AI value creation for PE-backed mid-caps and growth participations across Europe. The discipline that defends the multiple at exit.
Discuss your situation →Operator first. CDO at John Paul (acquired by Accor 2018). CTPO at Weekendesk under Otium Capital. Group CDO at Effy Group post-merger. Most recently MD Tech and Chief Transformation Officer at Altavia Group, 40+ countries. INSEAD GEMBA, CentraleSupélec engineering.
Funds that treat it as a vendor-selection exercise will discover the cost at exit. Funds that treat it as an operating-model question will defend their multiple.
Five Comex mandates as Chief Transformation Officer, Group CDO, CTPO. AI governance framework deployed across 40+ countries. 800M€ group restructured.
The discipline of execution · I kill more initiatives than I launch, I tie every initiative to a P&L line in 90 days, I remove layers before deploying tools.
Strategy without execution is fiction.
Execution without strategy is noise.
Technology without purpose is waste.
France Invest 2025: portfolio holds are extending. Exits are concentrated.
AI is rewriting the operating model under everyone at the same time.
AI requires deep operating-model rewiring · 18 to 24 months minimum. Most PE holds left in flight have less than 36 months before exit. The math is unforgiving: start now or write down the AI thesis at exit.
Wrong AI moves show up in three quarters and get corrected. Inaction shows up at exit and cannot be. The Partners who confuse prudence with paralysis will discover prudence has a price tag · paid in the multiple they will not defend.
The pressures above don't kill portcos by themselves. These three patterns do · because they prevent the portco from absorbing the pressures in time. Each shows up in the multiple at exit if left untreated.
The first sign is rarely operational. It is verbal · the value creation plan stops being mentioned in the weekly Comex routines.
The signal · the value creation plan is precise, the equity story is clean, but the routines managers actually run don't match either.
What happens · the mid-hold review surfaces 4 quarters of activity that didn't move EBITDA. The exit narrative becomes harder to evidence quarter after quarter.
The first sign is rarely the budget. It is governance · the AI line item stops having a single owner answerable at the next board.
The signal · twenty pilots running, three of them serious, none of them owned at GM level. AI spend is now a line in the budget · returns are not.
What happens · every quarter that passes makes the gap harder to explain to the LPs. The fund's AI thesis at acquisition becomes a liability at sale.
The first sign is rarely the org chart. It is the customer · NPS plateaus while internal coordination meetings multiply.
The signal · layers built for a pre-digital information flow, middle management overloaded with coordination, customer signals trapped in reporting cycles.
What happens · the portco still works. But at exit, EBITDA reflects activity, not value · and buyers can see it in diligence.
Each pattern has a recovery window. None of them gets cheaper to fix in month 24 of the hold than in month 6. I take ownership at Comex level · defined scope, measurable thesis, clean exit when the work is done.
The structure adapts to the moment in the holding cycle and the level of execution required.
Full Comex accountability. Day-to-day execution alongside the management team. Mandates of 6 to 24 months.
Taken when · the value creation plan needs hands-on delivery, not advice. Declined when · no GM-level P&L ownership of the AI initiative is possible.
Reference · Effy Group 2021 to 2022 (Group CDO, 5 entities, 80+ team).
Pre-deal tech and AI diligence. Sparring partner for management teams across the portfolio. Board advisor on digital transformation. Cross-portfolio AI playbook.
Taken when · the fund holds 4+ portcos with shared AI questions. Declined when · the mandate is bolt-on advisory without portfolio-wide leverage.
Reference · Otium Capital 2009 to 2010 (Operating Partner, VC fund).
Short, structured diagnostic on a specific portco or thesis. Output · a clarified bottleneck, ranked value creation levers, named owners, decision-grade recommendations.
Taken when · the fund needs to test whether deeper engagement is warranted. Declined when · the request is a desk study without management access.
Reference · 4Q Method applied across 5 mandates.
Selected operating engagements across PE-backed mid-caps and large groups in France and Europe.
40+ countries. 70+ BUs. €800M GMV.
MD Tech and Chief Transformation Officer · 40+ countries, 70+ business units, family-owned with minority PE backing. The 80M€ roadmap was the easy part. Getting 70 BUs to converge on one operating model · that was the work.
$150M exit to Accor. 100 → 1,000 FTEs scaled.
Built the data and AI foundation that scaled HNWI customer experience across 12 countries. The exit to Accor in 2018 was the punchline. The eighteen months that preceded were the work.
5 entities consolidated post-merger.
Took the Group CDO seat as the merger was still settling. Five tech stacks were duplicating each other. The first 90 days were diagnostic. The 18 that followed were rationalisation. The market shifted mid-mandate · the discipline survived.
GMV €15M → €45M during the scale-up.
Chief Product & Technology Officer during the scale-up phase. The product was working. The architecture was not. Time-to-market dropped 50% before the GMV tripled.
Cross-portfolio diligence on 7 portcos.
Operating Partner inside the VC fund. Cross-portfolio digital diligence on 7 portcos. The discipline of looking under the hood before forming an opinion · this is where the operating reflex was built.
The discipline before the slides. The principles before the contract.
Where AI creates value. And where it destroys it.
If your fund holds a portco where AI capex is rising faster than EBITDA,
it is worth a conversation.
