Scaling Revenue with Predictable Forecasting ft. Jonas Samsioe

From Consulting to Sales to RevOps: A Non-Linear Path

Jonas didn't start his career planning to become a revenue operations leader—partly because the role didn't exist yet. Twenty years ago, the titles "sales operations" and "revenue operations" weren't part of the lexicon. What existed were people solving problems at the intersection of sales, systems, and strategy.

His path began with consulting at two of the Big Five firms, where he learned to get up to speed quickly on unfamiliar businesses and help them solve problems. But something was missing. He wanted to build something of his own, so during the early 2000s dotcom era, he co-founded a B2C marketplace. It was ahead of its time, and ultimately didn't succeed—but the learning was invaluable.

From there, Jonas moved into sales, carrying his own book and learning firsthand what it's like to live on commission. That experience—not knowing if you'll make rent beyond your draw—gave him empathy for salespeople that would shape his entire approach to operations.

After earning his MBA in Europe (he's originally from Sweden), Jonas moved into sales management for an educational company, then returned to the U.S. and landed at American Express. That's where everything started to come together.

American Express: The Foundation of Field Effectiveness

At Amex, Jonas worked in what they called "field effectiveness"—essentially the early version of sales and revenue operations. His team supported roughly 150 salespeople whose job was to travel to different markets and ensure that American Express cards were accepted for payment.

The work was fundamentally operational. When the corporate card team landed a new client in Austin, Texas, and discovered there weren't enough restaurants accepting Amex, Jonas's team would route that to the territory owner—or if the territory was too large, assemble a strike team to go sign up merchants on the ground.

The key constraint: they only targeted businesses that already accepted credit cards. Convincing someone to accept credit cards at all is a much harder sell than convincing them to add Amex to their existing mix.

This required building a sophisticated territory model using data from Dun & Bradstreet—work that honed Jonas's Excel skills and taught him how to use data to drive field productivity. But the real insight came when he realized he could flip the model on its head: if you know all the companies in the U.S., and you know which ones already accept Amex, the gap is your lead generation universe.

That model became a powerful lead engine for American Express—and it caught the attention of Dun & Bradstreet, who recruited Jonas to help them pitch similar solutions to other customers.

Consulting for Equity: Building Revenue Engines for Startups

After a stint at D&B, Jonas started his own consultancy in New York, trading services for equity and helping early-stage companies set up and scale their sales organizations. It was a formative period—working with startups that had great ideas but no infrastructure to convert those ideas into revenue.

Jonas would attend startup events, talk to founders about their problems, and swoop in to help them hire their first salespeople, build their first sales processes, and set up backend systems that could scale. The goal wasn't always an IPO—often it was positioning the company as an acquisition target or simply getting to sustainable revenue.

This experience taught him how to build lean, effective systems without the resources of a Fortune 500 company. Every dollar mattered. Every process had to deliver immediate value. There was no room for theoretical best practices that didn't move the needle.

When his daughter was born, Jonas took a full-time role at DCR Workforce in Florida (health insurance matters when you have kids). The company had landed Google as a customer, and Jonas's job was to reverse-engineer how a 250-person company had pulled that off—then replicate it to land more enterprise clients. The company eventually sold to a public company, opening the door to Jonas's next chapter.

Scaling from 100 to 1,300: Building a Go-to-Market Engine

Jonas joined a SaaS company in Miami when it was still relatively small. Over the following years, the go-to-market organization grew from roughly 100 people to 1,300-1,400, including all instrumentation: sales, renewals, specialists, overlays, and operations.

The company went from $180 million to $1.6 billion in ARR. It was a hypergrowth environment fueled by acquisitions, private equity investment, and a prolific CEO who set an aggressive path forward. There were growing pains—lots of them—but the scale of the challenge is what makes Jonas's insights so valuable.

When you're managing a 1,300-person revenue engine doing 60,000+ deals a year, everything breaks. Quote-to-cash processes that worked at 100 people collapse under the load. Salesforce instances get bloated with acquisitions. Forecasting becomes a battlefield of conflicting data. The tech stack sprawls beyond anyone's ability to manage it.

Jonas learned how to build systems that could operate at that scale, and more importantly, how to know when to rebuild them.

Bottoms-Up Forecasting: How a Sales Factory Predicts Revenue

The company Jonas worked with ran what he calls a "sales factory"—a high-velocity, lower ASP model where individual deals don't make or break the forecast. This is fundamentally different from enterprise sales, where a single deal can swing the quarter.

In a sales factory, the last thing you want to do is forecast whales. Big deals break the forecast in both directions. They create false optimism when they're in the pipeline, and devastating misses when they slip.

The core of the process was a weekly ritual called the "territory report card." Every rep met with their manager to walk through their deals. The meetings were recorded—even when in-person—both to ensure managers actually did the work and to enable AI analysis of how reps talked about their pipeline.

The manager's job wasn't to accept the rep's forecast. It was to interrogate it. "You say this deal closes next week, but you don't know who the financial approver is. If it's a $1,000 deal, fine. If it's a $15,000 deal, that's a problem." The manager had to understand the structure of every deal in their forecast, then commit to it.

That forecast rolled up to directors, who reviewed it with VPs, who had full instrumentation to see every deal, every rep, and every AI-generated insight. The system used Power BI dashboards layered with AI scoring: the rep's stated probability, the AI's calculated probability based on activity and sentiment, and flags when the two diverged.

For example: a rep says 90% likely to close, but the last conversation was two weeks ago and the sentiment analysis was negative. The AI surfaces that conflict, and leadership can make an informed decision about whether to trust the rep's optimism or discount it.

This bottoms-up discipline, combined with AI augmentation, created forecast accuracy that improved throughout the quarter. Since 70% of deals were found and closed within the same quarter, early-quarter forecasts had more variability—but by week 8 or 9, the numbers were tight.

The Hybrid Forecast: AI as Supplement, Not Replacement

Jonas is clear about the role of AI in forecasting: it's a supplement to human judgment, not a replacement. In what he jokingly calls "vibe forecasting"—where you just feel your way to a number—AI adds rigor. But it can't run the show.

The reason is context. If a rep hasn't talked to a customer in two weeks and it's a $1,000 deal, that might not matter. If it's a $15,000 deal in the final week of the quarter, it probably does. AI can flag the gap, but a human has to interpret the significance.

This hybrid model worked exceptionally well in the U.S., where the company had strong conversational intelligence coverage. In EMEA and APAC, GDPR restrictions limited call recording, so the AI had to rely more heavily on Salesforce data—which may or may not be as accurate. That regional difference reinforced the need for human oversight.

Jonas isn't a fan of "vibe selling" or "vibe forecasting," but he's a strong believer in using AI to enhance the information humans are working with. The best forecasts come from combining data discipline, AI insight, and experienced judgment.

Aligning RevOps with What Sales Actually Cares About

One of the perpetual tensions in revenue organizations is the gap between what RevOps thinks matters and what sales actually cares about. RevOps can get hyperfocused on systems, dashboards, and process. Sales just wants to hit quota and close deals.

This was especially important because the company hired aggressively from non-traditional backgrounds. Many reps were in their first or second job out of college, or coming from retail environments into their first office role. They had no Salesforce discipline, no experience with structured sales processes, and needed very different training than someone with five or ten years of SaaS experience.

What mattered more than documentation was fixing the systems that slowed sales down. The company had acquired a business that could turn around a quote in 20 seconds—the kind of frictionless experience you get signing up for Netflix. After the acquisition, that same process took 24 hours. That's not progress. That's a disaster.

Sales and revenue operations had to identify that problem, diagnose the root cause (in this case, a clunky quote-to-cash integration), and fix it. Sometimes RevOps could solve it directly. Sometimes they had to escalate it to executives with a plan for how to dig out.

The point: RevOps earns credibility with sales by making their jobs easier, not by creating more reports.

Compensation as Behavior Change: The Evergreen Problem

Nothing changes sales behavior like compensation. Jonas saw this firsthand when analyzing why so many reps were selling evergreen (non-term) agreements instead of one-year or three-year contracts.

The reason was simple: evergreen was easier. No negotiation over contract length, no pressure to commit, just sign and go. But it was leaving money on the table—for both the company and the reps.

The fix wasn't training. It was compensation design. RevOps restructured accelerators so reps made more money on term agreements. At the same time, customers saved money by signing three-year deals versus paying month-to-month.

Once reps saw that a three-year deal put more money in their pocket and gave them a value proposition to sell, behavior shifted. They started proactively structuring deals with terms instead of defaulting to evergreen.

The lesson: when sales behavior isn't aligned with business objectives, look at the incentives first. You can train people all day on why term deals are better, but if evergreen pays the same and takes less effort, that's what they'll sell.

From CPQ to Revenue Cloud: Why the Shift Matters

For years, the company relied on CPQ (Configure, Price, Quote) as the center of its revenue tech stack. It worked fine when deals were straightforward, but as the business scaled and deals got more complex, the cracks started showing.

The biggest issue: most revenue leakage happened outside CPQ. A quote would generate cleanly, but then negotiations, approvals, and discounting would happen in side channels—email, Slack, verbal agreements—and CPQ couldn't track or control any of it.

Sales reps started gaming the approvals process. At 60,000+ deals a year, it was hard to catch every instance, especially at end of quarter when everyone was rushing. Deals that shouldn't have been approved slipped through. Too many evergreen agreements got signed because they were faster. Discounts that violated policy were granted because someone wanted to hit their number.

CPQ didn't have the guardrails to prevent this. It could generate a quote, but it couldn't enforce pricing policy, automate approvals based on deal structure, or provide real-time insights into where the process was breaking down.

That's why the company migrated to a full revenue cloud platform. The new system had better policy automation, tighter pricing guardrails, and visibility into the entire quote-to-cash cycle—not just the quote generation step.

The transition wasn't seamless. They were still discovering issues from the old system that were baked into processes and habits. But the shift from point solution (CPQ) to end-to-end platform (revenue cloud) was essential for managing deal velocity and margin control at scale.

Dividing Responsibilities: Sales Ops vs. Revenue Ops vs. Deal Desk

As revenue organizations scale, new roles proliferate: sales ops, revenue ops, deal desk, enablement. The boundaries between them aren't always clear, and that can create confusion about who owns what.

Jonas's framework is grounded in company size and immediate need. A $50 million company doesn't need this level of specialization. But at $250 million and above—especially with high deal velocity—you need to divide and conquer.

In a hypergrowth environment, you bring in sales ops and deal desk first because they solve immediate pain. You bring in revenue ops later because they operate at a higher altitude and need systems to already be in place to optimize.

This doesn't mean revenue ops is more important—it means the needs are sequential. You can't strategize about pipeline health if your reps can't get quotes out the door.

Tech Stack Discipline: The Cost Beyond the Price Tag

One of the most eye-opening insights from Jonas's experience is how tech stack bloat happens—and why the sticker price is only part of the cost.

The company ran a relatively lean tech stack compared to peers, partly by choice and partly out of necessity. They built many tools in-house because they had strong R&D and Salesforce development teams, and because they were stuck on Salesforce Classic (which modern tools couldn't integrate with).

When they evaluated third-party tools, the calculus wasn't just features and integrations.
It was: what's the admin burden to keep this thing running?

Jonas gives the example of a widely-used conversational intelligence platform they inherited from an acquisition. It was feature-rich and worked well—but it was expensive, and they were only using a fraction of its capability. They built a lighter-weight solution in-house that covered 30-40% of the functionality but cost a fraction of the price and required less ongoing maintenance.

That trade-off won't make sense for every company. But for an organization with 30% annual turnover, complex integrations, and constant change, a tool that requires extensive training and configuration can become a liability. The people who implemented it aren't there anymore. The context is lost. The tool sits underutilized while the company pays full freight.

When to Pause and Audit Your Stack

In hypergrowth, there's never time to pause. You're always fighting the next fire, closing the next deal, fixing the next broken process. But at some point, you have to take a breath and audit what you've accumulated.

Jonas recommends asking three questions:

He also suggests that bringing in fresh RevOps leadership periodically can be valuable. After seven years in a role, part of your brain still operates at the $200 million company level even though you're now managing $1.6 billion in ARR and serving mature EMEA and APAC markets. New eyes see the legacy assumptions that insiders don't question.

The CISO Factor: Why Security Now Drives Tech Decisions

In the last year, Jonas has seen a dramatic shift: CISOs are now heavily involved in tech stack decisions, especially around AI and automation.

The reason is simple: every tool now has a "copilot" or "AI feature," and many of those features involve sending customer data to external models. If you're in a regulated industry—finance, healthcare, government—or subject to GDPR, that's a potential catastrophe.

Jonas half-jokes about sitting at his desk and receiving notifications that yet another vendor has exposed his data. It's not theoretical. It's happening constantly.

This means the old model—see a tool, get a demo, buy it, deploy it—no longer works. Now it's: see a tool, get a demo, get CISO approval, navigate security reviews, ensure data residency compliance, and then deploy. By the time you're done, the business need may have evolved.

It's frustrating, but necessary. Revenue operations leaders have to build relationships with security teams and factor compliance into every decision.

For people looking to break into revenue operations or scale their impact in the role, Jonas offers straightforward guidance:

For business owners

Follow immediate needs. Start with salespeople. Add a sales manager. Then a sales leader. When they're wasting time on reports, hire sales ops. When deals need better structure, bring in deal desk. Revenue ops comes last because they're strategic—they help you tomorrow, not today.

For aspiring RevOps professionals

Recognize that the role requires comfort with details. You'll work closely with IT, debug integrations, build data models, and need to understand systems at a technical level. You don't need a computer science degree, but you can't be intimidated by complexity.

At the same time, you need relationship skills. You're the air traffic controller of the sales engine. Finance needs forecasts. Product wants performance data. Marketing needs attribution insights. Sales needs tools that work. You're the bridge between all of them, and that requires trust, communication, and credibility.

For companies scaling revenue

Don't confuse sales ops with revenue ops. They serve different functions. Sales ops is tactical and immediate. Revenue ops is strategic and forward-looking. You need both eventually, but not at the same time. Hire for today's pain first, tomorrow's optimization second.

Jonas closes with a topic he wishes more companies discussed: the gap between ARR (annual recurring revenue) and actual recognized revenue.

Investors say they care about ARR, but what they really want is revenue. And when there's a wide gap between the two—driven by deal structure, payment terms, or revenue recognition rules—it creates problems.

Revenue operations and sales enablement have a role to play in closing that gap by educating reps on what "good deals" look like. Not just deals that close, but deals that convert to recognized revenue quickly and cleanly.

This requires structuring processes and compensation so reps are incentivized to build deals that are good for the business, not just easy to close. It's the difference between a sustainable revenue engine and one that's constantly playing catch-up.