Red flags

Twelve red flags to check before an acquisition.

A free checklist to prepare your questions for management. Each flag describes a possible risk and the check that lets you measure it.

Flags 01 to 03

Revenue exposed to AI.

01

Part of the revenue pays for work that AI can already produce

Writing, translation, data entry, first-line support, content: when competitors produce for less with AI, the target’s prices or volumes fall in turn.

CheckBreak revenue down by service and by billing model.

02

Time-based billing dominates

A client who knows a task takes less time with AI renegotiates the rate or the number of days billed.

CheckMeasure the share of revenue billed on time spent and track the average daily rate over 24 months.

03

Clients ask for price cuts and cite AI

The signal shows up in tenders and renewals long before it appears in the annual accounts.

CheckReview lost tenders, loss reasons and discounts granted over the past year.

Flags 04 to 07

AI tools poorly measured or poorly controlled.

04

Savings attributed to AI are not measured

Management presents productivity gains without documenting the baseline, the measurement period or the volumes processed.

CheckAsk for the basis of calculation, then reconcile it with time spent and headcount.

05

A critical automation depends on a single person

A departure or a disabled personal account can stop a process that touches billing or client relationships.

CheckList the automations in production, their owners and the accounts that run them.

06

Paid AI licenses are barely used

Low usage points to a poorly chosen tool, untrained teams or a process that has not changed.

CheckCompare active accounts, frequency of use and licenses billed.

07

Teams use AI tools that management does not know about

Client or personal data goes to unlisted services, outside any contract.

CheckCross-check declared tools, expense reports, subscriptions and team interviews.

Flags 08 to 10

Data the target does not fully control.

08

Client contracts restrict the reuse of data

An AI project in the business plan may rely on data the target processes for its clients without the right to reuse it.

CheckReview the ownership, confidentiality and purpose clauses of the main contracts.

09

Data needed for AI projects is missing

History too short, fields poorly filled in, duplicates: the announced value assumes data the company does not have yet.

CheckMeasure the depth of history and the fill rate of the fields the projects need.

10

An AI vendor hosts data outside the European Union

The transfer may require contractual safeguards missing from the contracts the target signed.

CheckIdentify the hosting locations, subprocessors and transfers for each tool.

Flags 11 and 12

A business plan too optimistic about AI.

11

The business plan counts on AI-driven value with no budget

Future margin is then overstated: the investment, licenses and operating cost that make this value possible are missing.

CheckFor each announced saving, ask for its cost, its timeline and its owner.

12

No one is assigned to lead the AI projects

Without an available owner, a project falls behind and its value arrives later than the business plan expects.

CheckFor each project, identify the owner, the decision-maker and the skills available.

A red flag in your deal?

Pre-LOI AI screening · €15,000 excl. VAT · Before the letter of intent

Discuss an AI screening
FR