Last updated: July 2026
Good reps usually stay when the pay plan is understandable, the role is winnable, account decisions are fair, the support lets them sell and they can see a future. Most of the time a resignation is the end of a slow build, not a sudden decision. Audit those conditions while the rep is still happy, rather than reaching for a counter-offer once they have already taken a call.
I place sales reps into building products manufacturers across Australia. BDMs, Account Managers, Specification Managers, State Managers. I see who leaves, who stays, and what the companies that hold onto good people do differently. This is the audit I would run before assuming money is the answer.
The eight-lever retention audit at a glance
Work through the levers in this order. The point is not a return-on-investment league table. It is a sequence that moves from the conditions you can fix cheaply to the ones that take real structural work.
| Lever | Problem it detects | Evidence to check | First action | Measure after 90 days |
|---|---|---|---|---|
| 1. Comp plan transparency | Rep assumes they are underpaid because the plan is opaque | Can the rep model their own outcome without asking finance | Publish the plan in writing with worked examples | Reps stop guessing what they earn at each level of target |
| 2. Uncapped upside with controls | A cap tells your best rep to slow down | Where the plan caps or the accelerators flatten | Replace caps with accelerators inside clear controls | Top-quartile reps are not leaving money-capped |
| 3. Territory decision rights | Accounts get reallocated without a conversation | How account credit and shared accounts are decided | Write down decision rights and credit rules | No account moves without a documented conversation |
| 4. Career path clarity | Rep stops seeing a next step | Can each senior rep name their next move | Write the roles, the gates and the review cadence | Every senior rep can name their next step |
| 5. Role redesign | A good person is stuck in a role that no longer fits | Whether the role still matches the person | Build a role around their strengths, not a title | The person is doing more of their best work |
| 6. Recognition | Recognition is transactional or absent | Can the manager point to specific recent recognition | Make it regular, specific and from the manager | Recognition given in the last month is visible |
| 7. Package calibration and non-salary options | Market moves faster than the annual review | Whether pay and terms match the current market | Calibrate twice a year, add non-salary options | No strong rep discovers the market moved before you do |
| 8. Service load and operational support | Number looks fine but the role is unwinnable | How much of the week actually reaches a customer | Fix the support and systems that block selling | Selling time as a share of the week is improving |
Is it a pay problem, a role-design problem, a manager problem or a support problem?
In my experience the real reason a good rep leaves is that they are unhappy. It is normally something to do with the culture or the system of operations around the role. Salary is part of the picture. More often the other stuff gets them to the point of taking a call, and pay is the reason they give on the way out.
So before you assume it is money, work out which of four things is actually broken.
Pay problem. The plan caps their upside, the territory has been carved so the same effort earns less, or their package has drifted behind the market. If salary genuinely is the key driver, say so and deal with it directly with the rep. Do not wait for the annual cycle or a resignation to have that conversation.
Role-design problem. The job has become harder to win in. Support failures, an unreachable target, an account base that no longer produces. The rep is working harder for less.
Manager problem. When several reps under the same manager tell a recruiter the same thing, that is data, not coincidence. The gap between what a manager assumes and what the rep says out loud is often the whole story.
Support problem. The rep spends the week on service issues, quoting and order chasing instead of selling. Their number can still look acceptable while the role quietly becomes unattractive.
Name the real problem first. A pay rise that papers over a broken role buys you a few months, not a career.
The eight retention levers
Each lever follows the same pattern: the failure mode, the action, the guardrail and what to measure.
1. Comp plan transparency
Failure mode. The rep knows their on-target earnings. They do not know what happens at 120% or 140% of target, how the threshold is set this year, or whether the bonus pool has been adjusted. So they assume the worst.
Action. Publish the plan in writing, with worked examples at 80%, 100%, 120% and 140% of target.
Guardrail. The plan still has to define margin, returns, house accounts, shared credit and windfalls, so transparency does not become a promise to pay on bad revenue. How the mechanics actually work is covered in what your commission structure actually pays.
Measure. Reps can model their own outcome without asking finance.
2. Uncapped upside with defined controls
Failure mode. A cap tells your best rep to slow down. A rep running at 180% on a plan capped at 150% is being managed out.
Action. Replace caps with accelerators. As an illustration only, you might pay more above 120% of target and more again above 150%. You spend more in a great year and you keep the rep who produced it.
Guardrail. Set the controls first. Margin floors, shared credit on national and house accounts, inherited pipeline and delayed projects. Uncapped upside without controls just rewards the wrong deals. The full design logic is in designing a sales comp plan for a building products BDM.
Measure. Your top-quartile reps are not leaving capped-out conversations with recruiters.
3. Territory decision rights and account credit
Failure mode. A rep loses their two best accounts to a national-accounts structure and nobody has the conversation. The trust goes and does not come back.
Action. Write down decision rights and credit rules. Which accounts are owned, which are shared, what happens when a national customer wants a direct line. Review it with the rep in the room.
Guardrail. This is about clear decision rights and consultation, not contractual ownership of customers. A rep does not own a customer, but they should never be surprised by a reallocation.
Measure. No account moves without a documented conversation first.
4. Career path clarity
Failure mode. The structure is flat. The State Manager is 38 and going nowhere, the GM is the founder. The rep stops seeing a five-year future and starts taking calls.
Action. Write the actual roles, the gates and the review cadence. Where there is no management seat to offer, build paths into strategic accounts, technical ownership, mentoring or project leadership.
Guardrail. The path has to be real. Do not dress a sideways move up as a promotion.
Measure. Every senior rep can name their next step and what it takes to get there.
5. Role redesign
Failure mode. A good person is stuck in a role that no longer fits them, and a competitor offers one that does.
Action. I have seen employers keep or re-engage people by building a national role, a product-management role or a genuinely new position around the person's area of passion and expertise.
Guardrail. Redesign the work, not just the title. A new business card over the same broken role fixes nothing.
Measure. The person is doing more of what they are best at within two quarters.
6. Recognition
Failure mode. Recognition is transactional. A "great deal" email after a win and nothing in between.
Action. Make it regular, specific and from the direct manager, not saved for the CEO's speech at the annual conference.
Guardrail. Recognition does not repair broken pay, workload or management. It is one lever, not a substitute for fixing the others.
Measure. Managers can point to specific recognition given in the last month.
7. Package calibration and non-salary options
Failure mode. Annual reviews are too slow. By the time you review in November the rep has been on the wrong side of the market for months and has already taken a call.
Action. Calibrate twice a year against current market data, not your last hire's package. Beyond salary, the options I see hold people include a four-day week, extra annual leave and long-term incentives that genuinely pay and compound. The honest name for that last one is golden handcuffs, an incentive that vests over time so leaving means walking away from real money.
Guardrail. Twice-yearly calibration is a market and performance check, not a promise of two pay rises a year. Any incentive has to be transparent and fair, or it breeds the resentment it was meant to prevent.
Measure. No strong rep is discovering the market has moved before you do.
8. Service load and operational support
Failure mode. The quietest flight risk on a team is often the rep whose number still looks fine. Two reps in the same market can have very different results because one is supported to sell while the other spends the week on service issues, quoting and chasing orders.
Action. Measure how much of the week actually reaches a customer. Then fix the support, the systems and the internal execution that make the role unwinnable.
Guardrail. A healthy quota can hide a broken role. Do not wait for the number to drop before you look.
Measure. Selling time as a share of the week is tracked and moving in the right direction.
How the levers change by role
The same eight levers apply everywhere, but the first place to look shifts with the seat.
Specification reps. The risk is delayed and long-tailed. Protect the long project cycles and the specification influence they carry, because the loss shows up a year late when the whole downstream project goes with it.
Account managers. The risk is the relationship they built being reallocated without a conversation. Territory decision rights and account credit matter most here.
New-business BDMs. These leave first when the plan caps their upside or the territory is carved. Comp transparency and uncapped upside are the levers to check.
Sales leaders. They stay for scope, autonomy and a business they can shape. A flat structure and a founder who will not let go are the usual causes of a quiet exit.
Match the lever to the seat rather than running one retention policy across all of them.
How do you run a stay interview that actually helps?
The useful conversation happens six months before the resignation, not six hours after. Thirty minutes, run by the direct manager, every six months, built around three questions:
- What would make you take a call from a recruiter this quarter?
- What is the one thing you would change about the role?
- Where do you want to be in two years?
The reps you would most regret losing are usually the most articulate about what is making them consider leaving. The part that matters is what happens next. Acknowledge what you heard, investigate it, decide what you will and will not change, explain that plainly, and follow up with a name against each action and a date. When you cannot give the rep what they ask for, say so honestly and explain why. A stay interview that asks, records the concern and then does nothing does more harm than never asking.
What if a rep is already talking to someone else?
Prevention beats a counter-offer. Once a rep is in a live process, matching the money rarely fixes the reason they started looking. The territory is still carved, the progression is still blocked, the manager who would not pay them last year is only paying them now because they threatened to leave. If you want the case for why a top performer is worth this effort and what losing one actually costs, that is set out in why your best salesperson is your most undervalued asset.
What does a 12-month retention rollout look like?
Four quarters, in this order.
Q1, audit. Map current tenure, comp ranges and where each of the eight levers sits today. Identify the three weakest.
Q2, quick wins. Comp plan transparency, territory decision rights in writing and the first round of stay interviews. Low friction and they start producing signal inside 90 days.
Q3, structural. Career path documentation, any accelerator restructure, the twice-yearly calibration cycle and a real recognition cadence.
Q4, measurement. Track regretted attrition, tenure by role, internal moves, stay-interview actions actually closed, the change in selling time, quota attainability and time to productivity. Adjust the levers that did not move the metric. Avoid inventing benchmark targets. Judge yourself against your own baseline.
One honest note. Retention is not about keeping everyone. This framework is for capable people you have a sound reason to keep, not for holding onto a poor fit or standing in the way of a rational career move.
What losing a good rep actually costs
Losing a strong rep costs more than the fee, but the fee is a fair starting point. My current recruitment guidance is a fee of 15% to 22%, which is roughly $20,000 to $35,000 for a BDM. On top of that sit the vacancy, the onboarding and the ramp before a replacement is contributing. Then there is the part employers miss most often. A departure can take lost sales, lost specifications and lost customers with it, because in this market the relationship is a large part of the product. Those are separate risks, not one blended number.
To model your own figure, the cost-of-a-bad-hire calculator walks through it line by line. For the full methodology behind the cost, I set it out in the cost of a bad sales hire in building products. Both are there so this page does not have to repeat the maths.
If a strong rep is becoming hard to hold, the useful first step is to separate a pay problem from a role, support, territory or management problem before you assume recruitment is the only answer. That is work I can help with. The hiring managers page has the quickest way to reach me.