespeople-what-really-works
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-b2b-digital-inflection-point-how-sales-have-changed-during-covid-19
• https://www.acca.org/
• https://www.bls.gov/ooh/installation-maintenance-and-repair/heating-air-conditioning-and-refrigeration-mechanics-and-installers.htm
• https://www.salesforce.com/products/field-service/overview/
• https://www.servicetitan.com/
• https://www.gartner.com/en/sales/topics/sales-performance-management
• https://sloanreview.mit.edu/article/the-hidden-costs-of-sales-incentives/
• https://www.quotapath.com/
• https://www.xactlycorp.com/

Related on PULSE

• [How Do I Know Where, When, and How Many People to Schedule at Each of My Multi-Unit Retail Locations?](/knowledge/tl0001)
• [How Do I Figure Out How Many People to Schedule Each Day and at What Times for My Single Store?](/knowledge/tl0002)
• [How Do I Know How Many Cooks and Servers to Schedule Each Shift at My Pizza Restaurant?](/knowledge/tl0003)
• [How Many Salespeople Should I Schedule Each Day on My Furniture Store Floor?](/knowledge/tl0004)
• [How Do I Decide How Many Reps to Schedule at Each Store in My Mattress Retail Chain?](/knowledge/tl0005)