At the end of the day, sales compensation is an exercise in applied human behavior.

Change a measure, threshold, accelerator, or pay mix and you change the calculations sellers make on repeat throughout the year: Which opportunity deserves another hour? How aggressively should I push this deal? Is the next dollar of effort still worth it?

Even small design choices in compensation don’t just affect end-of-year outcomes—they shape day-to-day selling behavior in real time.

Which is why incentive plans can’t be entirely understood by a look in the rear view mirror. And yet they're often evaluated much further downstream, once we can see attainment distributions or whether revenue landed where expected.

Fortunately, our latest podcast guest, Johannes Habel, brings a more research-based, academic perspective to sales performance questions.

An Associate Professor at the University of Houston, Johannes researches sales management and the behavioral psychology behind selling. Before academia, he worked in management consulting, where his interest in high-value, consultative selling eventually led him to pursue a PhD in sales management. His perspective combines practical exposure to complex selling with a career spent studying what really makes sellers perform.

In this episode of the show, Forma.ai CEO and founder Nabeil Alazzam and Johannes move past familiar questions about whether incentives “work” and into the more interesting territory:

What exactly are they causing people to do?

Their conversation covers what research tells us about top performers, how an incentives intensity can affect the quality of a sale, and why the same comp mechanic can land differently across a sales force. Below we've summed up some of the most interesting ideas at a glance.

Episode resources

• Connect with Johannes on LinkedIn

Optimize for the quality of seller effort, not just the quantity

The traditional case for variable compensation is intuitive: make performance economically meaningful and people work harder. And research supports this relationship.

In fact, Johannes points to evidence connecting incentives with increased effort and persistence, which—in many selling environments—translates into stronger performance.

But, as Johannes indicates, this only captures one dimension of what the plan is changing. An even more important question is what kind of effort you’re creating.

Johannes is currently co-authoring research demonstrating that as an incentives intensity increases, sellers don’t just work harder, but they often change how they sell too. In one study context, for example, highly incentivized reps moved faster but they also asked fewer discovery questions, leading to weaker customer understanding and, ultimately, higher product return rates.

Here's Johannes on this point:

The key implication is that traditional dashboards can be misleading. Revenue may rise, activity may increase, and the plan may look successful, all while customer outcomes quietly degrade.

Which is why plan evaluation needs to extend beyond attainment and payout to include downstream signals like:

• returns or cancellations

• discounting behavior

• deal quality

• customer retention

• product mix

• renewal potential

• sales-cycle movement

These signals don’t all belong in the compensation plan itself, but they are critical diagnostic indicators of whether the incentive is producing healthy (vs. harmful) behavior.

Johannes also notes that most companies naturally settle into moderate variable pay ranges (often 40–60%) rather than extreme commission structures. And his research helps explain why: higher incentive intensity can indeed improve effort, but it also introduces trade-offs in consultation quality, hiring, retention, and income stability.

So when pressure builds to “make the plan more aggressive,” the better executive question becomes:

What behavior do we expect another dollar of incentive power to create—and what might deteriorate as a result?

This type of framing is more useful than debating pay mix in isolation.

Design for how your sales motion actually creates value

Early on in the conversation, Johannes also challenges a persistent stereotype about sales talent.

We tend to picture the great salesperson as the most extroverted person in the room. They're charismatic, persuasive, energized by conversation...they command a room. Right?

Well, in reality, the research points elsewhere.

In one study, sellers with a medium level of extroversion outperformed those at either extreme:

Johannes insists sellers need enough social energy and confidence to persuade, but they also have to be willing to stop talking, ask questions, and learn.

This balance matters even more as the sales motion becomes complex.

The takeaway here isn't to start personality-testing your sellers and designing plans around introversion. It's more that the selling motion itself should be a design input for the type of sellers you likely need.

Johannes describes curiosity, adaptiveness, persistence, and effort as important predictors of sales success. Adaptiveness in particular is one of the more established predictors found in academic sales research.

And different motions place different demands on these characteristics.

In a complex consultative sale, the rep has to create room for discovery, listen carefully, understand the customer's situation, and adapt. A more transactional motion, however, may place somewhat more weight on persuasion and pace, although customer understanding still matters.

This has a direct connection to compensation design.

If your go-to-market strategy depends on deep discovery, multi-threading, long-term customer value, and matching the right solution to the right problem, a plan that relentlessly emphasizes velocity can create friction with the very selling behavior the business needs.

The same applies in reverse. A high-volume transactional role shouldn't necessarily inherit the economics of a long-cycle enterprise seller.

Ultimately, per the research in this area, start with the economics and psychology of the job. Then decide what incentive architecture supports it. While it sounds straightforward. In practice, it is one of the easiest principles to lose when organizations standardize plans for administrative simplicity.

Stop expecting one incentive curve to motivate everyone equally

Most sales compensation teams already segment plans by role. But Johannes pushes this idea further: even within a selling population, research suggests that different performers can react very differently to the same incentive mechanic.

Take caps, for example. For a high performer who expects to reach the cap, the mechanic can entirely remove the economic reason to keep pushing. A lower performer who is nowhere near it isn't making the same calculation.

Same plan feature. Very different behavioral effect.

This leads Johannes to a broader conclusion from academic literature: there is no universal incentive structure that optimally motivates every seller.

And it creates a fascinating opening for where compensation design could go next.

Today, most enterprises can only take personalization so far before governance, explainability, administration, or fairness concerns create a new set of problems.

But, as Nabeil brings to light, imagine the infrastructure catches up.

In the full episode, Nabeil and Johannes explore a future in which incentives could respond more dynamically to the characteristics of a deal, the needs of the business, the seller involved, or changing market conditions. Instead of setting the economic signal once a year and hoping it remains relevant, the incentive itself could become much more adaptive.

Johannes finds the possibility compelling. He also gives an important warning: get the basics right first.

Many companies still struggle to operationalize relatively conventional incentive programs cleanly. So before experimenting with AI-adjusted incentives at an individual or deal level, organizations need reliable underlying data, clear strategy, strong governance, and plan mechanics sellers actually understand.

Johannes also points out there is also a new class of gaming risk introduced as things become more complex with dynamic incentives.

If sellers learn, for example, that avoiding a certain product today could cause the system to offer a richer incentive tomorrow, a dynamic model could inadvertently teach them to manipulate timing rather than respond to the intended signal.

For now, though, you can apply the principle without creating individualized compensation plans.

That is, when reviewing a major mechanic, pressure-test it across four dimensions:

1. ‍Who: How will top, middle, and lower performers experience this differently?
2. ‍Behavior: What action does the mechanic make economically attractive?
3. ‍Context: Does that behavior still make sense across different markets, products, customer types, or deal cycles?
4. ‍Side effects: How could a rational seller exploit the mechanic, and would you still like the business outcome if they did?

The future may be dynamic incentives. But the immediate opportunity is simply to stop evaluating every plan feature through the eyes of an “average” rep who may not actually exist.

Before changing the plan, prove you have an incentive problem

Perhaps the most useful reminder in the conversation is also the simplest. Johannes is unequivocal that incentives work (people respond to them, pursue the outcomes attached to them, and increase effort).

But because compensation is such a powerful lever, it’s often the first response when performance slips:

Pipeline weak? Change the incentive.
Product not moving? Add a kicker.
Strategic priority lagging? Adjust the measure.

Unfortunately, however, compensation is only one tool in a broader salesforce effectiveness system. When performance drops, the root cause may sit elsewhere (leadership, culture, pipeline quality, pricing, competition, or enablement, for example). An incentive change won’t fix a problem it didn’t create.

For comp leaders, your value isn’t in how quickly you can redesign a plan every time a gap appears. Sometimes the most strategic move is to not change the plan at all.

Instead, diagnose the system:

• Is poor attainment a motivation issue or an opportunity issue?

• Is a product underperforming because of rep economics or customer demand?

• Are sellers misallocating time because of incentives—or because of gaps in coaching, data, or frontline management?

That last question is especially important. Nabeil and Johannes note that compensation is rarely reinforced in ongoing manager–rep conversations. Even months after rollout, sellers may not fully understand how their plan works or how to use it to guide behavior. Yet these conversations often don’t happen, partly because money is sensitive and managers avoid it.

This is an operating model gap, and one of the highest-leverage things a modern sales compensation function can get a handle on.

Treat compensation as an ongoing behavioral hypothesis

Taken together, Johannes' research and observations throughout the episode suggest a useful way to think about incentive design.

Every compensation plan is really a set of hypotheses:

• If we make this outcome economically attractive, sellers will change their behavior.

• If sellers change their behavior, performance will improve.

• If performance improves, the business and customer will ultimately benefit.

The first link is relatively easy to measure. The real strategic work is in the next two: whether behavior actually changes as intended, and whether that change produces the right kind of performance and customer outcomes.

Did sellers respond as expected? Did different groups respond differently? Was the effort high quality? Did customer outcomes actually improve, or did the incentive simply shift behavior in unintended ways?

These questions will only become more important as AI increases visibility into seller activity and enables more adaptive approaches to performance management and incentives.

Johannes and Nabeil explore this future in the episode, including how AI may reshape productivity and sales-force structure. Johannes is skeptical of the idea that AI will only augment salespeople. As agentic systems take on more of the workflow, he expects companies will eventually need fewer people to produce the same output, even if early adopters temporarily expand to capture more opportunity.

That future raises many open questions—but one principle applies today:

The more precisely you influence seller behavior, the more precisely you need to understand the behavior you're trying to create.

Catch the full episode of The Sales Compensation Show with Johannes Habel and Forma.ai Founder and CEO Nabeil Alazzam for more on seller psychology, incentive design, dynamic compensation, and the future of selling.

Want more insights like this? Subscribe to The Sales Compensation Show on Spotify or Apple Podcasts, or YouTube for bi-weekly episodes featuring the revenue leaders behind today’s fastest-growing companies.