By Ross Robinson, director of propositions and partnerships at Optima Partners
Over the past few years, organisations of every shape and size have poured money into AI, personalisation and smarter decisioning platforms. Many now have an impressive list of pilots on their books – from personalised content tests to decisioning experiments and GenAI-powered workflows.
On paper, it looks like progress. But in reality, very few organisations are seeing tangible commercial outcomes sparked by these pilots. What begins as ambitious transformation often deteriorates into fragmented activity, slow approvals and disconnected efforts that never reach the customer.
It’s AI theatre: lots of dramatic gesturing, very little measurable impact. And as pressure grows to prove ROI, demonstrating this impact, not just its promise, is becoming non-negotiable. Because no matter how compelling the performance, the P&L decides whether it was worth staging at all.
The barriers to scale
Most organisations aren’t short on talent, innovation or creativity. Marketing, digital and customer teams can already design highly engaging personalised experiences.
The challenge lies in the execution – in delivering these experiences consistently and at scale.
Some of the key barriers include:
- Fragmented ownership: When teams build their own solutions to support personalisation, it often leads to inconsistent processes and misaligned KPIs.
- Lengthy manual approval processes: Decisions that could be made in minutes using the right AI tools, take days due to unnecessary manual oversight.
- Misconceptions around investment. Scaling AI is often assumed to entail significant upfront costs around new technology, supporting infrastructure and upskilling staff – but this is rarely true.
Why unlocking real value is important
Too often, AI success is measured by technical capability. People focus on what it can do, versus what it actually delivers. Instead, we need to focus on measurable outcomes:
- Revenue growth
- Improved efficiency and productivity
- Reduced costs
- Faster time-to-market
- Stronger compliance
These outcomes are driven not by AI alone, but by how its insights feed into decisioning – and how those decisions translate into personalised actions at scale.
That’s why we often encourage clients to focus on Time-to-Value (TtV) as a measure of AI progress: how quickly a strategic idea generates measurable commercial or operational benefit. We track this through three specific areas:
- How quickly an idea becomes a live experience
- The number of assets delivered per team over time
- The percentage of initiatives that create measurable uplift – revenue, retention or cost efficiency.
This shift in focus moves teams away from chasing pilots and towards delivering tangible impact.
Closing the gap between strategy and delivery
To break out of the AI pilot trap, businesses need more than a robust strategy. They need the operational foundations to execute ideas at speed, with consistency and control.
That means a value-first approach, designing customer experiences around high-impact opportunities. But the real acceleration comes when governance, automation and operating-model reform are embedded in AI initiative execution from day one:
- Governance for speed: Pre-approved templates, clear guardrails and risk-aligned workflows allow teams to move quickly without waiting for manual approval.
- Automation: Tools like Guardian Agents can handle regulatory checks, tone-of-voice reviews and content compliance, cutting manual review time and improving decision accuracy.
- Workflow acceleration: Streamlining the content supply chain and added automated approval triggers means teams can move from AI ambition to business impact far faster.
The result is not just better personalisation, but better decision making, delivered at scale with measurable outcomes – whether that’s reducing time-to-market, improving decisioning accuracy or realising incremental revenue.
Moving beyond “AI theatre”
The organisations that will pave the way over the next decade are those able to operationalise AI sustainably, confidently and at pace – turning ideas into outcomes, not intentions. But to do this, we need to focus on the execution and organisational impact.
That’s when AI stops being a promise and starts being a P&L driver.
