In brief: AI does not repair a broken workflow; it automates its failures faster. This briefing helps leaders connect measurable value, one transformable process and a paced route from pilot to sustainable operation.
When people talk about AI in business, they point at features and tools. The bigger impact is in processes — how you discover problems, make decisions, ship work, and measure results. Processes are the invisible operating system of the organisation, and AI applied to a broken one doesn’t fix it. It automates the dysfunction faster.
The Process pillar exists to prevent that. It breaks transformation into three questions, each with its own briefing thread:
- Total Value Transformation — does this AI initiative tie to a measurable business outcome?
- Transformable Process — which workflow do we transform first, and how?
- Transformation Journey — how does the organisation get from pilot to scale without breaking?
Each connects back to the People pillar: Targets shape direction, Transformers carry execution, and ignoring either stalls everything.
Total Value Transformation
The first question for any proposed AI initiative: what value does this create? Not “does it look innovative” — many pilots exist for exactly that reason and die as lab experiments. Value means a lever you can measure: revenue growth through retention or personalisation, cost savings through automation, risk reduction, or user satisfaction you can see in the numbers.
Frame it for the audience in front of you. For Users, translate to relieved pain. For Buyers and Deciders, translate to ROI and strategic fit. Without that translation, Buyers dismiss the initiative as a cost centre, Deciders see a risky experiment, and Transformers end up building something with no defined purpose.
Daily Harvest is a clean example of value-first deployment: AI applied to recommendations, support, and shipping logistics — including calculating dry-ice needs per shipment — with impact measurable in efficiency and satisfaction rather than in press coverage.
Transformable Process
Knowing why AI matters, the next question is where to start. Look for processes that are painful twice over: for users (slow onboarding, clunky search) and for the team running them (manual reporting, bug triage, repetitive QA). The best first candidates sit at the intersection — relieving internal drudgery in a way users can feel.
Two honest complications. First, automation and augmentation are different answers: removing repetitive work versus helping people decide faster, and each fits different processes. Second, transformation shifts burden as often as it removes it — automate bug triage and your Maintainers inherit a new monitoring job. Count that cost before declaring the win.
DHL shows the pattern at scale: AI translating delivery instructions, training new staff, and routing over a million calls a month — rolled out with staff involved, framed as a colleague rather than a replacement, which is much of why adoption held. Camunda shows it at the tooling level, with generative AI folded into process modelling itself.
Transformation Journey
Scaling is where every tension surfaces at once. Users worry about learning curves and their own jobs. Buyers watch costs widen with scope. Deciders demand governance before granting the green light. Influencers rally adoption or sink it. Meanwhile Implementers fight legacy integrations, Organizers try to synchronise pilots into one rollout, Maintainers discover that monitoring at scale needs several times the pilot’s resources, and Creators ask for time to retrain models before the next department goes live.
A concrete version: your support pilot cut ticket resolution 30%. Leadership wants it company-wide. Scale fast and pain spreads faster than value; scale slow and you lose enthusiasm and funding. There is no formula here — pacing is the leadership decision, made checkpoint by checkpoint, with phased funding and tested infrastructure rather than a switch-flip. Most organisations are now making exactly this move from proving pilots to embedding AI in operations, which means most of your competitors are mis-pacing it in one direction or the other.
The leader as process architect
Across all three questions, the leader’s job is the same translation described in the People pillar, applied to workflows: convincing Buyers and Deciders with value cases while ensuring real user pain drives the choices, navigating the pull between visible quick wins and defensible ones, and pacing the journey so trust builds as fast as capability.
Without that architect, pilots stay pilots, value cases get lost between finance and engineering, and teams burn out automating the wrong things. AI transforms processes only when someone decides which ones, in what order, at what pace — and that someone is you.
Explore the Process pillar
Start with mapping the work before automation and making process pain visible. Then use the Use-Case Canvas delivery conversation to carry an agreed route into delivery.