What's Changing Underneath the Simulation Licence

Every Head of Digital at a mid-sized European process-engineering consultancy has had the same Monday. Aspen AI assistant adoption at sixty-eight percent across the mid-tier. Feasibility-study drafting time down thirty-two percent. Project margin flat against Q4. Your VP Engineering opens with the dashboard. He is mid-sentence when your phone buzzes. Email from the CTO at your largest specialty-chemicals client, a fourteen-year account: "We're inviting three firms to quote the next plant-expansion feasibility study, including a digital-engineering shop we met at ACHEMA. Your proposal is welcome by month end." Your most senior unit-operations engineer mentioned his retirement timeline at the Christmas party. Fourteen months. His reasoning on that client's reactor geometry has never been captured.

You have run digital and engineering technology for this firm for ten years. The pattern used to be simple. License the right simulators. Hold the hazop bar. Ship feasibility studies on time. Meet the board on licence utilisation and billable-hour margin. The business was understandable.

Something changed. Aspen and AVEVA both ship AI on top of their platforms now. What used to take fifty mid-level engineers on a feasibility-study pipeline runs on five seniors and an AI-copilot layer. Ninety-five percent of corporate AI deployments produce no measurable P&L impact, and the gap between licence adoption and project margin is where your seat now lives or loses. A digital-native process-advisory shop at forty percent of your fee just landed a first named quote in your specialty-chemicals book. Your VP Engineering carries correlation-set reasoning that has never been written down, and retirement is fourteen months out.

You are not running one R&D budget. You are running two, and only one is on your scorecard. One funds the Aspen AI assistant seats your engineers already use. The other funds what has to exist by 2028: the reasoning your three most senior unit-operations engineers take with them when they retire, the correlation library no simulator vendor ships, the client-project archive your firm owns the rights to. On one hurdle rate the first wins every quarter. On one scorecard the second does not exist.

This briefing covers what is happening in your sector, what it means for the R&D and digital-platform function you run, and the three questions your team should answer before the next board cycle.

The AI-natives in your category are not experimenting. They are quoting. Small digital-engineering shops built around GPT-5 enterprise, CAPE-Open fluency, and a handful of senior process engineers are pitching specialty-chemicals and refining clients on three-week fixed-fee feasibility studies at forty percent of your rate-card quote. They do not have your fifty-year operator references. They do not need to. Their deliverable lands hazop-defensible enough, fast enough, cheap enough that your client's procurement team is asking your account partner to explain the eight-week timeline in terms their CFO can defend.

The moat your firm ran on for two decades has moved. Mid-tier engineering hours commoditise fast now. A feasibility study that used to require a twelve-person team across three months can be drafted by a three-person team and an AI-copilot layer in three weeks, on modern simulator architecture, priced fixed-fee on an outcome the client can verify. Your engineering-hours edge was never going to hold. What can hold is the layer underneath the simulator: the unit-operations reasoning your seniors carry for specific reactor geometries, the correlation library tuned against twenty-five years of plant data, the client-project archive your firm owns the rights to. Those compound. They take decades to build and they cannot be bought. A digital-native with three people and a CAPE-Open licence cannot ship them because it does not have your project history.

Your feasibility study drafted on Aspen AI assistant is on an eighteen-month productivity curve that every competitor also has. Your captured correlation library against your top ten client reactors is the asset the digital-native cannot replicate without your project archive.

The industry's speed mismatch is the third force. Digital-native process-advisory shops build from scratch with no legacy, no heritage simulator investment, no senior-partner billing model to protect inside their own house. Your firm has all three. You cannot pivot at their fee point on their terms. What you can do is identify what is genuinely defensible in your engineering output, stop defending what is not, and ship the moat they cannot build. The firms already moving are quiet about it. The ones rolling out Aspen AI assistant seats and calling it an AI strategy are the ones the board will replace in 2028.

Your build engine. Aspen AI assistant adoption is at sixty-eight percent. Feasibility-study drafting time is down thirty-two percent. Project margin is flat. Your mid-tier engineers are pushing AI-drafted PFDs through review with minimal annotation. The correlation reasoning behind the numbers never gets written down. Twelve months in, your team cannot defend a hazop against a senior challenge because the documentation trail stops at the AI output. The review discipline that worked in 2022 is training the wrong practice on AI-generated volume now.

Your platform and data. Aspen is pushing an AI-assistant uplift on your licence renewal. AVEVA AI Assist is in your mid-tier's inbox. Siemens Gproms is quoting an AI add-on. Every competitor can buy the same seats. What compounds is the layer underneath the licence: the unit-operations reasoning your seniors carry, the correlation library tuned to your top ten client reactors, the client-project archive with hazop history preserved, the data rights renegotiated into the next twenty-four months of master service agreements. A capture-platform built on your project archive is portable across GPT-5, Claude 5, and Gemini 3. The simulator layer commoditises. The capture layer compounds.

Your product thesis. Your board asked for the firm's three-sentence AI thesis last quarter. Your answer was the one from three years ago. Your rate card assumes engineering-hours will price like engineering-hours did in 2020. They will not. Digital-native competitors are quoting fixed-fee feasibility studies against outcomes the client can verify. You quote hours against scope. That is a capital-allocation question about where the moat gets built, not a pricing one. Your CFO and commercial leader need a joint document with you before the digital-native forces the conversation on their terms.

Your R&D bench. You quietly stopped hiring juniors in 2024. The routine simulator work that trained juniors into mid-tier engineers is now running on Aspen AI assistant. No routine work, no training ground, no mid-tier bench in 2030, no internal senior candidate when the current seniors retire. OpenAI and Anthropic both resumed junior hiring in late 2025 specifically to pair juniors with seniors on judgment-intensive work. The process-consultancy that cuts for this year's utilisation number has a cheaper mid-tier in Q4 and an empty senior bench in five years.

Your board-level authority. Thirty percent of European AI-strategy ownership sits with the CIO, twenty-three percent CDAO, twenty-one percent CAIO, one percent COO. You are not on the list. Not because you abandoned the mandate. You have not yet claimed it. Your CEO has a CAIO shortlist. Every month without a visible claim, the thesis drifts further from the seat that can defend it. Your Q1 boundary agreement with the CEO lands before the CAIO shortlist, not after.

Your sector is repricing from below because digital-native process-advisory shops are quoting fixed-fee feasibility studies at forty percent of your rate card. Your function is compressing because five senior engineers with an AI-copilot layer now produce what twelve did, and the hazop bar that worked for twelve cannot hold the AI-drafted volume five produce. Same force, two altitudes. But there is a third thing happening that most mid-market process-consultancy Heads of Digital have not named yet.

The moat that matters in 2028 is not inside the simulation licence. It is underneath it. Aspen will ship AI on top of its simulator. AVEVA will ship AI on top of its platform. Every competitor buys the same seats. What does not commoditise is the unit-operations reasoning your senior engineers carry, the correlation sets your firm has tuned to specific client reactors over twenty-five years, and the client-project-data estate you own the rights to. A digital-native with three people, GPT-5 enterprise, and CAPE-Open access cannot build that layer because it does not have your project archive. You do. The window closes in roughly eighteen months, which is the time between now and when the digital-native has enough project footprint of its own to start building the same layer, or when your most senior unit-operations engineers start retiring and the capture window closes.

In the late 1980s, process plants ran Programmable Logic Controllers on a point-to-point architecture. Honeywell, Yokogawa, and Foxboro shipped the Distributed Control System. Not a bigger PLC. A different control architecture. Plants that upgraded PLCs piece by piece looked fine for five years. By the mid-1990s they could not integrate with the MES, ERP, and advanced-process-control layer that became the competitive unit. Plants that rebuilt around DCS architecture captured twenty to thirty percent throughput gains and held them for a decade. The engineering leaders who could not make the architecture call lost their operating references to the firms that did.

You have twelve to eighteen months. The Heads of Digital who claim the authorship of this decision in Q1 author the decade. The ones who wait until the CAIO hire arrives inherit a seat that has narrowed to licence utilisation and platform uptime, and a scorecard that stopped mattering the quarter their biggest account opened the quote to a digital-native.

Every Head of Digital has had the AI conversation. Most are making the same four decisions. A smaller group is making different ones.

Three questions every process-consultancy Head of Digital is tracking. The third is the crux. The first two are how you earn the right to answer it.

Four concrete moves any process-consultancy Head of Digital can start this quarter. Each addresses one sub-function. Pulled together, they become one architecture the firm builds on.

The process-consultancy Head of Digital has four places to go. None can do what we do.

The big tech-transformation practices ship a standardised AI-governance framework scaled across a sector book. Their cost structure requires reusability. Your seat is idiosyncratic. A reusable framework on an idiosyncratic seat produces governance theatre you cannot defend to the board.

The simulator-vendor professional-services arms (Aspen, AVEVA, Hexagon pro-serve) price the engagement as enablement for licence expansion. Their incentive structure requires more seats. Telling you to spend freed capacity on a firm-owned capture moat instead of expanded seats is structurally impossible for them to say.

The simulator AI add-on vendors (Aspen AI assistant, AVEVA AI Assist, Bentley AssetWise AI) ship drafting and documentation productivity on a per-seat quote. Their product is the productivity gain. Telling you the gain is worth nothing if you take it to margin rather than to capture is not a conversation they are built to have.

An in-house or fractional CAIO splits the seat. Strategy CAIO is revenue-measured; Platform CAIO is adoption-measured. One person cannot hold both. Two-thirds of newly-appointed CAIOs spend year one compiling inventories instead of setting direction.

The four options you have already looked at. None can run the capital-allocation conversation the seat requires. GRAIL is the fifth.

We start with your digital and engineering leadership team because the change cannot outrun the team's personal understanding of what is possible. Three workshops. Each delivers standalone value. Each earns the right to the next.

The questions where your digital leadership cannot agree on an answer are the ones worth an hour on the agenda. Ask these honestly.