How Do I Motivate Reps to Sell More Than the Core Product in 2026?

The floor that hit quota every month and still lost margin

Picture a twenty-rep floor selling a hardware-plus-service package. The core unit carries roughly a 22% margin. The service plan attached to it carries something closer to 60%. Accessories land somewhere in between, and the multi-year support agreement is the highest-margin line on the sheet. Every month the floor hits its revenue number. Every month the CFO asks why gross profit is flat while top-line revenue climbs.

Pull the rep-level report and the pattern shows up immediately. Two or three reps are attaching service plans on most deals. The rest are attaching on maybe one in five. Nobody is selling the multi-year agreement at all, because it requires a longer conversation, a pricing exception, and a second signature. The floor is not lazy. The floor is optimizing perfectly against the only number anyone publishes: total revenue closed.

This is the moment where most leaders reach for a speech. They call an all-hands, they talk about "solution selling," they put a poster on the wall about the customer journey. Two weeks later attach rate is exactly where it was. The speech failed because it never touched the scoreboard. A rep who can hit their number by selling one comfortable product will keep selling one comfortable product, and no amount of enthusiasm changes that arithmetic.

The reframe that actually moves the floor is this: you do not have a motivation problem, you have a measurement problem. Motivation is downstream of measurement. Publish one number and you get behavior aimed at that one number. Publish a weighted composite and you get behavior aimed at the composite. The rep did not change. The target changed.

Notice how far this extends past software sales. A car dealership that only tracks units moved gets a floor that ignores the finance and insurance office, where the actual profit lives. A gym that only tracks new memberships gets a staff that never sells personal training packages. A managed services provider that only tracks new logos gets account managers who never expand a footprint they already own. Same failure mode, different vertical: a single published metric, and a portfolio of neglected margin sitting right next to it.

The scenario also reveals a second, quieter problem. The reps who *are* attaching service plans are usually doing it through personal habit, not process. They developed a phrasing that works, they use it every time, and nobody has ever asked them to write it down. That knowledge sits in three heads on a floor of twenty. When one of those three leaves, the attach rate for their territory falls off a cliff and nobody can explain why. Fixing the scoreboard without capturing that tacit knowledge just tells eighteen reps to do something they have never seen done well.

How the weighted scorecard actually changes behavior

The mechanism is a weighted multi-KPI composite. You list every product line and behavior a complete rep should produce — usually eight or nine lines, because fewer than six is too coarse to distinguish anything and more than ten stops being legible on a wall. Core product revenue. Add-on attach. Accessory attach. Service plan attach. Retention or renewal on the book they own. Offer rate. Activity or pipeline coverage. Maybe referral generation, depending on your motion.

Each line gets a weight, set with leadership rather than by the sales manager alone, because weights are a strategy statement and strategy belongs to the executive table. Each line also gets a level from 1 to 5, where 3 is "meets expectation for a rep at this tenure" and the anchors for 1 and 5 are written down in plain language before anyone is scored. The composite is the sum of weight times level across every KPI.

The arithmetic does the work a speech cannot. A rep at level 5 on core product and level 1 on everything else lands at a low composite, and the gap is visible to that rep, to their peers, and to leadership on the same sheet. There is nowhere to hide behind a big single-product month. A rep at level 3 across the whole book — solid, not spectacular, on every line — outscores the specialist. That inversion is the entire point, and reps understand it the first time they see the numbers.

The second mechanical property matters just as much: weights are editable overnight. When a product launches, when a supplier changes terms, when margin shifts, you re-weight and republish, and the floor re-aims the next morning. There is no re-education campaign, no new comp plan document, no quarter-long change management project. The math changed, everyone can see it changed, and behavior follows within a week or two.

Publishing is not optional, and it is the step most leaders quietly skip. A private scorecard is a performance review. A published scorecard is a market signal. When every rep can see every other rep's composite and the weights that produced it, three things happen: the top performer's method becomes an object of curiosity rather than resentment, the bottom performer knows exactly which line to work on, and the manager stops being the sole source of feedback. The matrix argues on the manager's behalf.

One structural caution on the mechanism itself. If your comp plan pays a flat rate on all revenue while your scorecard weights add-ons at three times core, you have built two contradictory systems and the wallet wins. The scorecard has to feed the money. That can be a composite multiplier on the variable component, a tier gate where a certain composite unlocks a higher accelerator, or a direct per-line commission schedule that mirrors the weights. Any of those works. What does not work is a scorecard that lives on a wall while comp lives on a different logic entirely.

Real numbers, ranges, and what to expect

Set expectations with ranges rather than promises, because attach economics vary enormously by category and nobody's benchmark transfers cleanly to your floor. What follows is how to build your own numbers rather than borrow someone else's.

Start from your own baseline, measured three ways. Pull ninety days of closed deals and compute attach rate as (deals with the add-on) divided by (deals eligible for the add-on). Eligibility matters — if a third of your deals cannot take the service plan for contractual reasons, including them in the denominator makes every rep look worse and hides the real spread. Then compute the same rate per rep, and then the rate per rep per product line. The three views usually tell three different stories, and the per-rep-per-line view is the one that shows you whether you have a training problem, a friction problem, or a comp problem.

Look at the spread, not the average. A floor averaging 30% attach is not one thing. It might be twenty reps all sitting near 30%, which is a systemic problem — likely friction or pricing. Or it might be four reps at 75% and sixteen at 15%, which is a knowledge-transfer problem with a ready-made solution sitting in your top four reps' call recordings. The average is identical; the intervention is completely different.

Weight by margin, not by revenue. The most common weighting error is assigning weights proportional to the revenue each line produces, which just re-creates the total-revenue scoreboard with extra steps. Weight by gross profit contribution and by strategic priority. If a service plan carries roughly triple the margin of the core unit, its weight should reflect that even though its ticket price is a fraction of the core.

Cap the number of scored lines at nine and revisit weights quarterly. More lines dilute attention until nothing is prioritized. Weights revisited more often than quarterly make the target feel unstable and reps stop trusting it; revisited less often and the matrix drifts away from where the business actually is. Quarterly is the compromise most floors settle on, with an exception clause for genuine market shocks.

Give any change a full sales cycle before you judge it. If your average cycle is three weeks, look at week six. If it is four months, do not draw conclusions in month two. The most common way a good scorecard gets killed is a leader declaring it a failure before a single cohort of deals influenced by it has had time to close.

Track offer rate separately from close rate on add-ons. These are different skills with different fixes. Low offer rate is a confidence or friction problem — the rep never brings it up. Low close rate on offers made is a positioning or pricing problem — the rep brings it up and loses. Fixing the second when you have the first wastes a quarter.

Instrument the time cost. Time how long it actually takes a rep to add a service plan to an order — with a stopwatch, watching a real rep do it, not by asking. If the answer is more than a minute or two of extra clicking, form-filling, and approval-chasing, you have found a bigger lever than any incentive. Reps do the arithmetic on effort-to-reward without being asked, and a small add-on commission does not survive contact with a three-screen approval workflow.

Budget for the recognition, not just the commission. Spiffs and contests have a real cost, and it is worth deciding the number in advance rather than improvising monthly. Whatever you set, keep it consistent — an incentive that appears and disappears teaches reps to wait rather than to change habits.

Trade-offs, alternatives, and where each one breaks

The weighted scorecard is not the only way to solve this, and it has genuine costs. Here is the honest comparison.

Pure commission re-rating. Simply pay a much higher commission rate on add-ons and let the money do the talking. This is fast, requires no new system, and reps understand it instantly. It breaks when the behavior you want is not directly commissionable — offer rate, retention, discovery quality — and when reps discover that discounting the core to force an add-on still pays them well while destroying blended margin. It is also blunt: you cannot re-aim it without opening the comp plan, which is a legal and administrative event, not an overnight change.

Spiffs and short contests. Cheap, energizing, and genuinely effective at moving a specific number for a specific month. They break as a permanent strategy because reps learn to save deals for contest weeks, the lift disappears the day the contest ends, and the floor develops a tolerance that requires ever-larger prizes for the same effect. Use them to launch a behavior, not to sustain one.

Manager-enforced process. Require the add-on conversation in every deal review, inspect it, coach it. This works and it is the most durable of the alternatives, but it scales with manager quality and manager time. On a floor with one strong manager and two weak ones, you get three different attach rates and no way to tell whether the difference is territory or coaching.

Product bundling. Stop selling the add-on separately and make it part of the standard configuration. This is the most reliable of all — attach rate approaches 100% by definition — but it is a pricing and packaging decision that sits outside the sales org, it removes optionality that some customers genuinely need, and it can raise the entry price enough to cost you deals at the top of the funnel.

The weighted scorecard. Flexible, re-aimable overnight, covers non-commissionable behavior, exposes gaps that revenue hides. Its costs are real: it takes real work to build, it requires leadership to actually agree on weights (which surfaces disagreements people were happy to leave buried), it needs clean data to score honestly, and it fails completely if it is published but not connected to compensation. A scorecard with no money behind it is a poster.

Most floors need two or three of these at once. The friction fix and the scorecard pair especially well: the scorecard tells the rep what to do, and the friction fix makes doing it cheap. Running the scorecard alone on a system that takes three minutes to add a service plan just tells reps to do something the software punishes them for.

There is also a sequencing argument. Fix friction before you change comp. Friction fixes are cheap, uncontroversial, and reversible; comp changes are expensive, political, and hard to walk back. If a one-click bundle recovers a meaningful share of the gap on its own, you have just made the comp conversation much smaller.

Where this goes wrong, and how to keep it from going wrong

Weights set by one person. If the sales manager sets weights alone, the matrix becomes that manager's opinion with a spreadsheet around it, and the first time a rep disagrees it turns into a personality conflict. Set weights with the executive team, document the reasoning in one paragraph per line, and publish that reasoning alongside the weights. When a rep asks why service plans are weighted heavily, the answer should be a margin fact, not a preference.

Scoring on dirty data. If attach rate is computed from a CRM field reps fill in manually, they will fill it in favorably, and within a quarter your matrix is measuring data-entry behavior rather than selling behavior. Score from the system of record for the transaction — the order, the invoice, the signed contract — not from a self-reported field.

Punishing the attempt. The most damaging pitfall. If a rep offers the full stack and the customer says no, and the only thing the scorecard records is a miss, reps rapidly learn that offering is risky and stop offering. Track offer rate as its own weighted line so the attempt earns credit. A rep offering to 90% of eligible customers and closing 30% of those offers should outscore a rep who never offers and never fails. This single change flips the internal calculation from fear of rejection to credit for effort, and it is usually the fastest-acting item on this list.

Calling it an upsell in internal language. The word carries a whiff of extraction, and reps who think of themselves as consultants resist it. Use "completion" or "solution layer" instead, and frame a core-only sale as leaving the customer under-served rather than as leaving money on the table. That is not spin — a customer who buys a system without the support plan genuinely has a worse outcome when something breaks. Reps sell what they believe.

No script, so no confidence. Most reps avoid add-ons because they do not know how to raise them without sounding transactional. Give them three sentences: what most customers do, what the add-on changes, and one concrete consequence in the first month. Role-play it in a weekly huddle, record the versions that land, and keep the recordings somewhere reps can find them. Competence produces motivation far more reliably than motivation produces competence.

No cheat card at the point of sale. One index card per add-on: the problem it solves in a sentence, the customer-side value in a sentence, the most common objection with a rebuttal. Put it in the CRM as a popup on the relevant screen, not in a shared drive nobody opens. A rep who can answer "why should I buy this?" in under ten seconds will offer it far more often than one who has to go looking.

Recognizing revenue instead of process. When you celebrate only the biggest deal, you teach the floor that size is the goal. Celebrate the complete sale specifically — name the rep, name the layers they attached, name the objection they handled. Social currency for comprehensiveness is what shifts the floor's definition of a good rep, and it costs nothing.

Letting the top performers' method stay tacit. Designate a rep each quarter who has genuinely mastered the full-stack sale and give them fifteen minutes in a team meeting — not a lecture, a walkthrough of one real deal, start to finish. Peer proof outperforms manager instruction because it removes the "easy for you to say" objection. Pair a low scorer with a high scorer for a week of joint calls; both usually improve.

Changing weights too often. Overnight re-weighting is a capability, not a habit. If reps cannot predict the target from one month to the next, they stop optimizing against it and revert to whatever pays reliably. Quarterly, with a written rationale, keeps the flexibility without destroying trust.

Shipping the matrix without the money. A published composite that has no effect on anyone's paycheck is a wall decoration. Reps will read it for a month and then ignore it. Decide up front how the composite touches variable pay — multiplier, tier gate, or per-line rates — and ship both changes together.

Judging it too early. Give the change a full sales cycle plus a few weeks before you assess it. Killing a working system in week three because the numbers have not moved is how organizations end up cycling through five motivational programs in two years and concluding that none of them work.

Related questions

Does this work for account managers and customer success, not just new-business reps?

Yes, and often better. Expansion and renewal are naturally multi-line, so a composite covering retention, expansion revenue, product adoption depth, and reference generation fits the role more honestly than a single renewal-rate number ever did.

What if my CRM cannot compute a weighted composite?

Score it in a spreadsheet for the first quarter. The matrix is arithmetic — weights times levels, summed. Prove it changes behavior before you pay for tooling or ask for engineering time to automate the report.

How do I handle reps who game the weights?

Gaming means the weights reward something you did not intend, which is a design bug, not a character flaw. Watch for discount-to-attach behavior specifically, add a margin or discount line to the matrix, and the exploit closes.

Should the composite affect base pay or just variable?

Variable. Base pay changes are slow, hard to reverse, and legally fraught. A composite multiplier on the variable component moves behavior quickly and can be re-tuned each quarter without reopening anyone's employment terms.

Does public ranking demotivate the bottom of the floor?

It can, if the only public artifact is a rank. Publish the composite alongside the per-line detail so a low scorer sees a specific fixable gap rather than a verdict, and pair the publication with a named path to improvement.

FAQ

What if reps only focus on the core product because it is easiest?

That is exactly the behavior a weighted composite corrects. When total revenue is the only published metric, chasing the fastest dollar is the rational move. Assign weights to add-ons, service plans, retention, and offer rate, and the easy single-product path stops producing a strong score — or a strong paycheck.

How do I decide which KPIs to include and how to weight them?

Start with leadership alignment on eight or nine lines spanning core product, attach, service plans, retention, and activity. Weight by gross margin contribution and strategic priority rather than by revenue size, so a high-margin service plan carries real weight despite a small ticket. Document one sentence of reasoning per weight and publish it with the matrix.

Will reps feel overwhelmed tracking that many metrics?

Not if the presentation is simple. A 1-to-5 level per line with written anchors keeps scoring legible, and the composite collapses everything into one number a rep can watch. Publishing the full matrix replaces guesswork with clarity — reps see precisely which line moves their pay most.

What about a rep who is excellent on the core product and weak everywhere else?

The composite surfaces that immediately. Level 5 on core with level 1 elsewhere produces a low total no matter how large the core number looks in isolation. Because variable pay follows the composite, the gap has financial consequence — and the per-line detail hands that rep a specific improvement path rather than a vague instruction to broaden.

How fast can I change the weights when priorities shift?

Overnight. The matrix is a published arithmetic rule, so you update the weights, republish, and the floor re-aims within a day or two. Use that capability sparingly — quarterly adjustments with written rationale, exceptions for genuine market shocks. Constant re-weighting teaches reps the target is unstable and they stop optimizing against it.

Do I need software to run this?

No. A spreadsheet runs a weighted composite for a floor of twenty reps without difficulty, and starting there forces you to get the weights and anchors right before automating. Once the behavior change is proven, move the scoring into whatever system already holds your transaction data so the numbers stop depending on manual entry.

Sources

• https://hbr.org/2012/07/motivating-salespeople-what-really-works
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights
• https://www.gartner.com/en/sales/topics/sales-performance-management
• https://www.shrm.org/topics-tools/topics/compensation
• https://www.salesforce.com/resources/articles/cross-selling-and-upselling/
• https://www.worldatwork.org/resources/publications
• https://www.bain.com/insights/topics/customer-strategy-and-marketing/
• https://www.sec.gov/edgar/searchedgar/companysearch

Related on PULSE

• How Do I Tie Commission to More Than One Product?
• How Do I Score My AEs on More Than Just Closed-Won?
• How Do I Get My Reps to Sell the New Product Line?
• How Do I Get My Sales Reps to Sell the Full Product Line Instead of Just One or Two Products?
• How Do I Get My Real Estate Agents to Capture More Referrals?