What Leaders Wish They Knew Before Going In
It’s a sobering reality: across industries, roughly 70% of digital transformation programs fail to achieve their intended outcomes. For organizations embarking on a Sales Performance Management (SPM) implementation, that statistic is particularly damning. Sales compensation is especially vulnerable because it lives at the crossroads of multiple departments — Sales, HR, Finance, and sometimes IT. It’s where data streams, human behavior, and executive priorities inevitably collide.
As one comp leader put it, “You don’t just automate spreadsheets… you automate politics.”
For every successful rollout, there are two more that looked promising on paper but quietly collapsed under process debt, misaligned expectations, and sheer exhaustion. Talk to leaders who’ve survived the journey, and the same themes emerge – hard-won lessons from the other side of SPM implementation. Here are the top three things they wish they’d known from the start.
1. Get the Process Right Before the Platform
SPM implementations rarely fail because software can’t calculate. They fail because organizations haven’t clearly defined what they want calculated, or more importantly, why. The biggest hidden cost isn’t buried in the fine print of your vendor contract; it’s in internal ambiguity – the volume of comp plans, their complexity, unclear metrics or ownership, etc.
What we see too often is teams rushing into technology, hoping it will solve structural problems. But if you don’t think critically about your internal processes and effectively “clean house”, what you’ll automate is the chaos you have today. The calculations will just run faster.
Across dozens of implementations we’ve analyzed, seasoned leaders agree: success as an outcome of clarity, not configuration. One director said bluntly, “If your comp philosophy isn’t nailed down, your configuration will just codify confusion.”
That truth was echoed by John Waldron, now Total Rewards Director at PepsiCo, whose early SPM rollout taught him the perils of leading with technology. “There was this naive thought that if we implement this high-tech option, it is going to automate the whole process,” he told Workspan Daily. When PepsiCo reframed their second iteration around vision, ownership, and user experience (not system features) automation improved, accuracy followed, and trust soon after.
The takeaway is as simple as it is nuanced: you need to align on governance, intent, and the human experience you want to enable before shopping around for a technology solution. A well-designed spreadsheet with purpose will outperform a multi-million-dollar platform built on confusion, every day.
2. Fix the Data Plumbing Before You Build
It seems obvious that clean data is table stakes, especially in a calculation heavy space like SPM. Despite this, poor pluming remains a source of much of the implementation frustration for many strong organizations. Upstream data from CRMs, ERPs, and HRIS systems often arrives late, improperly labeled, or inconsistently defined.
What seasoned teams have learned is to run data-readiness sprints well before building: mapping ownership, refresh cycles, and reconciliation logic. They also establish a “go-live + 60 days” monitoring cadence, treating early mismatches as an expected phase, and not a crisis. As one senior comp admin put it, “We stopped pretending the data pipes were perfect and started measuring leak rate instead.”
3. Plan for Human Change, Not Just System Change
In nearly every implementation postmortem, we hear the same refrain: “We underestimated change management.” What they really mean is people didn’t understand — and therefore stopped trusting — the system.
PepsiCo’s first rollout offers a case in point: Their calculation accuracy hit 99%, but payout accuracy lagged at 90% (a gap that translates into millions in potential errors, annually.) The math was sound, but the ecosystem surrounding it was not. Processes lacked clear ownership, users were unsure of how to navigate the system, and transparency was patchy. A robust calculation engine alone could not (and cannot) build confidence or competence.
And that’s the common trap: assuming that once the system “works,” adoption will follow. But the reality is that enablement and trust decay together. When users don’t fully understand a system, mistakes creep in. This, of course, erodes confidence, which in turn reduces engagement and leads to the reintroduction of practices like shadow accounting.
Leaders who know better stage rollouts in waves, pair training with feedback cycles, and make sure both data and people are ready before automation takes over. Communication isn’t top down; it’s continuous, responsive, and shared.
The takeaway isn’t simply that change management matters (of course it does, enormously). It’s that operational and human change must mature together… or neither will hold for long.
Closing Thoughts: Design for Imperfection
Every large-scale tech transformation will hit turbulence – that’s a given. The real test of maturity isn’t whether you can prevent it, but whether your systems, processes, and people can adapt when it happens.
Lasting programs are built with resilience by design. They balance governance with flexibility, clarity with pragmatism, and structure with empathy for the people who will live inside them. The organizations that get this right treat SPM not as a one-time project but as a living framework that keeps evolving as the business does.
That kind of resilience doesn’t come from software. It comes from the discipline to align early on purpose and process, and to design for imperfection rather than pretend it won’t happen.
As John Waldron put it, “All three of these items — strategy, vision, and experience — should be unique to your company. You can’t outsource ownership.”
Diana Romanovsky
Diana Romanovsky is a technical content creator specializing in sales performance management (SPM). She covers industry insights, emerging trends, and best practices to help organizations optimize their sales compensation strategies.
