How Do I Get My Telecom Reps to Sell Bundles?

The end-to-end process from KPI inventory to a paid composite

Bundling failures are almost never a training problem. Reps in telecom stores can recite the fiber tiers and the protection plan tiers in their sleep. They skip broadband and security because those motions cost more minutes per sale, carry more objection surface, and — critically — because the number on the wall behind them counts lines. Fixing that is a sequenced operations project, not a pep talk, and it runs in six stages.

Stage one is the KPI inventory. Write down every revenue-bearing motion a complete sale contains. In a typical retail telecom footprint that's wireless activations, broadband or fiber, voice or VoIP, home or business security, equipment protection plans, device upgrades, and accessory attach. Anything you leave off the sheet is a service reps quietly stop pitching by the end of the week, because there's no reason to spend the extra six minutes on a motion nobody grades. Include the awkward ones — the security attach nobody likes pitching is precisely the one that needs a weight.

Stage two is weighting. Sit with leadership and hang a number on each KPI reflecting real strategic value: margin, contract length, churn resistance, and whatever the carrier is currently pushing through spiffs. Fiber usually earns a heavy weight because it's high-margin recurring revenue with a long install-to-churn horizon. Accessory attach earns a light weight — it matters, but a case sale shouldn't move a rep's standing as much as a two-year fiber contract. Keep the single largest weight under roughly 30 percent of the total, or you've just built a different single-metric scoreboard with extra steps.

Stage three is rating. Score each rep 1 to 5 on each line, honestly. A 1 means the motion essentially doesn't happen. A 5 means it happens on nearly every eligible account. The point of the 1-to-5 band rather than raw counts is that it normalizes across traffic volume — a rep in a low-footfall store isn't punished for the mall's parking situation, and a rep in a flagship location can't coast on walk-in volume alone.

Stage four is the composite. Sum weight × level across every KPI. That single formula is the whole mechanism. A rep at level 5 on activations but level 1 on broadband, security, and protection can no longer hide inside a fat activation count. The composite drags them down and points at the exact three lines they're skipping — converting "he doesn't really bundle" from a vague manager impression into a specific, arguable, coachable gap with a number attached.

Stage five is anchoring pay and coaching to the composite. This is where most implementations die. If the scorecard is published but commission still pays flat per line, reps read the scorecard as a suggestion and the pay plan as the truth — and they're right. The meaningful money has to track the composite. That doesn't require blowing up the whole comp plan on day one; a common on-ramp is to route 20 to 30 percent of variable pay through a composite multiplier and leave the rest on existing per-unit rates, then shift the ratio over two or three quarters as trust builds.

Stage six is the re-weight loop. Carriers move. A fiber promo lands mid-month, a security partner rewrites its terms, a device launch eats every Saturday. Because you own the weights, you re-weight overnight and the floor pivots by the opening shift. The one discipline: announce the change before it takes effect. A re-weight reps discover after the fact reads as a pay cut, and that's the fastest way to burn the whole system.

The loop is deliberately circular. A scorecard built once and frozen becomes stale within a quarter, and a stale matrix is worse than no matrix — it actively steers reps toward last quarter's priorities.

Where bundling creates revenue and where it quietly leaks

The financial case for bundling isn't really about the incremental ARPU on the second service, though that's the number most operators lead with. It's about the churn math underneath. Every additional service on an account raises the switching cost — porting a phone number is a fifteen-minute errand, unwinding a phone, fiber, and monitored security relationship means coordinating three installs, three cancellations, and probably a truck roll. That friction is the asset. A single-service account is a rental; a four-service account is closer to an annuity.

Where the revenue actually gets created is at three points in the account lifecycle. At point of sale the attach is nearly free — the customer is already sitting in the chair, already in a buying posture, already handing over ID and payment. At upgrade the account reopens, and a device upgrade is the single best natural excuse to audit what else the household is paying for elsewhere. At service events — a repair, a plan change, a billing call — the rep has permission to review the account, and reviewing the account is how the fiber gap gets found.

Now the leaks, which are more interesting because they're where operations work actually pays.

The eligibility leak. Reps don't pitch fiber to addresses they assume aren't serviceable, and they're frequently wrong. Footprints expand quarterly and the rep's mental map is from whenever they last checked. If your point-of-sale system doesn't surface serviceability automatically at the address field, a real share of eligible accounts never hear the offer. Fix this at the tooling layer, not the coaching layer — no amount of scorecard pressure overcomes a rep who genuinely believes the offer doesn't exist for that customer.

The provisioning leak. A broadband or security sale that clears in the store but falls out during install scheduling looks like a sale to the rep and like nothing to finance. If the scorecard credits the order and the pay plan credits the install, reps see a phantom gap between what they sold and what they got paid, and they lose faith in the whole apparatus. Decide explicitly which event counts — order, install, or 30-day survival — and make the scorecard and the comp plan agree on it.

The bundle-discount leak. Bundles usually carry a promotional discount. If reps are graded purely on service count and not at all on realized revenue, you get the classic RevOps failure mode: attach rate climbs, revenue per account doesn't, and margin actually declines because you've discounted services the customer would have bought anyway. Guard against this by weighting on margin contribution rather than raw unit count where your data supports it, and by auditing the discount rate on bundled versus unbundled accounts quarterly.

The queue leak. A full bundle sale takes materially longer than a phone-only sale. In a busy store with three reps and eleven people waiting, the rep who bundles is the rep whose queue backs up — and if you're also grading on transactions per hour or customer wait time, you've built two KPIs that fight each other. This is the most common structural contradiction in retail telecom scorecards. Either accept the throughput hit explicitly, or staff for it, or move the bundle conversation to a scheduled follow-up appointment where the clock isn't running against the rep.

The downstream cost leak. Bundled accounts generate more support contacts, more truck rolls, and more billing complexity. If the bundle push isn't coordinated with care and field operations, you win the sales number and lose it back in cost-to-serve and early churn from bad install experiences. Loop customer success into the weight-setting conversation for exactly this reason — they know which service combinations generate the support tickets.

Concrete numbers, benchmarks, and how to instrument them

Rather than borrow external benchmarks that may not describe your footprint, the more useful move is to instrument your own baseline before you change anything, because the deltas are what you'll defend in the QBR. Pull ninety days of history and compute five figures.

Services per account at point of sale. Not the installed base average — the average count of billable services attached to newly opened accounts, measured at sale. This is your true bundling rate and it's usually lower than leadership assumes, because the installed base number is inflated by years of accumulated adds. Track it as a distribution, not just a mean: knowing that 60 percent of new accounts are single-service tells you more than knowing the average is 1.8.

Attach rate per service, per rep. For each secondary service, the percentage of eligible accounts that took it. The word *eligible* does the heavy lifting — a rep can't be graded on fiber attach in a non-serviceable address, and if you don't strip ineligible accounts out of the denominator you'll produce a scorecard reps can correctly call unfair, which kills adoption on day one.

The spread between your top and bottom quartile. This is the number that sizes the opportunity. If your top quartile attaches broadband at three times the rate of your bottom quartile in comparable stores, the gap is behavioral and coachable, and closing half of it is a concrete revenue target. If the spread is narrow, your constraint is upstream — pricing, serviceability, or product — and a scorecard won't fix it. Run this check before you build anything.

Revenue per account and margin per account, bundled versus unbundled. Run both. If bundled accounts show higher revenue but flat margin, your promotional discount is eating the entire gain and the bundle strategy is a wash financially even as the attach numbers look heroic.

Retention curve by service count. Cohort accounts by number of services at open and track survival at 6, 12, and 24 months. This is the number that justifies the entire program to finance, because it converts attach rate into lifetime value rather than a one-month revenue bump.

On the scorecard mechanics themselves, a few practical calibrations. Use a 1-to-5 level band rather than raw percentages — it's coarse enough that reps trust it and fine enough to show movement. Set the level thresholds off your own quartiles: level 3 is the median performer, level 5 is roughly top-decile, level 1 is bottom quartile. Recalibrate the thresholds every couple of quarters, because a static threshold turns into a participation trophy once the floor improves.

Keep the KPI count between six and nine. Below six you're not really measuring a bundle; above nine reps can't hold the matrix in their head, and a scorecard nobody can recite is a scorecard nobody optimizes against. Keep the largest single weight under about 30 percent of the total so no one line dominates. And publish the full matrix — every rep should be able to see their own levels and every weight without asking a manager. Opacity is what makes scorecards feel like surveillance instead of a map.

For the pay side, model the composite multiplier against last quarter's actuals before you launch it. Run every rep's historical numbers through the proposed plan and look at the distribution of outcomes. If nobody's pay changes, the weights are too flat to drive behavior. If half the floor takes a meaningful cut, you've built a pay cut wearing a scorecard costume and you'll lose your best people to the carrier store across the parking lot. The target is a plan where the reps already bundling well come out modestly ahead and the phone-only reps see a clear, achievable path back to their current earnings.

Pitfalls that kill bundle scorecards, and the fixes

Launching the scorecard without touching comp. The single most common failure. Reps compare what the scorecard says to what the paycheck does, and the paycheck wins every time. If you can't move comp in the first cycle, at minimum attach a real spiff to the composite so there's money — any money — flowing through the new mechanism. A scorecard with no financial consequence is a poster.

Grading reps on services they can't sell. If a rep's role or location doesn't carry security or business voice, those KPIs carry zero weight in that role's version of the matrix. Run role-specific weight profiles — retail, inside, field, business — off the same KPI inventory. Same framework, different dials. Grading someone on a product they have no path to sell is the fastest way to get the whole system dismissed as head-office nonsense.

Overweighting one line. Reps will absolutely stampede toward the heaviest weight, and that's the intended mechanic — but if fiber carries 45 percent of the weight you've rebuilt a single-metric scoreboard and reps will drop protection and accessories entirely. Cap the top weight and audit the actual behavior mix each month against the intended mix.

Counting orders when comp pays installs. Covered above but worth restating as a pitfall, because it's the one that generates the most floor-level resentment. Pick one event and make every system agree.

Silent re-weights. Changing weights without announcement reads as a bait-and-switch even when it isn't. Announce, state the effective date, and never apply retroactively to a period reps already worked.

No coaching attached to the gap. The matrix tells you a rep is a level 1 on security attach. That's diagnosis, not treatment. Without a specific intervention — a call review, a ride-along, a scripted transition line, a paired shift with the top attacher — the rep sits at level 1 and just earns less, which produces resentment rather than behavior change. Budget the coaching time explicitly when you launch; the scorecard's value is that it tells you exactly where to spend it.

Manual data entry. If managers key bundle numbers into a spreadsheet at close every night, the matrix will be stale within a month and wrong within two. Feed it from the order system. Whatever tooling you pick, vet the integration before you sign — confirm it ingests wireless lines, broadband, and security attach straight from the source of record.

Ignoring the customer experience side. A bundle sold under pressure to a customer who didn't want it shows up sixty days later as a cancellation, a chargeback on the rep's commission, and a bad review. Pair the composite with a retention or 60-day-survival guardrail so the scorecard rewards bundles that stick, not bundles that book.

Choosing the tooling and rolling it out

The matrix is the product; software is just the delivery surface. Build the KPI inventory and the weights *before* you shop, because every category of tool performs better against a matrix that already exists — and shopping first tends to let a vendor's data model dictate your priorities.

Tools split into three functional camps. Visibility platforms — sales scorecard, leaderboard, and gamification products — put multi-metric performance on store TVs and in chat, which is what keeps softer motions like broadband and security front-of-mind during a chaotic Saturday. Compensation platforms — incentive-comp tools that administer multi-component plans — are where the matrix grows teeth, paying distinct rates on lines, broadband, security, and protection with the audit trail finance needs. CRM-native scorecards host the weighted view right next to the pipeline using data already in the object model; nothing is gift-wrapped, but every ingredient the composite needs is already there. A conversation-intelligence layer is a useful fourth complement for inside teams — it reveals whether reps are even putting broadband and security on the table before they close, which raw counts can never show.

At the entry end, a cleanly built spreadsheet costs nothing and hides nothing: list the KPIs, dial the weights, rate 1 to 5, let one formula roll the composite. The real cost is your time to maintain it and the standing risk of a sheet nobody has touched since the last promo. Plenty of teams start here deliberately and graduate once the model is proven.

Decide where the teeth belong — visibility, pay, or a deliberate stack of both — and keep the weights in your own hands rather than a vendor's, so you can swing priorities overnight when a carrier package moves. Then pilot on one store or pod, not the district. Prove the weighted matrix lifts broadband and security attach on a single floor for a full quarter against a comparable control store, and roll the same weights and payout rates wide only once the gain clearly holds.

Sequence the rollout over roughly a quarter: weeks one and two to build and socialize the matrix, weeks three and four to run it visibly with no pay impact so reps can see their own numbers without risk, then switch on the composite component of pay at the start of the next full comp period. Never switch on mid-period. And publish a worked example on day one — show a phone-only rep landing under target while a rep who folds in broadband and protection clears it comfortably. Veteran reps flip the moment they see their own earnings ceiling move up, and they flip against you the moment the math looks like a trick.

Related questions

How is this different from a plain commission plan?

A commission plan pays per unit and is blind to mix. The weighted composite grades the *balance* of an account, so a rep can't max earnings on one easy motion. Comp still pays the money — the composite decides how the mix is valued before it does.

Should retail and inside telecom teams use the same weights?

Same KPI framework, different weight profiles. Inside teams typically weight voice and business broadband higher; retail weights device upgrades and accessory attach higher. Run role-specific profiles off one shared inventory so the composite stays comparable in structure, if not in dials.

Does this apply outside telecom?

Yes — any multi-product sale with an easy anchor product has the same failure mode. Cable, insurance, banking, and equipment dealers all see reps park on the high-volume line. The weighted composite is a RevOps pattern, not a telecom-specific one.

How long before attach rates actually move?

Expect one full comp period of noise, then measurable movement in the second. Behavior change follows the first paycheck that visibly reflects the composite, not the launch meeting. Judge the pilot on a full quarter against a control store, never on the first month.

What if leadership won't change the pay plan?

Start with a composite-based spiff pool funded from existing promo budget. It proves the mechanism with real money at small scale, and the attach-rate data from that pilot is the argument that gets the base plan changed next cycle.

FAQ

Does this work for any size telecom team?

It does — the weighted scorecard behaves identically whether you run a five-person kiosk or a five-hundred-rep region. The only thing that scales is the delivery surface: a small crew can live inside one shared spreadsheet reviewed at Monday huddle, while a large region needs a real dashboard fed from the order system to keep the matrix current. The underlying logic of weighting each service line and rolling to a composite never changes.

How often should I update the weights?

Re-weight whenever the business moves under you — a carrier promo drops, a package gets restructured, a strategic priority flips — which in practice can mean monthly or occasionally weekly. The one non-negotiable discipline is announcing the change before it takes effect, so reps know what they're chasing going in. Overnight re-weights are completely fine as long as the refreshed matrix is visible and explained at the next opening shift.

Won't reps just chase the highest-weight item and ignore the rest?

Partly, and that's the intended mechanic — you want them moving toward the lines that carry real bundle value. The guardrail is structural: never let one KPI's weight dominate, so keep the largest single weight under roughly 30 percent of the total. Then audit the actual mix monthly. If protection and accessories have collapsed to zero, your weights are too lopsided and need flattening.

What if my reps can't sell broadband or security?

Those KPIs carry zero weight in that role's version of the matrix. The scorecard should only ever grade services a rep has a genuine path to sell. If a rep is limited to wireless and voice, grade exactly those two — the composite still rewards selling them together rather than leaning on one. Serviceability gaps get handled the same way: strip ineligible accounts out of the denominator.

How do I get buy-in from veteran reps who hate new scorecards?

Show the math on a real example rather than arguing the philosophy: a phone-only rep clearing well under target while a rep who folds in broadband and protection clears it comfortably. Then run a full trial period with the matrix visible but no impact on the check. Most veterans flip once they see their own earnings ceiling move up — the objection is almost always about risk, not about measurement.

Is this just a more sophisticated form of micromanagement?

It's the opposite in practice. It replaces a manager's subjective read of who "really bundles" with a transparent number every rep can inspect. Reps see precisely which action moves their score and self-correct without a supervisor hovering. The manager's job shifts from policing the floor to helping one rep lift one specific weak KPI — which is a better use of a manager and a far less irritating experience for the rep.

Sources

• https://hbr.org/2011/12/how-to-really-motivate-salespeople
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/the-secret-to-making-it-in-the-digital-sales-world
• https://www.bain.com/insights/topics/customer-loyalty/
• https://www.salesforce.com/resources/articles/sales-performance-management/
• https://www.gartner.com/en/sales/topics/sales-performance-management
• https://www.fcc.gov/general/measuring-broadband-america
• https://www2.deloitte.com/us/en/insights/industry/technology/technology-media-and-telecom-predictions.html
• https://www.pwc.com/us/en/industries/tmt.html

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)