What's Changing in the Book You Sign
Your next €10M of capital. The project margin you forecast. The thesis your board can recite back. Three questions that look separate, but are actually one. A briefing for CFOs at mid-market European industrial automation firms.
Thursday morning, audit committee due next Thursday
Every CFO at a mid-sized European industrial automation firm has had the same Thursday morning. Four things on the screen at 8:30. Q3 project-margin variance printed at 4.2 points against a 2-point plan, the second consecutive quarter, driven by two commissioning projects that overran. Your largest integrator at a key OEM lost a €12M bid to a competitor quoting thirty percent shorter timelines on the back of Siemens TIA Copilot. Your CTO wants €2M next quarter for AI engineering-productivity tooling, routed through IT as a Copilot-licence line plus a data platform. Your sponsor's partner forwarded the Bloomberg Klarna story yesterday with a single line: "thinking about your bench plan?"
You have signed twenty forecasts for this board. The pattern used to be simple. Hold the day rate, hold the percentage-of-completion, convert the backlog. Close the quarter clean. Keep the auditor happy on the project-asset recognition. The business was understandable.
Something changed. Your project-margin variance is drifting and your team cannot reconstruct the project-level story without reading the commissioning notes. Your AI capital is running through the IT budget as a productivity line. Your AI-augmented competitors are quoting fixed outcomes on standard PLC work, your rate card is day-based, and procurement teams at your best OEM clients are putting both quotes side by side. Your Controllers are closing faster than they ever have, and the audit committee is asking about traceability on the agentic tooling. The close is clean. The story underneath it is not.
The job has not changed in fifty years. The bar has risen on every measure of doing it, simultaneously, and the contradictions you have been navigating since you took the seat are suddenly re-openable. The CFOs who treat them as engineering problems run the offensive play. The ones who treat them as permanent trade-offs lose capital-allocation authority to the CTO within eighteen months.
This briefing covers what is happening in your sector, what it means for the finance function you run, and the three questions your team should answer before the next audit committee.
Mid-market automation just lost the equation that priced delivery for a decade
Every Siemens TIA Copilot or Schneider EcoStruxure AI deployment at a competitor is compressing standard PLC development thirty to fifty percent. Programming time drops. Commissioning time drops. Documentation drops. The hour baseline your fixed-fee bids were priced against is now wrong. The AI-augmented integrators in your segment are not making this mistake. They quote the same capability on fixed outcome, not estimated hours. Procurement teams at your best OEM clients are now running comparison quotes where your day-rate number sits next to their outcome-priced number with a thirty percent shorter timeline. You are running one pricing mechanic. They are running three. Across your installed base the mechanical difference is three to five points of project margin a year, and it is already showing up in the quarterly prints.
Your competitors are not waiting. The roll-up pattern in European automation is accelerating, and the firms moving first are buying AI-native integrators as tuck-ins rather than trying to rebuild their own delivery model. Some are migrating pricing on their application-engineering work first and the commissioning book second. A small set of the more patient ones are building the application-engineering knowledge asset and the software-content stream that AI-native entrants cannot replicate in eighteen months. They are the quiet compounders. The rest of the sector is shipping Copilot licences to the engineering team at no incremental price and wondering why project margin is compressing.
Your customer value is rising. Your project margin is not. That gap is a pricing decision the finance seat has not yet claimed, dressed as a delivery-team problem.
The sector is repricing underneath everyone at the same time. Siemens and Schneider are compressing standard work. The specialists who carry twenty years of client-environment knowledge are aging out, with no pipeline behind them. The installed-base service stream and the software-content mix are the two places where value is actually compounding, and most firms still report them as rounding-error line items on the monthly pack. The ones that treat this as a delivery-efficiency problem rather than a capital-allocation-and-moat problem are walking into a board conversation they will not win.
Your capital, your forecast, and your narrative are feeling it every day
Your capital allocation. Your CTO's €2M AI-tooling request is sitting in the IT budget, routed through the same mechanism you use to sign off on engineering-station upgrades and software licences. It is not a tooling request. It is the largest single capital-allocation decision you will make this decade, and it deserves the same discipline you bring to a tuck-in acquisition. The finance seats that ship this as a capital line to the board, with alternatives analysis and an ROIC hurdle, keep the authorship for the next ten years. The ones that sign the requisition hand the authorship to the CIO and the CTO by default.
Your forecast. The team built the annual plan on a fourteen-percent project-margin assumption. The Q3 print came in at ten. Your Head of FP&A cannot reconstruct the project-level story because the variance signal lives in commissioning notes in the engineering system, the utilization signal lives in a time-tracking tool that does not distinguish standard PLC work from complex integration, and the pricing-pressure signal lives in lost-bid reports her team has never had access to. The forecast is not wrong because the estimate was aggressive. It is exposed because the data architecture underneath it was built for monthly percentage-of-completion reconciliation, not for continuous reconstruction of the project-margin story.
Your rate-card architecture. The board voted on a commercial strategy that assumed AI tooling would mean efficiency for the engineering team. It is that, and also a pricing question most mid-market automation firms have not opened. Your AI-augmented competitors are on three pricing mechanics. You are on one. The day rate that ran clean from 2014 to 2024 is now the reason your bookings-to-revenue conversion is drifting on the tenders where the comparison is clearest. Rate-card architecture belongs in the capital-allocation conversation, not in the sales-team meeting. The question is where margin compounds, not how to sell more hours.
Your application-engineering and service-engineer bench. The standard PLC programming, commissioning checklists, first-draft documentation, and diagnostic work that used to train a junior engineer into a senior application engineer are now compressing under AI. The Q1 billable-hours win is real. The 2031 senior-engineer bench is empty. The firm that cuts the junior layer without rebuilding the entry pathway has a cheaper quarter this year and no internal technical lead in five.
Your economic thesis. Your board has asked for the strategy-day preview. Your sponsor's partner has asked for an AI-and-headcount note. Your chair has asked how the firm stays ahead of Siemens and Schneider as they compress standard work. Three audiences, three different documents, each drafted by a different team. None of them are anchored in the same data. The CFOs pulling ahead write the thesis once and let the three audiences draw from the same architecture. The ones behind them are editing three decks and hoping the numbers match.
You do not have three scorecard problems. You have one.
Your sector is repricing from below because AI-augmented competitors bring three pricing mechanics to the tender and the rest of the market is on one. Your finance function is stretched because the forecast credibility, the capital-allocation authority, and the economic thesis are each being managed as a separate programme, each drawing on a different data layer, each drafted by a different team. Same force, two altitudes. But there is a third thing happening that most mid-market automation-firm CFOs have not named yet, and it is the one that matters most.
Capital allocation without forecast credibility is guessing. Forecast credibility without a narrative is scorekeeping. A narrative without a data architecture underneath it is marketing. The finance seat that holds all three holds one integrated asset, not three separate programmes. Every capital-allocation question, every forecast-credibility gap, every narrative weakness is a symptom of the same missing layer: the backlog-intelligence infrastructure that feeds all three from the same data, the same governance, the same discipline.
This is the PLC-to-DCS architectural decision of 1998 replayed on a compressed timeline. Every CFO in your cohort is making it right now, whether they see it or not. The integrators that personally owned that 1998 architectural call kept the customer conversation for the next twenty years. The ones who let ABB and Emerson decide because "that was an engineering decision" watched the process-control relationship move to the vendors, and the next automation tender came in at a commoditised rate. The shape is the same. The asset under decision is different. The cost of getting it wrong in 2026 is compressed into eighteen months instead of five years.
You can tell which camp a firm is in from a single question. Can your team answer any material board question by walking the same backlog graph, in the same meeting, with the same data? Most mid-market automation firms cannot. The ones that can have already made the shift, and they are the ones the sponsor is already marking for the next hold cycle.
You have twelve to eighteen months. The CFOs who own this decision personally run the offensive play. The ones who do not become the CFO the board replaces when the successor walks in with a thesis.
What most firms are doing. What the winners are doing instead.
Every automation-firm CFO has had the AI conversation. Most are making the same four decisions. A smaller group is making different ones.
| The default path | The winners' path |
|---|---|
| Route AI capital through ITSign the CTO's requisition. Productivity line in the annual plan. CIO owns the roadmap inside eighteen months. | Route AI capital through the M&A hurdleAlternatives against tuck-in, software-content build, working capital, buyback. One capital line, five options, one choice. |
| Hold the day rate, ship AI gains freeDelivery time drops thirty percent. Price flat. Project margin drifts three to five points. Bookings conversion softens. | Migrate to three pricing mechanicsOutput-based on standard PLC work. Outcome-share on commissioning. Subscription on software content. Margin holds. |
| Run the forecast on a rolling spreadsheetMonthly cadence. Three backlog scenarios take a week. Below-plan quarters produce no project-level story until post-mortem. | Rebuild the forecast as a live backlog graphTen scenarios in an afternoon. Variance reconstruction in the same meeting as the board reads the print. |
| Cut the junior engineer layer for Q1Billable hours up fifteen percent. Empty senior bench in 2031. External technical-lead hire the year after. | Redesign the junior pathway around judgmentJuniors pair with agents on commissioning, application design, client dialogue. Bench rebuilds faster than it would have grown. |
The default path shows a cleaner Q1. The winners' path has visible friction for three quarters and compounds from month fifteen. The difference is whether the CFO is authoring the capital or administering it.
Capital Allocation. Forecast Credibility. Value-Creation Narrative.
Three questions every automation-firm CFO is tracking. The first is the crux. The other two are how you earn the right to answer it.
Where does the next €10M actually compound?
AI-augmented engineering infrastructure? Software-content build that lifts recurring revenue? Tuck-in of an AI-native integrator? Working-capital financing for a lengthening sales cycle? Share buyback? Today's default: the CTO asks, the IT budget funds, the requisition gets signed, and the capital-allocation authority slides away from the seat. The question is not which tool. The question is whose decision this is.
Why did project margin print below plan, and what is the project-level story?
Your board voted on a fourteen-percent plan. Your firm is printing ten. Your Head of FP&A cannot walk the audit committee through the project-level story because the variance signal sits in commissioning notes, the utilization signal sits in a time-tracking tool that does not separate standard from complex work, and the pricing-pressure signal sits in lost-bid reports the finance team has never seen. Three backlog scenarios used to take a week. They now run in an afternoon. The bar has risen on what credible means.
What does this firm do that compounds faster than peers?
Your board wants the strategy-day preview. Your sponsor wants the AI-and-bench note. Your chair wants the how-we-stay-ahead answer. Three audiences, three documents, three teams, and the numbers underneath them do not match. The CFOs pulling ahead write the thesis once and refresh it continuously from the same data architecture that runs the backlog forecast. Three audiences read excerpts of the same document.
What the firms winning this play are actually doing
Four concrete moves any automation-firm CFO can start this quarter. Each addresses one lever. Pulled together, they become one backlog-intelligence asset the firm runs on.
Capital: route the AI decision through the M&A hurdle, not the IT requisition
Your CTO's €2M request is a capital-allocation decision. Treat it like one. Present alternatives on one page: AI-augmented engineering build, software-content pivot, tuck-in of an AI-native integrator, working-capital extension, share buyback. ROIC-scored against the same five-year horizon you use for acquisitions. Name what you are choosing not to fund to protect the thing you are funding. That sentence keeps authorship at the seat.
Forecast: rebuild the forecast as a live backlog graph, not a rolling spreadsheet
Your project-margin model should draw from live bookings, live utilization, live commissioning-note variance, live service-attach data, and live lost-bid reports, as one graph rather than five spreadsheets stitched together. Three backlog scenarios compress from a week to an afternoon. Below-plan quarters produce the project-level story in the same audit-committee meeting, not in the post-mortem. The forecast stops being a deliverable on a cadence.
Pricing: migrate from one rate mechanic to three
Output-based pricing on standard PLC work where AI compresses the hour baseline and the client wants a fixed outcome. Outcome-share on commissioning efficiency or plant-uptime improvement where AI diagnostics make the measurement cheap. Subscription on the software-content layer. The day rate holds the base. The three new mechanics convert the AI delivery gains into sustained project margin on the tenders where AI-augmented competitors are already quoting.
Narrative: write the economic thesis once, refresh continuously
Three sentences on one page. What does this firm do that compounds faster than peers? What capital gets starved to protect it? What gets fed? Refresh the thesis continuously off the same backlog graph that runs the forecast. Sponsor-monthly excerpts the thesis. Audit-committee prep excerpts the thesis. Strategy-day deck excerpts the thesis. Three audiences, one artefact, four registers.
The only firm that joins authorship to working infrastructure
The automation-firm CFO has four places to go. None can do what we do.
The big finance transformation practices sell a twelve-to-twenty-four-month close-acceleration or ERP-modernisation programme staffed by a partner and analysts billing hours. Audit-arm conflicts constrain their governance advice. Their billable-day model requires selling substrate that agents commoditise.
The industrial finance-tech vendors sell platform licences and a long S/4HANA or IFS implementation. Your backlog-intelligence has to sit on firm-owned infrastructure or it becomes a vendor asset at renegotiation.
The FP&A and planning specialists sell a point solution for one layer: scenario planning, close-acceleration, reporting. None can run the capital-allocation-plus-narrative conversation the seat requires.
The AI-native engineering-tooling vendors sell Copilot licences and data platforms where depreciation is steepest. Buying them does not address who authors the capital decision. Faster delivery, the same authorship gap.
The market offers four things you have already said no to. GRAIL is the fifth option.
- Senior advisory that speaks board-room, not deck-room
- Working infrastructure your team uses the same week
- Custom agents built on your bookings, backlog, and data
- The start of a backlog-intelligence layer your FP&A owns
- No vendor lock-in at any phase of the engagement
- Founded by operators who have carried CFO-adjacent seats
- Business first, technology second
- Three weeks from first meeting to working infrastructure
- Your team keeps the infrastructure, you own the IP
- Every next step earned by what just shipped, never assumed
Three sessions with your finance leadership. Working infrastructure from week one.
We start with your finance 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.
How to think
Walk through what is happening in your sector and inside your finance function. Build the first piece of infrastructure in the room: a live backlog agent on your real bookings, running against your real utilization, commissioning-note, and project-margin variance data. Your leadership team leaves with a shared view of what to do next and the first piece of working infrastructure in their hands.
Your finance infrastructure
A half-day build. Your Controller, Head of FP&A, Treasurer, and Head of IR each leave with five to ten pieces of working infrastructure tuned to their function. Percentage-of-completion agents. Continuous backlog scenarios. Covenant monitoring. Board-pack narrative drafting. Infrastructure your team is using the same week, not decks about infrastructure.
Ready to run
Refined infrastructure based on real use. The rollout plan to the wider finance organisation. The capital-allocation-thesis document for the next board meeting. The pricing-migration roadmap for the commercial team. And the CEO and sponsor conversation with numbers they can check: ROIC alternatives, backlog scenarios, software-content progression.
Three weeks. No IT project. No board approval needed to start. The infrastructure your finance function is already running on when we finish.
For your next forecast review
The questions where your finance leadership cannot agree on an answer are the ones worth an hour on the agenda. Ask these honestly.
Your CTO wants €2M for AI engineering-productivity tooling. Is it in the IT budget as a productivity line, or on the capital page as one option among four? If the former, what would it take to move it before the next board cycle?
Your board voted on a fourteen-percent project-margin plan. Your print is ten. Can your team walk the audit committee through the project-level story in the same meeting, or does it take a post-mortem? If the latter, what data architecture is missing?
Your AI-augmented competitors are quoting on two or three pricing mechanics. You are on one. What is the twelve-month plan to migrate from day rate to output-based, outcome-share, and subscription on software content, and who on your team is authoring it?
Your sponsor, your chair, and your CEO have each asked for a different AI-and-bench document in the last ninety days. Are they excerpting the same thesis, or are your teams drafting three documents that do not reconcile?
Of the junior application engineers you are cutting to Copilot-augmented seniors this year, how many are the people who would have made your 2031 technical-lead bench? What are you doing instead to build that bench?
If any of these produces a disagreement in your finance leadership, that is the conversation worth having before the board asks.
Continue reading
Sources and limitations
This briefing draws on GRAIL's engagement experience with leadership teams and on public company disclosures. Figures are illustrative unless a source is named beside them. A sourced edition is in preparation.