The Forward Deployed Engineer, Chapter 15: The Economics of Forward Deployment

This is Part 15 of a series walking through my book The Forward Deployed Engineer. In the previous chapter, we placed the function in the org. This one turns to the question every founder eventually has to answer: do the unit economics work?

The economic question is simple to state and hard to answer: does the function’s gross margin, properly accounted for, compound or degrade as you scale customers? The chapter spends some time on “properly accounted for,” because most companies that conclude the math works have miscounted at least one of the three relevant cost centers. What counts as FDE cost? What counts as customer success cost? What counts as platform engineering cost that the FDE function should be credited for unlocking? Each company answers these differently, and most answer them inconsistently across quarters, which is why the FDE function’s economics tend to look better in the quarter the analysis is run than they actually are over the year.

There are three distinct ways to price the function, and each has its own gross margin profile. The function can be bundled into the platform subscription, which preserves software-business economics on the surface but moves the FDE cost into COGS in ways that can quietly invert the math at scale. It can be charged as a separate professional-services line, which makes the cost visible but tends to push the function toward consulting-shop behaviors and produces the wrong incentives for the sales team. Or it can be priced against outcomes, via the Outcome-Level Agreements we’ll get to in Chapter 17, which is the model best aligned with the function’s actual value but which requires real measurement discipline to make work. Each pricing model carries its own characteristic distortion when the sales team gets compensated against it, and the chapter walks each in operational detail.

Whatever pricing model is chosen, gross margin depends on the relationship between the cost of an engagement and the platform leverage produced by it. That relationship is governed by two specific levers: the platform commit rate (the share of field learnings that become roadmap items within a quarter) and productization velocity (how quickly gravel roads pave). When both metrics trend up, gross margin trends up. When either trends down, gross margin trends down, and it tends to do so faster than the leadership team realizes until a board meeting reveals the slope.

The Services Trap, Multi-Year Contracts, and What the Math Implies

The consulting trap I named in Chapter 13 has an economic signature, and it deserves its own treatment here. The signature is a gross margin profile that trends slowly but inexorably toward services-business levels rather than software-business levels — 40-50% instead of 70-80% — and it tends to be invisible for several quarters before it becomes alarming. The diagnosis is straightforward: customer-specific work has grown faster than general platform work, and the function is now paying senior FDE wages to deliver services rather than to produce leverage. The cure is unromantic and consists almost entirely of the operational levers I’ve already named: enforce platform commits, reject deployments whose patterns don’t graduate, and measure productization velocity as if it were a revenue metric, because economically it nearly is.

Two commercial structures shape the function’s economics in non-obvious ways. Multi-year contracts can stabilize the function’s revenue (good) while encoding pricing decisions for years (potentially very bad), and the chapter walks how to structure them so that the customer gets the price predictability they want without locking the vendor into 2025 economics in 2028. Volume discounts can quietly invert the unit economics if the function’s cost curve doesn’t match the discount curve — a discount that assumes the second deployment costs half as much as the first only works if the second deployment actually does, which is a question the productization metrics either answer or don’t.

The chapter closes with the small set of operational decisions the economic analysis implies. Enforce platform commits as a first-class metric, tracked at every quarterly review and visible to the function’s leadership. Measure productization velocity, and act on it when it falls. Price for outcomes when you can, because the OLA-priced function has the cleanest math and the deepest customer trust. And never let the function’s gross margin remain unmeasured for more than a quarter — the math does not work itself, and the function’s economic signature is the single best leading indicator of whether the operating model is healthy. Tomorrow: when not to build any of this at all.