A poorly framed letter of instruction is one of the most common — and most avoidable — sources of weak expert evidence. Experts can only answer the questions they are asked, within the scope they are given, on the basis of the material they are provided. If that framing is vague, overbroad, or misaligned with what the tribunal actually needs to decide, the resulting report inherits those problems no matter how skilled the expert is. This piece sets out what a well-constructed instruction should contain.
Start with the questions the tribunal needs answered
The single most valuable discipline in drafting an instruction is to work backward from the issues the tribunal will actually need to decide, rather than forward from whatever material happens to be available. A delay expert instructed simply to "review the programme and advise" will produce a report with far less forensic value than one instructed to determine, for example, the cause and quantum of critical path delay to a specific milestone, using a specified or reasoned methodology, by reference to defined categories of concurrent delay.
Precision here does two things. It gives the expert a clear brief against which to test their own analysis, and it gives the tribunal a report structured around the actual questions in dispute rather than a general narrative the tribunal must then mine for relevant conclusions.
What the instruction should specify
A complete letter of instruction to a construction or engineering expert should address:
- The precise scope of the opinion sought — the specific factual and technical questions the expert is being asked to answer, not a general invitation to "assess the claim"
- The applicable procedural framework, including the institutional rules (DIAC, arbitrateAD, DIFC Courts) and any tribunal directions already in place governing the form, timing, and exchange of expert evidence
- The documents and information being provided, with a clear statement of what has and has not been made available, so the expert can identify gaps early rather than discovering them mid-report
- Any facts or assumptions the expert is being asked to take as given, clearly distinguished from facts the expert is expected to establish or verify independently
- The relevant contractual framework, including the applicable FIDIC form (Red, Yellow, Silver, or Emerald Book, as relevant) and any bespoke amendments that bear on the questions in dispute
- Confirmation of the expert's overriding duty to the tribunal, and an explicit acknowledgment that this duty applies regardless of who is instructing or paying
Common instruction failures worth avoiding
Certain patterns recur often enough in practice to be worth naming directly:
- Scope creep by omission. An instruction so broad that the expert effectively defines their own remit invites inconsistency between what the tribunal ordered and what the report delivers.
- Buried assumptions. Handing an expert a set of "agreed facts" without flagging which are genuinely agreed and which are simply the instructing party's characterisation of disputed matters. Experts who accept these uncritically produce reports vulnerable on cross-examination.
- Selective document sets. Providing an expert with a curated subset of the project record, rather than the full relevant set, is one of the fastest ways to produce an opinion that later has to be qualified or withdrawn once the complete record surfaces in disclosure.
- Instruction drift. Supplementary instructions delivered informally over the course of the engagement, rather than being documented and incorporated formally, create exactly the "off the record" instruction problem discussed in Part 3 of this series.
Why this matters commercially, not just procedurally
For lawyers managing cost-sensitive clients — a recurring concern for contractors and engineering businesses in particular — a well-framed instruction is also the more efficient one. An expert working from a precise, well-documented brief spends less time seeking clarification, produces a first draft closer to what is ultimately needed, and is less likely to require substantive rework once gaps in the instruction surface late in the process. The upfront care invested in the letter of instruction is, in practical terms, time and cost saved later in the engagement.
Part 6 turns to the practical work that follows instruction: site inspections and evidence gathering on live construction and infrastructure projects, and what a defensible inspection record looks like.