# Adrien de Malherbe: consulting deliverable examples

Reviewed 17 November 2026. These fictional examples illustrate a deliverable format. They are not client results, quotations or delivery commitments. The written proposal defines scope, responsibilities, timing and fees for each engagement.

## 1. Europe market-entry decision memo

Decision: Should a fictional B2B SaaS company fund its first local seller or validate demand with its founder?

Evidence register:

| Observation | Source to retain in a real engagement | Interpretation | Unresolved question |
| --- | --- | --- | --- |
| Two French evaluations have named owners | Dated buyer-confirmed evaluation records | The team can investigate a specific buying process | Budget, pilot commitment and delivery scope |
| One German enquiry needs an integration | Customer requirement and product assessment | Delivery feasibility is not yet established | Owner, effort and support implications |
| Founder can conduct initial discovery | Agreed capacity plan | A bounded test may be feasible before hiring | Time allocation and product support |

Recommendation: resolve buyer commitment and delivery requirements before authorising the first local hire. Record the decision owner and spending limit in the real engagement. Compare a founder-led test, a first seller and a larger team using editable monthly cash assumptions. Keep bookings, recognised revenue and cash receipts separate.

Decision gates:

- A buyer confirms the problem, evaluation owner and next decision.
- Product and service owners confirm whether the required delivery is feasible.
- Finance checks cash exposure under delayed payment.
- The founder approves the next commitment and review date.

Handover: decision memo, evidence register, editable cash model, unresolved assumptions and a 90-day experiment plan with owners. The 90-day plan describes the customer's experiment; it is not the promised duration of the consulting engagement.

## 2. Sales-execution diagnostic finding

Question: Why are opportunities not progressing to a buyer-owned next step?

Observation: four of six opportunities in a fictional selected review sample lack a recorded, buyer-confirmed next step. The sample is small and selected; it does not establish a team-wide rate or the cause of stalled deals.

Competing explanations:

| Explanation | Evidence to inspect | Owner |
| --- | --- | --- |
| Buyer has not agreed to continue | Actual meeting record and buyer communication | Opportunity owner |
| A step was agreed but not recorded | Original dated source and CRM entry | Opportunity owner |
| Stage definition does not require a next step | Current definition and review practice | Sales manager |

Proposed action: verify the six records, separate missing evidence from missing commitment, and agree the stage rule. Review a fresh comparable sample at the next operating meeting.

Decision rule: if the main issue is missing documentation, test a recording and review change. If buyers have not committed, inspect qualification and value before prescribing a CRM redesign or additional headcount.

Handover: prioritised findings, evidence references, alternative explanations, action owners, review dates and criteria for retaining, revising or stopping each experiment.

## 3. AI workflow pilot specification and evaluation

Task: create a reviewed account-preparation brief from approved sources.

Input: fictional Northstar Logistics CRM note describing manual exception reviews; a dated public vacancy for an operations analyst. Budget, procurement owner and buying date are unknown.

Output format:

- Observed fact: manual exception review is recorded in the supplied CRM note. Include the source ID and date.
- Observed fact: an analyst vacancy appears on the supplied public page. Include the source URL and capture date.
- Hypothesis: reducing review effort may be relevant. A vacancy alone does not establish purchase intent.
- Unknown: budget, decision owner, quantified workload and timing.
- Discovery question: how much effort does the review require, and who owns improvement of that process?

Evaluation:

| Measure | Method | Acceptance to agree before testing |
| --- | --- | --- |
| Factual support | Check each material claim against its source | No unsupported material claim in accepted output |
| Missing information | Include absent, outdated and conflicting input cases | Unknowns and conflicts remain visible |
| Time and effort | Compare preparation plus human review and correction on the same task set | Meet the agreed net-effort target |
| Usefulness | Reviewer applies an agreed rubric | Meet the agreed task-specific threshold |

Use a held-out task set for the final evaluation. Record sample size, failure cases and correction effort. Do not infer commercial uplift from faster drafting alone.

Handover: workflow specification, scoped prototype, evaluation sheet, failure log, owner checklist and adoption/revision/stop recommendation. Production integrations, software costs, customer messaging and ongoing operation require a separate agreed scope.

## Scoping a real project

Describe the decision, deadline, current evidence, team and constraints at https://adriendemalherbe.com/contact/ . Do not include confidential customer records in the initial enquiry. Price on request for all three engagements. Fees and delivery dates are confirmed in a written proposal after scope and dependencies are checked.
