Why your biggest pricing mistake is copying competitors and going cheaper | Mark Walker @ Nue

Mark Walker has taken an unusual path into SaaS leadership. Before becoming CEO of Nue, he studied biology, went to law school, worked in litigation, and spent time in the music industry helping navigate the early streaming era. Today, he leads a revenue orchestration company that powers pricing, quoting, self-service, and billing workflows for many of the fast-growing AI businesses reshaping software.

That background gives him a perspective that is both practical and unusually blunt: most startups underprice, many companies overcomplicate the wrong things, and far too many discover their revenue systems are leaking money only after the damage is already done.

This is a closer look at how he thinks about pricing, usage models, revenue infrastructure, acquisitions, and the one sales principle that keeps showing up across every stage of company building.

Mark’s path to SaaS wasn’t linear

Like many experienced operators, Mark did not set out to become a software CEO. His first startup launched in 1999 and was, in his words, early rather than perfectly timed. The company, Blanketware, tackled a kind of web integration problem that resembled what Ajax later made mainstream. The idea was strong, the market was real, and the customer list included major enterprises, but the technology ecosystem had not caught up yet.

The business became solid without ever becoming explosive. It was one of those classic startup stories where the thesis was right, but the timing was a few years off. That experience shaped a recurring lesson: being early can look a lot like being wrong, even when the market eventually proves the concept out.

What Nue actually does

Nue positions itself as a next-generation RevOps or revenue orchestration platform. In practical terms, it covers the order-to-cash stack for recurring revenue businesses, including:

• CPQ
• Self-service purchasing
• Usage-based billing
• Enterprise billing workflows
• Payments and collections

What makes the company distinctive is not just feature breadth, but architecture. Rather than forcing companies to stitch together separate quoting, billing, and pricing systems, Nue is designed as a unified stack. That matters most when pricing gets complicated, which increasingly happens as software companies move beyond flat subscriptions into hybrid models, contract flexibility, and usage-heavy offerings.

It is especially visible in AI. Many high-growth AI companies now run at least part of their enterprise revenue stack on Nue. Some use it for CPQ only. Others layer in self-service, billing, or broader revenue architecture. The common theme is complexity: these businesses are experimenting with pricing in ways older systems were never designed to support.

Why AI companies are forcing revenue systems to evolve

One of the most interesting points Mark makes is that there is no single “AI pricing model.” The market is still in an experimental phase, and companies are testing different approaches at the same time.

That is important because legacy revenue systems were built around more standardized assumptions. They often expect pricing to be predictable, contracts to fit familiar shapes, and billing logic to remain relatively stable over time. AI companies break those assumptions quickly.

Instead of dictating a model to customers, Nue learns from the experiments happening across those businesses. Some innovations come from large enterprise-scale companies, but many come from smaller, faster-moving teams. Mark is clear on this point: innovation usually does not start with the largest players. It starts with the fastest ones.

That learning then flows in both directions. Smaller companies may pioneer a PLG-to-sales-led motion or a new pricing mix, and larger businesses can adapt those patterns later as they add self-service on top of an existing enterprise base.

How Nue decides which customer requests matter most

Prioritization gets tricky when customers vary not just by size, but by strategic importance and willingness to experiment. Mark describes three categories of strategic customers:

• Large customers by spend: Some customers pay millions annually and naturally receive a different level of support.
• Strategically important companies: These may matter because of their influence on broader market direction.
• Highly innovative customers: Some businesses are not huge yet, but they are solving new problems in new ways and pushing the product into future use cases.

That third group is especially important. A smaller company can shape roadmap decisions if it is exploring a pricing or packaging challenge that points toward where the market is going.

Salesforce-native, but not limited by Salesforce

Nue’s CPQ runs natively on Salesforce, but the broader platform sits on AWS. That distinction matters because not every transaction needs to flow through Salesforce. For product-led or hybrid businesses, the system can route only the right accounts or thresholds into the direct sales workflow.

For example, a company might send deals into Salesforce when:

• Seat count exceeds a certain threshold
• Usage or capacity reaches a defined level
• A strategically important customer account appears, regardless of deal size

This flexibility is part of why Salesforce still matters even in fast-moving AI companies. Once businesses become complex, they need a robust data model and a system of record that can support enterprise selling.

There is also an interesting market dynamic here. Salesforce sunset its own CPQ product, which created room for vendors like Nue. At the same time, Salesforce remains both a key partner and a common competitor. That sounds contradictory, but in enterprise software it is not unusual: a platform can be essential infrastructure and a competitive force at the same time.

When to invest in revenue infrastructure

This is where Mark’s advice gets especially useful for founders.

He separates the answer into two main paths.

If you start as PLG

If the business begins with a product-led or self-service motion, revenue infrastructure matters immediately. In that model, the buying flow is part of the product experience.

For early-stage companies, his advice is straightforward: start with Stripe Billing. It is usually the right fit when:

• You are not yet on Salesforce
• Your product catalog is still simple
• You need speed and API-first flexibility
• You are still proving the business before layering in enterprise complexity

The point is not that Stripe is permanent. It is that early companies should not burden themselves with heavyweight infrastructure before they need it. Later, when sales-led growth enters the picture, the business can integrate or expand into more powerful systems.

If you start with direct sales

For companies selling a smaller number of high-value contracts from day one, the answer depends on growth expectations.

If the market is relatively tight and growth looks like a classic SaaS curve, there is time to grow into more sophisticated systems. But if the company believes it could scale at extreme speed, delaying infrastructure becomes risky.

Mark’s view is simple: if you think you may become one of those companies that reaches enormous revenue in a very short time, adopt the systems built for that future earlier than feels comfortable. Replatforming in the middle of hypergrowth is painful. Replatforming while also handling VAT, GDPR, taxation, self-service, enterprise sales, and usage complexity can become a genuine operational threat.

On the other hand, if the company is only doing a handful of transactions a year, almost any setup will work for a while. The infrastructure decision becomes more urgent when operational complexity starts compounding, not merely when revenue appears on the board.

How to price your product without leaving money on the table

One of Mark’s sharpest observations is that many founders price too low in the early years. They discount heavily to win initial customers, reduce perceived risk, and create early proof. That is normal. The mistake is staying there too long.

His rule of thumb is memorable: ask what the highest-paying customer is currently paying, then test a price that is roughly double that amount.

The underlying logic is not arbitrary. In the first year or two, most startups do not have the confidence to charge market rates. They optimize for adoption instead. Once product-market fit starts to appear, pricing should catch up.

But his bigger argument is about how pricing should evolve in the current market.

Why seat-based pricing often misses the point

In older software markets, companies frequently inherited pricing logic from adjacent tools. If CPQ products were commonly sold per seat, new CPQ products also got sold per seat. The model spread because it was familiar, not necessarily because it reflected customer value.

Mark challenges that assumption directly. If the customer’s goal is revenue growth, process speed, or commercial flexibility, why should the vendor charge based on seat count alone?

That leads to a more flexible pricing philosophy: tie price to value, and offer customers different ways to buy depending on how they think about that value.

Nue does this by offering multiple pricing structures for the same product line. Customers can buy based on:

• Seats
• Percentage of revenue
• Percentage of revenue growth
• Dollars processed
• Usage lines processed
• Transaction count
• Transaction value

This creates a useful advantage. Different companies naturally gravitate toward different models, and that choice reveals how they see value.

Consider two businesses, each doing $100 million in revenue:

• One sends a tiny number of very large invoices.
• The other sends massive volumes of low-value transactions.

Those companies will not want to buy the same way. The high-volume business may prefer a small percentage-of-revenue model over per-transaction fees. The large-invoice business may prefer paying per invoice, while investing more in analytics or tooling that helps identify expansion opportunities upstream.

The lesson is clear: pricing should not be a borrowed template. It should reflect the customer’s economics.

Usage-based pricing: hype, reality, or something in between?

Usage-based pricing gets enormous attention, especially in AI, but Mark argues that truly pure usage businesses are still rare.

Most companies end up with blended models because customers need some combination of:

• Platform access
• Predictability
• Baseline commitment
• Service layers around the core product

That is why many AI companies now offer a mix of:

• Pure usage pricing
• Committed spend agreements
• Prepaid commitments
• Subscriptions
• Professional services
• Application-style software sold more traditionally

The hard part is not describing these contracts. It is operationalizing them. Once a customer commits to a spend level, a new set of questions appears:

• What if they do not use the full amount?
• Which products count against the commitment?
• What can be credited back?
• Who was responsible for the original forecast?

That is where commercial complexity starts bleeding into finance, customer success, and contract management.

For earlier-stage SaaS companies, Mark advises restraint. Do not make pricing more complex than the business can support. A combination like subscription plus included usage credits can often deliver the best of both worlds. Full-scale committed spend structures may be too much for a company with just a few million in ARR.

The biggest pricing mistake: copying competitors and going cheaper

Asked to name the worst pricing move a founder can make, Mark does not hesitate: looking at competitors, deciding to undercut them, and slapping a low price on the product without a serious value thesis.

There are narrow categories where cheaper really is the strategic edge, particularly in highly infrastructural markets where cost efficiency dominates. But most SaaS products do not win that way.

For most startups, the real advantage has to be that the product is better. If it is not better, the company already has a distribution disadvantage versus incumbents, and competing on price alone only makes the economics worse.

Cheaper may help in a negotiation. It should not be the core identity of the business.

What revenue leakage is—and why it can cost millions

Revenue leakage is one of those problems that sounds small until someone quantifies it.

Mark defines it as the gap between what a company could or should have billed and what it actually billed. In some cases, the issue extends all the way through to collection.

There are two major forms of leakage.

1. You had the right contract, but failed to bill correctly

A common example is seat expansion. A contract may allow customers to add seats without renegotiating every time, with the expectation that the vendor will measure usage and bill the difference. If the measurement logic is not there, that incremental revenue quietly disappears.

Recovering it later is still better than never recovering it, but delayed billing distorts metrics along the way. It can affect:

• NRR
• Growth rate
• Perceived churn
• Company valuation

2. Your systems disagree about what should have been billed

This is even more common. The CPQ or contract system may describe the commercial terms one way, while the billing system interprets them differently. The source of truth for what was sold lives in one place; the source of truth for what was invoiced lives somewhere else.

When those systems drift apart, leakage becomes almost inevitable, especially as usage-based pricing adds more variables.

Mark says Nue has found leakage in nearly every implementation. Sometimes customers already suspected it. Sometimes they had no idea.

In one case, the missing amount reached $2 million in unbilled ARR. The issue was not bad contract data or bad calculations. It was a mismatch between the old contracting logic and the old billing system’s understanding of it.

That kind of miss is not just a billing issue. It is a valuation issue. Missing ARR can be worth many times its face value once growth multiples are applied. Worse, if the business looks slower-growing because of the miss, the valuation impact can spread across the whole company.

Why unified systems reduce leakage

Mark’s argument for unified revenue architecture follows naturally from the leakage problem. When a company implements separate CPQ and billing tools, it creates translation risk. Someone has to map one system’s logic into the other. If the integration is incomplete or brittle, money falls through the cracks.

In Nue’s model, companies implement CPQ and then activate billing on top of the same commercial logic. That removes the daylight between what was sold and what gets invoiced.

The broader point is not just about tooling. It is about preserving pricing as a strategic lever. If finance cannot bill a new packaging model, sales cannot sell it confidently. If sales sells it anyway and the system cannot support it, the company invites leakage and confusion.

Pricing flexibility is only useful if operations can keep up.

Lessons from acquisitions: pay attention to people, not just terms

Mark has been through multiple acquisitions, and his strongest advice is strikingly human: watch how the acquiring team behaves during the deal process, because that is often the best they will ever treat you.

If the relationship already feels wrong while everyone is on their best behavior, it is unlikely to improve after closing.

He contrasts that with a positive experience selling a company into Netrix, backed by TA Associates. The fit worked because the people were solid, expectations were discussed openly, and the post-deal path allowed for a sensible transition rather than forced integration.

That leads to a second lesson: integrating businesses is hard, especially when products, customer bases, and cultures differ. A rushed cultural merge can do more damage than good. Sometimes the best path is to let the acquired business operate independently for a while, identify where talent and processes can cross-pollinate, and integrate gradually.

Good acquisition outcomes are rarely just about economics. They depend on trust, culture, and honest alignment about how the future should work.

Biggest failure: relying too heavily on one partner

When reflecting on his biggest failure, Mark points back to his first startup. The company became too dependent on one large partner. That relationship opened doors and accelerated customer access, but it also created fragility.

When leadership changed at the partner, the strategic logic behind the relationship disappeared. The startup was left with a product optimized around a distribution path it no longer controlled.

The lesson is not to avoid partnerships. It is to avoid building a business whose survival depends on a single one. Distribution matters enormously, but healthy distribution strategy should involve an ecosystem rather than one lifeline.

Biggest win: the people and the shared hard moments

On the positive side, Mark does not point first to money. He points to the people.

Startup work is intense, difficult, and often unreasonable in the best and worst ways. The reward is the kind of camaraderie that comes from solving hard problems with committed teams. In his view, some of the moments worth remembering most are not the easy wins, but the painful situations everyone had to rally around and fix together.

That may sound romantic, but for experienced founders it tends to ring true. The hard moments create the strongest stories, the strongest teams, and often the deepest sense of meaning.

The best sales hack: address what customers are afraid of

The most practical takeaway from the conversation may also be the simplest.

Customers usually are not trying to discover reasons to buy. They are trying to uncover reasons not to buy.

That changes how good selling works.

Instead of delivering an enthusiastic monologue about features and strengths, start by surfacing the concerns already sitting in the customer’s mind. Name the risks. Acknowledge the weak spots. Clarify what the product does not do well. Help them decide whether those gaps matter for their situation.

That approach builds trust quickly because it matches how people actually make buying decisions. They want differentiation, judgment, and honesty, not generic persuasion.

Mark gives the example of recommending adjacent or partner products when they are the better fit. Nue’s team is trained to help customers determine whether Nue’s native usage capabilities are enough or whether a partner product is the right next step. That candor can feel unusual in software sales, which is precisely why it works.

The broader principle extends well beyond sales: if you can identify what the other person is truly worried about and address that directly, you will usually have a far more productive conversation.