A signed contract was once a stable anchor. Your deal had closed, credit was assigned, and revenue followed a predictable path.

But for some time now, consumption-based pricing has changed this straightfoward route.  

A customer can sign today and take months to ramp, while another may spike early and then stall. And so, the largest long-term opportunity may not look like the strongest short-term performer.

Much more than a trendy pricing change, this pricing structure fundamentally alters what your revenue teams need to plan around.

At the Sales Planning Summit, we brought together three leaders who have navigated this transition:

• Marnie Sprenger, Senior Director of Sales Compensation at Nutanix

• Dillon Anderson, Senior Director of Sales Compensation and Planning Strategy at Wolters Kluwer

• Sidd Jain, Vice President of Revenue Operations at HashiCorp (an IBM company)

Grounded in their real experiences, the discussion converged on a central question:

How exactly do you run a sales organization when revenue is no longer tied to the moment a deal closes, but to how customers actually adopt, expand, and sustain usage over time?  

What's more, can you trust that signal enough to base quotas, incentives, and performance decisions on it?

Distilling the biggest takeaways, here are four implications you need to consider before embedding consumption deeper into your planning model.

1. Build the consumption forecast before the plan

Most annual planning processes start by establishing targets and working backward into territories, capacity, and compensation. But consumption pricing demands more patience.

In fact, Dillon posits that in a consumption environment, forecasting may matter more than planning, largely because without a credible view of how customers will actually use the product, the downstream plan rests on assumptions the organization can't yet defend:

Dillon's point is that consumption reflects more unpredictable behavior than a traditional commercial commitment.

The behavior varies by customer maturity, implementation requirements, geography, industry, product, and the amount of organizational change required to begin using what was purchased. As Dillon explained, a sale into the public sector may occur when funding becomes available, but meaningful adoption follows six or eight months later. The planning signal is therefore found in the adoption cycle, rather than the date on which your contract was signed.

This changes the order of operations. So before placing a consumption number into a quota model, you need to build enough evidence to estimate:

• The typical delay between signing and initial use
• How long different customer cohorts take to ramp
• Which product, segment, industry, and geographic factors affect adoption
• How usage develops relative to the original commitment
• Which early signals indicate acceleration, stagnation, or eventual contraction

Product telemetry is the strongest source of this when it's available. Sidd also pointed to customer-success signals captured in CRM as a useful proxy when direct telemetry is limited, although he cautioned that compensation tied to consumption ultimately needs reliable usage data behind it.

Overall, you don't need a perfect prediction, but you will need a forecast mature enough to support the decisions that follow. Otherwise increasingly sophisticated quota and incentive design may just emphasize uncertainty.

2. Protect seller fairness while the performance signal matures

Building on the prior point, the forecasting challenge Dillon alludes to quickly becomes a people and sales performance problem.

Because when consumption determines attainment, the organization stops measuring only what a seller sold. It begins measuring how quickly the customer implemented, adopted, expanded, and sustained its use of the product.

These outcomes eventually provide a more complete picture of customer value, but early in the transition to this model they can scramble the organization’s understanding of sales performance.

A high-performing seller can theoretically close the company’s largest and most strategically valuable deal, yet appear behind plan because the customer needs a year to reach meaningful usage. Meanwhile, another account may ramp immediately, create a sharp consumption spike, and then fall away because adoption never happened.

This distortion is pretty significant.

Seller-performance data influences promotion, coaching, territory allocation, capacity investment, quota calibration, and decisions about whether a particular segment or motion is working. When the underlying signal is still immature, these decisions can compound the original measurement error.

As Dillon advocates for, a safer transition will treat the new model as an incubation period:

• Pilot the measure with a defined segment, product, or seller population.
• Preserve a bookings or commitment component while consumption history develops.
• Compare initial consumption against longer-term adoption before redefining “good” performance.
• Inspect whether attainment differences reflect seller influence, account mix, or timing.
• Establish correction mechanisms before the first dispute or unexpected payout occurs.

Sidd was direct about the likelihood of early errors. As he shared, it's normal for organizations to not get consumption quotas exactly right in the initial stages. In this case, hybrid plans can reduce the consequences by spreading performance across multiple components while the company gains evidence and makes planned corrections.

3. Align everyone around customer success, then pay each role for its leverage

With consumption-based pricing, you're expanding the revenue lifecycle. The initial deal still matters, but so does implementation, activation, adoption, and expansion.

As more functions contribute to revenue post-deal, traditional ownership boundaries soften.

Sidd argued this blurring can be productive: account executives, customer success, renewals, and other customer-facing roles begin operating more like a pod aligned around account-level success:

Importantly, the operating model can converge around a shared customer outcome without giving every role the same compensation measure.

That is: 

• a direct seller focused on acquisition may have considerable influence over the initial commitment (but limited control over usage twelve months later)
• An account manager may influence expansion and adoption
• Customer success may affect activation, value realization, and retention
• All while technical or product specialists remove barriers that determine whether usage can grow at all.

Dillon’s decision rule was simple: avoid continuing to incentivize someone on an outcome they cannot meaningfully control. He suggested that direct sellers may remain focused on acquiring and handing over the account, with account management and customer-success plans reflecting their different responsibilities later in the lifecycle. Pay mix may need to change along with those expectations.

For RevOps and compensation leaders, the design work should therefore begin with an influence map:

1. Define the customer outcome. What must happen for the account to produce durable consumption?

2. Identify each role’s leverage. Which decisions or actions can a given role directly affect?

3. Clarify handoffs and overlaps. Where does responsibility transfer, and where is shared execution intentional?

4. Select the measure and pay mix. Match the incentive to the role’s actual proximity to the outcome.

5. Look for gaps and conflicts. Does one role benefit from selling capacity another team cannot successfully activate?

Consumption plans become difficult when compensation is asked to solve unclear ownership. Our panel advocates for making the operating model coherent first, where the measures can then reinforce it.

4. Use only the metric your operating system can defend

Something our panel was not remiss to mention was that the market is still searching for a standard consumption-planning model.

While there's a temptation to find the company considered most advanced, copy its approach, and call it best practice, this simply isn't the case for consumption pricing.

As Dillon insisted, there are not yet enough mature consumption models to establish one universal definition of “best in class.” The relevant standard is what the individual organization can reliably support.

The right model depends on the maturity of the product, the reliability of the data, customer adoption patterns, role structure, contract design, forecasting capability, and the company’s willingness to absorb variability.

Which is why Marnie’s readiness questions offer such a useful starting point:

Before paying anyone on consumption (or any unfamiliar measure) leaders need to know where the information originates, whether it is accurate, whether internal audit accepts it, whether enough history exists to set quotas, and whether sellers can see their progress in the system they use.

These questions should be treated as gates. But according to this criteria, a consumption metric is really only ready for compensation when the organization can:

• Produce it consistently from an agreed source

• Reconstruct and audit changes over time

• Explain the calculation to a seller

• Establish a credible target or reference point

• Distinguish seller influence from external variability

• Monitor whether the measure drives the intended behaviour

• Govern exceptions without improvising new rules for every account

But again, the systems don't need to be absolutely perfect on day one. Our panelists all recommended starting small, sometimes with an incremental incentive or a limited population. Sidd noted that early testing could even begin in a spreadsheet while product, finance, and sales operations instrument the model more formally.

What matters is whether the organization can defend the outcome.

Consumption creates more questions about why numbers moved, when usage changed, and which version of the data a seller previously saw. Dillon emphasized that auditability becomes especially important here because teams must be able to return to an earlier point and explain the difference.

Governance is therefore a big part of plan design in this case.

Clear terms reduce gaming. Agreed definitions reduce disputes. Documented correction mechanisms allow leaders to adapt without undermining trust. And visible, human explanation gives sellers confidence that the model was designed intentionally rather than generated by a report they can't interrogate.

Planning around consumption pricing is ultimately a learning system

As our panel covered throughout the session, consumption pricing introduces variability into nearly every part of the revenue model.

Forecasts become behavioural. Seller performance becomes harder to interpret. Revenue ownership stretches across more roles. And quotas and incentives end up depending on signals that may be developing.

The response can't be to force consumption into the old planning architecture and hope the numbers settle.

Instead, the strongest organizations build a learning system around it.

They will begin with the customer adoption curve, test measures before raising their stakes, preserve room for correction, and distinguish shared accountability from identical compensation.

The central planning question has evolved from how much a seller can close to how customer behaviour turns a sale into realized value, and which teams can influence that progression.

Your job is to understand the customer journey well enough to place someone’s quota and earnings against it.

Want to hear the complete conversation? Watch session three of The Sales Planning Summit  featuring all three of our expert panalists.