Why Marketing Operations Transformations Fail at the People Layer

Every enterprise marketing operations transformation has two workstreams, whether or not the programme plan names them both. The first is technology delivery: requirements, configuration, integration, testing, go-live. The second is organisational change: the shift in how people work, what they are responsible for, and how they are measured. Most programme plans are detailed on the first and thin on the second.

This is not a resource problem. It is a framing problem. When change management is treated as a subset of the technology workstream — a communications plan, a training schedule, a go-live email — it is scoped too late, resourced too lightly, and closed too early. The technology workstream ends at go-live. The change workstream begins there.

Prosci's ADKAR model defines five sequential conditions for individual change: Awareness of the need to change, Desire to support it, Knowledge of how to change, Ability to demonstrate the change, and Reinforcement to sustain it. Most marketing operations programmes invest in Awareness, Knowledge, and Ability. Desire and Reinforcement are the stages most frequently under-resourced. Prosci's published research on change management best practices identifies under-resourcing of Reinforcement as a primary driver of post-go-live adoption failure.

Kotter's 8-Step Change Model, first published in Harvard Business Review in May 1995 and updated in Accelerate (2014), identifies 'generating short-term wins' (Step 6) and 'sustaining acceleration' (Step 7) as the steps most frequently cut when programme timelines compress — both fall in the post-go-live window. The pattern is consistent across frameworks: the work that determines whether change holds is the work that gets cut first. In Dean Brown's advisory work at Rarovera, this is the most consistent finding across enterprise marketing operations engagements: the technology lands; the change does not.

Every team has informal leaders — people whose adoption of a new platform signals to their peers that it is safe to commit. They are rarely the most senior people in the room. They are the ones others ask when they are not sure what to do, the ones whose scepticism becomes the team's scepticism and whose enthusiasm becomes permission.

A stakeholder influence map identifies these people before go-live. It plots every affected role against two axes: their current level of support or resistance, and their informal influence on the people around them. High-influence, high-resistance individuals are the programme's most important change management problem. High-influence, high-support individuals are its most underused asset.

The influence map is most effective when built in the design phase, not the deployment phase. The informal leaders identified are brought into the programme early — invited to review workflow designs, consulted on configuration decisions, briefed ahead of the wider team. By the time the platform goes live, these individuals have been part of the programme, not recipients of it. Their adoption is genuine, and their peers notice. This reflects Kotter's Step 4 — enlisting a volunteer army — applied at the team level.

The alternative — identifying resistance after go-live and responding with more communication — is slower, more expensive, and less effective. Resistance that has had time to organise is harder to shift than scepticism engaged before it hardens.

A training plan teaches the platform. Workflow redesign answers a different question: on Monday morning, what does this person do first, and how? These are not the same question, and conflating them is where most adoption programmes lose the thread.

Role-level workflow redesign means documenting, for each affected role, the step-by-step sequence of tasks that person performs differently on the new platform. Not at the system level — 'the campaign request now goes through the DAM' — but at the individual level: when a brief arrives, you open the intake form in the platform, complete the required fields, attach the approved brief document, and submit. The campaign manager receives an automated notification and has a defined window to accept or return with comments. That level of specificity.

Without it, people default to familiar patterns. The new platform becomes a parallel system rather than a replacement. The old approval email continues because nobody documented that it should stop. The spreadsheet persists because the new dashboard does not yet feel as legible as the one someone built over three years.

The redesign should be done with the affected roles, not for them. A structured working session with a campaign manager and a marketing operations lead, walking through a real campaign brief on the new platform and documenting what changes, produces more durable adoption than any amount of e-learning. This maps directly to Prosci's ADKAR Knowledge and Ability stages: knowledge of how to change, and demonstrated ability to do so in a real workflow context.

Go-live is not the end of the programme. It is the beginning of the hardest part. The 90 days after go-live are when adoption either compounds or collapses — and the difference is almost always whether there is a named person checking the signal and acting on it.

A minimum viable adoption cadence has three components:

• Weekly metric review: active users as a percentage of licensed seats, feature utilisation by role, and support ticket volume categorised by type. A spike in 'how do I' tickets signals a training gap. A spike in 'this is broken' tickets signals a configuration or integration issue. Flat active-user numbers signal a workflow or resistance problem. The metric tells you which problem you have.
• Fortnightly team-lead check-in: a 30-minute conversation between the programme owner and each named team lead, covering what is working, what is not, and what one thing would make the biggest difference this fortnight. The output is a written action, not a note to follow up.
• Monthly executive review: attended by the programme sponsor, covering adoption trend, open blockers, and any decisions that require executive authority to resolve. The sponsor's visible engagement with the adoption data is itself a reinforcement signal — it communicates that the programme is not over and that adoption is being measured.

Prosci's ADKAR model identifies Reinforcement as the final and most frequently under-resourced stage. Kotter's Step 7 — sustaining acceleration — is the step most often cut when programme teams move to the next initiative before the current change has embedded. This cadence is the structural mechanism for delivering both.

This piece addresses people-layer interventions in marketing operations transformation programmes, grounded in Prosci's ADKAR model and Kotter's 8-Step framework. It does not cover technology selection, integration architecture, data governance frameworks, programme budgeting, or transformations driven primarily by regulatory compliance requirements — such as those triggered by the EU AI Act (Regulation EU 2024/1689) or data protection obligations.

The interventions described are most applicable to programmes implementing or significantly reconfiguring marketing technology platforms. They are less directly applicable to pure process redesign programmes that do not involve a platform change, or to transformations where the primary constraint is technical debt rather than adoption.

Commercial interest disclosure: Rarovera is a marketing operations and technology consulting firm. This piece reflects our advisory perspective and we have a commercial interest in this topic. Readers should evaluate this framework alongside independent research from Prosci, Kotter International, Gartner, and Forrester.