Professional services business model: strategic imperatives for outcomes in an AI world

There's a line Martin Roxby (Director and Co-founder of J21A) likes to use: a business without a strategy is like a ship without a rudder. It doesn't matter how hard your crew works, you'll end up wherever external forces take you. Most PS leaders would nod along to that. Then Martin asked the uncomfortable follow-up question: if you'd never run a business without a strategy, why are so many companies running professional services, a business within a business, without one?

And the external forces aren't gentle right now. A year ago, Simon England (Partner at Garwood Growth) laid out three paths for services firms in the AI era: break down, break even or break out. A year on, he's seeing plenty of firms trying to break out, but not spending enough or moving fast enough to escape the "gravitational pull" dragging them back to break even. Working hard, no rudder, and a current pulling the other way. That's where a lot of PS teams are sitting right now.

The "Strategic imperatives for outcomes in an AI world" panel was the most strategic session of PropelX 2026 in London. Heather Pertel (Head of Partnerships and Alliances at Rocketlane) moderated a conversation with Simon, his Garwood Growth colleague Adam Maze (Managing Consultant), Martin, and his J21A co-founder Steve Beckley (Director and Co-founder). Garwood advises PS leadership teams and private equity investors, while J21A helps PS teams in tech companies build capability.

Why is AI different from every previous technology wave for PS?

Simon has lived through the previous big technology waves, from the PC to the internet to digital. Every one of them generated new growth and a need for more people, and demand for services kept going up. AI is different. For the first time, a major technology shift isn't creating net demand for people, and that's exactly why it's so disruptive for a business model built on selling people's time.

He sees the change as more sophisticated, bigger and faster than anything before it, with a lot more honesty now about which services are being disrupted, which are being augmented and which are becoming fully AI native. Simon also sits on the board of HCLTech, where he's watching a 250,000-person organisation transform in real time.

What changes first in the professional services business model?

Adam's answer was the economics of delivery. They're changing across the whole lifecycle, from scoping and onboarding to mobilisation and support. Using AI to cut the cost to serve is the obvious move, but the more forward-thinking PS firms are asking a different question: how do we deliver more for customers in the same amount of time, or faster? And customers aren't naive. They know AI should be making the work quicker.

Leading firms are redesigning delivery around their IP and around human judgement in hybrid human and agentic workforces, rather than bolting AI onto old processes. That has big implications for talent, resourcing, governance and the fundamental assumptions behind the cost base of the business.

Automate, augment or keep human

Adam's framework for redesigning delivery is simple to say and hard to do. Look at where AI can automate work entirely, where it can augment what your team already does, and where you need human judgement in the loop. Rebuild your delivery processes around those three categories. Then, once it's rolled out and adopted, judge it by how it moves your customers' outcomes, not by your own internal adoption metrics, because those "just tell you who's adopted the processes and the tools." That's the difference between AI ROI and AI activity.

Can your PS business break out, or will it break down?

Simon's three paths from a year ago still hold. Some firms will break down as AI eats their core offering. Some will break even, doing enough to survive but not enough to grow. And some will break out by differentiating, but the question he now asks is whether they're spending enough, and moving fast enough, to escape the pull back towards break even.

A lot of businesses are trying, he said, "but not quite hard enough." For PS leaders, that's a useful gut check: is your AI budget sized for a real change in how you deliver, or for a few experiments that keep the board happy?

Why does embedded PS need its own strategy?

Martin argued that services is now the real differentiator between software competitors, and that makes a PS strategy more important than ever. "People don't want to pay for output and effort," he said. If your services business is built on output and effort, you're selling something the market no longer wants. And if you'd never run a business without a strategy, it makes no sense to run a business within a business without one.

A real strategy isn't a vague vision statement people hope they understand. It translates what the market is demanding into the capabilities PS needs to build, and how it will build them. Martin's test: if your CxOs can't say what PS is worth to the business beyond billable utilization, there's no strategy. (He also urged the room not to be "the redheaded stepchild" of their organisation, which landed well given there were three redheads on stage.) It's the core of crafting a winning PS strategy.

Authority and budget, not just a mandate

If PS has a mandate to deliver outcomes, Martin said, it needs two things to make it real: delegated authority and discretionary budget. When he asked the room how many PS leaders had lots of investment coming into their teams, one person raised a hand. Businesses have traditionally not invested in professional services, and that needs to change if PS is expected to lead the shift to outcomes, starting with PS budget planning.

What new management disciplines do PS leaders need?

Heather asked what new disciplines firms need now that people and AI work so closely together. Simon named three, and they're exactly what private equity investors examine when they bring Garwood in to assess a PS business:

• Deep tech understanding at executive and board level, not just in the delivery team, so leaders can make informed AI investment decisions.
• Risk, governance and data management, areas that have often been delegated or, in Simon's words, "in some cases abdicated," and now need proper governance.
• Rewiring how you sell, moving from time and materials or fixed price to outcome-based, platform-based solutions, which is a fundamentally different sale that needs a software mindset.

He called the third one the hardest, and said it comes up in his conversations with executives every single day.

How is AI reshaping the spectrum of practice?

Steve brought in a model from David Maister's classic book, Managing the Professional Service Firm (not as good as J21A's own book, he joked). It plots every PS offer on a spectrum. At one end is procedural, commodity, high-volume, low-cost work with lots of competition. At the other is high-complexity, high-cost "brains" work with very few competitors, where advisory services used to sit: "a brain on a stick," paid for time, with a report as the output.

Over the last 10 to 15 years, SaaS PS built "a bit of a monoculture" at the commodity end. Part of the reason was investors: PE and VC firms didn't want services revenue weighing on the multiples they got from selling SaaS. AI will hit that commodity end first and hardest, and it's a real threat to the people's side of PS. But Steve also sees a big opportunity at the other end, where PS stretches past advisory into outcomes. He pointed to ideas from earlier in the day, like Sri's "go-live as day zero," as signs of that shift. His summary: "what we need now are outcomes engineers," which is a very different PS maturity goal from delivering more hours.

Why aren't companies getting verifiable ROI from AI?

Simon pointed to McKinsey's latest assessment that fewer than 6% of companies investing in AI are getting verifiable results. Part of the problem, ironically, is that most firms measure AI adoption by its inputs rather than its outcomes. The business change needed to make AI work is often underplayed, and data is frequently the real blocker.

He also sees a lot of AI washing in consulting and professional services, with fragmented pilots that never run end to end and aren't connected to a clear, business-specific AI narrative. It sounds obvious, he admitted, but it's often the most obvious things that get overlooked. Closing that gap between strategy and execution is where AI transformation actually starts to pay off.

If your AI dashboard only shows usage, you've built a very expensive attendance register.

How should you decide which services to automate, productise or exit?

Simon shared a quadrant Garwood uses with leadership teams and investors ("I'm a consultant, so I have to have a quadrant model"). It plots each service by how much value the customer perceives against how repeatable it is versus how much human judgement it needs. Each bubble is sized by the revenue it brings in, which makes priorities obvious:

• Low value, highly repeatable: automate it, hard. Some clients in software development and IT service management are already demanding 50 to 70% price reductions here, so this work needs to be fully AI-shifted.
• High value, repeatable: productise it by building IP and platforms and moving towards outcome-based pricing.
• High value, high judgement: enhance it. This is your moat, built on people, communication, expertise and the creative work AI can't do, and it's where value delivery differentiates you.
• Low value, people-driven: rethink, reposition or exit. Staff augmentation usually lands here, and it's where firms need to decide whether to stay in the box at all.

5 key takeaways on the professional services business model

1. Effort no longer sells. Customers want outcomes, so a model built on output and effort is selling the wrong thing.
2. AI hits commodity work first. Low-value, repeatable services face the biggest price pressure, including 50 to 70% cuts.
3. Embedded PS needs a strategy. If leaders can't say what PS is worth beyond utilization, there isn't one.
4. Measure outcomes, not adoption. Usage metrics don't explain why fewer than 6% of AI investments show verifiable results.
5. Move up the spectrum. The opportunity is in judgement, IP and outcomes, not more hours.

Best practices for rethinking your professional services business model

• Map your services. Plot your top 10 services on the Garwood quadrant, sized by revenue, using your PS KPIs.
• Redesign delivery in three buckets. Decide what to automate, what to augment and where human judgement stays in the loop.
• Write a real PS strategy. Translate market shifts into the specific capabilities your team needs to build.
• Ask for authority and budget. A mandate without delegated authority and discretionary budget stays on paper.
• Build consultative skills. Pair new tools with the consulting and commercial skills outcome conversations need.

About Rocketlane

Rocketlane is an agentic AI-powered professional services automation (PSA) platform built for services teams running complex implementations.

Its AI layer, Nitro, deploys named agents that do the work, not just flag it:

• Migration Agent — data preparation and transformation
• Workforce Agent — repeatable configuration at scale
• Documentation Agent — project context current across calls and emails
• Project Governance — delivery standards enforced proactively
• Nitro Signals — account risk surfaced before it escalates

The core distinction: Nitro agents produce the deliverable or enforce the gate. Most platforms advise. Rocketlane acts.

Teams using Rocketlane ship faster, recover margin through tighter governance, and scale delivery without proportional headcount growth.

Related sessions from PropelX

• Outcome based selling: how we rebuilt delivery around outcomes: how Pendo moved from selling hours to selling milestones, and why mindset mattered most.
• Outcomes vs outputs: translating customer outcomes into measurable success: David Dixon's framework for proving the outcomes PS can actually control.
• AI native organization: building the AI-native PS function: PS leaders on intent, org design and the end of staff augmentation.

The market has stopped paying for effort. See how Rocketlane helps PS teams run a business model built on outcomes.