Stop Pricing Growth Like It's Predictable: What Plinko Can Teach Agencies About Smarter Economics

I've spent over 20 years building, running, and advising agencies. I've priced projects by the hour. I've set flat retainers that made no one happy. I've argued about performance models over conference tables where everyone pretended they understood attribution. And after all of it, here is what I know for certain: most agencies price their work based on a lie.

The lie is that growth is predictable.

It's not. And that single flawed assumption is the root cause of almost every pricing problem agencies face.

Here's how it usually goes. An agency scopes a project. They estimate hours, set a monthly retainer, or pitch some version of "we'll take a percentage of results." The client agrees. Both sides shake hands. And then reality happens.

The first campaign underperforms. The second one is mediocre. The third one, for reasons no one fully understands, takes off. Revenue spikes. The client is thrilled. The agency is still earning the same flat retainer they locked in three months ago.

Or the reverse. The agency is doing excellent work. Strategy is sound. Execution is clean. But the market shifts, or a competitor launches something unexpected, or the client's product team drops the ball. Results stall. The client starts questioning the retainer. The agency starts padding reports with activity metrics to justify its existence.

Both scenarios stem from the same problem: the pricing model assumed a linear relationship between effort and outcome. Put X in, get Y out. But growth has never worked that way. Not once. Not ever.

If you've ever watched The Price Is Right, you know the game Plinko. You drop a chip from the top of a board. It hits pegs. It bounces left, right, then left again. It's chaotic. You have no control over where any individual chip lands.

But here's the thing. Drop a thousand chips, and something interesting happens. The distribution becomes remarkably predictable. Some chips land in the low-value slots. Most cluster around the middle. A few hit the jackpot. The individual path is random. The aggregate pattern is structured.

Growth works exactly the same way.

You launch a campaign, and it flops. You optimize it, and results improve. You catch the right timing with the right message, and it explodes. Any single initiative is unpredictable. But across a portfolio of well-executed efforts over time, the distribution of outcomes follows a reliable pattern. Most produce moderate results. A few produce outsized returns. Some produce nothing.

The question is not whether you can predict exactly where each chip lands. The question is whether your pricing model reflects how those chips actually behave.

For most agencies, the answer is no.

Agencies typically price in one of three ways. Each has its own flavor of dysfunction, but all three share the same underlying flaw.

Model One: Hourly billing.
This method punishes efficiency. The faster and better you get at your job, the less you earn. It incentivizes the opposite behavior you'd want from a partner. It turns every conversation into a negotiation over hours rather than outcomes. And it tells the client, in the clearest possible economic terms, that your value is measured in time, not results.

Model Two: Flat retainers.
They create a false sense of stability. They're predictable for both sides, which is exactly the problem. When results spike, the agency eats the upside. When the results tank, the client questions the expenditure. Nobody adjusts. The model just sits there, disconnected from reality, while both sides quietly resent each other.

Model Three: Pure performance models
These sound fantastic on pitch decks. "We only get paid when you grow." Except they ignore the dozens of factors outside the agency's control that affect outcomes. Things like: Market conditions. Product quality. Competitive moves. Seasonality. Internal politics. When you tie 100% of compensation to results you can only partially influence, you're not building a partnership. You're building a coin flip with extra steps. And the volatility in cash flow makes it nearly impossible to staff and plan responsibly.

All three models assume outcomes are predictable enough to price confidently upfront. They're not. Growth is volatile. Pricing usually isn't. That gap is where trust erodes, relationships sour, and good agencies get fired for reasons unrelated to the quality of their work.

So what's the alternative? You design pricing that reflects how growth actually behaves. You stop pretending you know exactly where the chip will land and start engineering economics that work across the full distribution of outcomes.

The framework has three layers. Each serves a distinct economic function. Together, they create a model that protects both sides on the downside, aligns incentives on the upside, and stays sustainable over time.

There is always a fixed retainer. Always. This fee is non-negotiable.

The base covers strategy, execution capacity, and operational continuity. It pays for the team that shows up every day, the tools that run the campaigns, and the strategic thinking that guides the work. It protects both sides. The agency can plan staffing and invest in senior talent. The client gets guaranteed attention and priority.

The base should cover your fully loaded cost of servicing the account, plus a modest sustainability margin. It should not include aspirational profit. That comes from Layer 2.

Think of the base as the cost of dropping the chips. You can't play Plinko without putting chips on the board. The base funds the game.

On top of the base, you define a clearly structured upside component tied to measurable outcomes. When results exceed agreed baselines, the agency participates in the incremental value created.

Such an arrangement can take several forms, depending on the business model: a percentage of revenue above baseline, tiered bonuses at specific milestones, margin participation, or bonuses tied to specific business events, such as a product launch or market entry.

The structure matters less than the principle: when the client wins are bigger than expected, the agency shares in that win. When results are at or below baseline, the client pays only the base. The agency absorbs the downside risk in the performance component.

The critical number in this entire model is the baseline. Get it wrong, and you either give away the upside you earned or create resentment on the other side. Baselines should be built on trailing actuals (not projections, not hopes), adjusted for reasonable organic growth, and reviewed on a regular cadence. If you can't agree on a baseline, you're not ready for performance pricing. Start with the base only and add the performance layer once you have enough shared data to set a number both sides trust.

This is the layer most agencies never think about. It's also the one that makes everything else sustainable.

Not every client will spike. Some will plateau. Some will underperform. One might break out significantly. That's the nature of the distribution. The discipline is in intentionally managing your client mix, the same way an investor manages a portfolio.

Your anchor clients, the ones with strong, stable base retainers and modest performance upside, provide cash flow stability. They fund your capacity to take on growth clients where the performance potential is higher, but the variance is wider. And a small slice of your book can be moonshot engagements: lean base, outsized upside if the distribution plays out.

The ratios matter. Too many anchors, and you're a boring, low-margin shop grinding out retainer work. Too many moonshots and one bad quarter kills you. The mix creates both resilience and optionality.

Most agencies manage their client roster by accident. Whoever signs, signs. Whatever the mix is, that's the mix. Portfolio discipline means being intentional about the types of engagements you pursue, the target ratio, and when to adjust.

The framework works because it stops pretending growth is something it's not.

It forces honesty. Both sides need to look at real data and agree on the baseline. No more hiding behind vague promises or inflated projections.

It aligns incentives. The most profitable thing the agency can do is maximize the client's results. Not minimize effort. Not pad hours. Not churn deliverables. Actually drive growth.

It creates asymmetric upside without reckless exposure. The base protects against downside. The performance layer captures upside. Neither side is overexposed.

It elevates the conversation. When the economics align, you stop arguing over whether the retainer is "worth it" and start talking about growth strategy, market opportunities, and how to push results higher. The relationship shifts from cost management to value creation.

And at the portfolio level, it makes the agency's business model fundamentally more resilient. You're not dependent on any single client outcome. You're managing a distribution across your entire book of business.

If there's a perfect application for Plinko thinking, it's dormant client reactivation.

Most agencies are sitting on databases of 100, 200, or 500 past clients and prospects they've lost touch with. Relationships that went quiet. Projects that ended. Contacts that drifted away. The database is like a Plinko board, ready for action.

You don't know which specific dormant contact will convert. But you know that across 200 outreach attempts, the distribution will follow a predictable pattern. Roughly 40-50% won't respond at all. Another 25-35% will be warm but not ready to re-engage right now. 10-15% will produce a small, quick re-engagement. And 3-8% will generate a significant new deal, often larger than the original engagement, because trust already exists.

You can't predict which contact falls in which slot. But you can predict the distribution. And the economics are beautifully clean because the baseline is literally zero. These clients are currently generating nothing. Every dollar of reactivated revenue is incremental. There's no argument about what would have happened anyway. They were inactive. Now they're not.

Price it with the same three layers. Base fee for building and executing the reactivation system. Performance component tied to revenue from reactivated accounts. Portfolio discipline across multiple reactivation campaigns so the aggregate distribution drives your business model, not any single engagement.

Let’s be clear: this is not gambling. Plinko works because the probability distribution is engineered. The board is designed. The pegs are placed intentionally. Agency pricing should be engineered with the same level of intentionality.

This is not ‘discounting’ your services. The base layer should fully cover your costs plus a reasonable margin. You're not giving away work for free in exchange for a performance lottery ticket. You're building a sustainable base and adding upside on top of it.

This is not a magic trick that makes bad agencies profitable. If the work isn't good, the performance layer won't activate. The framework rewards results. It does not create them. The quality of your strategy and execution is still the foundation of everything.

And this model is not a one-size-fits-all template. The specific numbers, structures, and baselines will differ for each agency and client. The framework provides the architecture. You engineer the details.

After 20 years of building and advising agencies, I've watched this industry cycle through the same pricing debates over and over. Hourly vs. retainer. Retainer vs. performance. Performance vs. value-based. Everyone's looking for the right model as if there's one answer that works universally.

There isn't. But there is a right principle: your pricing should reflect how growth actually behaves, not how you wish it would.

Growth is not linear. It has never been. Stop pricing it like it is.

When you combine downside protection, upside participation, and portfolio management, you stop acting like a vendor who gets paid for effort and start operating like a partner who gets paid for impact. The economic model alters the relationship, which in turn alters the conversation, leading to different results.

That's what partners do. They engineer the economics so both sides win across the full distribution of outcomes.

Drop the chips. Manage the board. Build something that works no matter where any single one lands.