Procurement change management: why training people at go-live guarantees failure

Every procurement transformation follows the same pattern. Configure the system. Train the users. Go live. Celebrate. Then, six weeks later, adoption metrics plateau. Users revert to old workflows. Spreadsheets reappear. The sweet-talked ROI begins to look like a rounding error. The problem is not the software. The problem is not the training content. The problem is the assumption that training someone once, at the moment of go-live, creates lasting behavior change. Adult learning science says otherwise.

Skills and behaviors are not the same thing. A skill can be taught in a two-day workshop. A behavior takes months of repetition, feedback, and reinforcement to form. Most procurement transformations fund the workshop. Almost none fund the reinforcement infrastructure required to turn a learned skill into a sustained behavior. That gap — between training attendance and behavioral adoption — is where the ROI of procurement technology goes to die.

---

The forgetting curve is eating your training budget

Learning research has known for decades that newly acquired knowledge decays rapidly without reinforcement. Within 7 days of a training event, retention drops below 30%. Within 30 days, it approaches 20% — unless the learner applies the knowledge repeatedly in realistic contexts. In procurement transformations, the training event happens at go-live. But go-live is the worst possible time to learn. Users face real transactions, real pressure, and real consequences for mistakes. Under stress, they fall back to their comfort zone — the old system.

The Aberdeen Group study, cited by Zycus, found that best-in-class organizations are 57% more likely to provide role-based training during e-procurement implementation. But even role-based training is not enough if it happens only once. The difference between best-in-class and average organizations is not the quality of the training content. It is the presence of a reinforcement system that extends beyond go-live.

---

Why procurement transformations fail at the adoption layer

Industry analysts consistently identify the same root cause. As Ivalua notes, "resistance to change and misalignment across departments can significantly impede the adoption process." But resistance is not the right frame. Most users do not resist change. They resist uncertainty. Training at go-live creates maximum uncertainty because users must simultaneously learn new processes and perform real work.

The Art of Procurement notes that "some digital procurement implementations fail not because the solution is bad, but because of poor adoption." A 2025 Oliver Wyman CPO trend report found that even with modern source-to-pay platforms, "several processes are still considered cumbersome by internal stakeholders." When users find a process cumbersome three months after training, it is not because the training was insufficient. It is because the training was a single event in a process that requires continuous support.

---

The four-phase reinforcement model

Sustained behavioral adoption requires a structured approach to learning that extends months beyond go-live. The most effective model maps to four phases:

Phase 1 is the most commonly skipped. The pressure to go live pushes organizations to train users in the production environment during the first weeks of operation — when every mistake has real cost. A sandbox environment where users can practice creating purchase orders, handling exceptions, and navigating the system without real-world consequences builds the confidence required for adoption.

---

In-the-flow learning: the mechanism that works

Digital adoption platforms have emerged as the most effective mechanism for post-go-live reinforcement. These platforms embed step-by-step guidance, checklists, and contextual tips directly into procurement workflows. A 2025 analysis of in-app learning describes how a manufacturing company "enforced procurement governance, improved cost efficiency, and strengthened supplier relationships" by embedding guidance into its SAP system.

In-app guidance works because it removes the need for users to remember training content. Instead of recalling the correct process from a workshop three weeks earlier, the user follows the guidance embedded in the system at the moment of use. Whatfix also recommends identifying "power users in each department who serve as internal champions" who "reinforce best practices and provide peer coaching." These champions create a distributed support network that does not depend on a central training team.

• In-app guidance: Step-by-step walkthroughs inside the procurement system. Reduces support tickets by 30-40% in documented implementations.
• Peer champion network: One power user per department who handles day-to-day questions and reinforces best practices. Extends training capacity without adding headcount.
• Analytics-driven intervention: Tracking adoption metrics and targeting coaching to users who fall below threshold. Treats adoption as a data problem, not a motivation problem.
• Manager-as-coach training: Equipping category leaders and team leads to observe behaviors and provide feedback. The single highest-leverage reinforcement mechanism.

---

What good looks like: post-go-live as the starting point

Organizations that achieve sustained adoption treat go-live not as the finish line but as the starting point. The ADKAR model (Awareness, Desire, Knowledge, Ability, Reinforcement), widely used in procurement change management, places Reinforcement as the final and most critical building block. Without it, Awareness and Desire create interest, Knowledge and Ability create competence, but nothing creates habit.

The Precoro guide on procurement change management emphasizes that "reward early adopters, mark key milestones, and continuously communicate benefits." An employee engagement analysis on procurement transformation puts it more directly: "If the same questions keep coming back weeks after go-live, it is not just a training issue. It may mean the process design or interface is not intuitive enough for real-world conditions." The best training program in the world cannot compensate for a system that users cannot navigate under time pressure. The two must be designed together.

---

What this means for buyers

If you are planning a procurement technology implementation in the next 12 months, your training budget is probably too small by a factor of 3-4x — because you are budgeting for the workshop, not for the reinforcement.

• Budget for reinforcement, not just training. A rule of thumb: allocate at least as much budget for post-go-live support (in-app guidance, coaching, refreshers, analytics) as for initial training. The workshop is 20% of the cost. The reinforcement is 80% of the outcome.
• Build the sandbox before you build the training deck. Users need a safe environment for deliberate practice before they face real transactions. Insist on a sandbox environment as a contractual requirement in your vendor selection process.
• Train managers before you train users. Managers cannot reinforce behaviors they do not understand themselves. Run a manager training session 2-3 weeks before end-user training, focused on observation, feedback, and coaching techniques.
• Define "go-value" before you define "go-live." The Zycus/Aberdeen research makes a critical distinction: go-live is the first moment of truth. Go-value — when adoption metrics hit target — is the real moment of truth. Set your go-value target before the implementation starts and track against it monthly.
• Treat recurring help desk questions as system design failures, not training gaps. If the same question comes up repeatedly, the system is not intuitive enough. Escalate the pattern to your vendor or internal IT team. Training cannot fix bad UX.

Frequently asked questions