Board liaison research

How should board information requests reach the right source owner? A routing study

A response-path study for director questions, authorized source owners, evidence, and reporting boundaries.

Research desk with charts, planning matrix, and executive support capacity materials
Key takeaway: Study each board or committee information request received outside the ordinary packet workflow with frozen definitions, source-lineage checks, explicit authority, privacy controls, and harm measures before changing the workflow.

Preserve the question and its authority

Create an immutable link to the authorized message, meeting note, or committee instruction that contains the request. Preserve the original wording and maintain a separate operational interpretation approved by the accountable executive. If the scope or intended audience is unclear, label the request provisional and seek clarification before source work begins. This prevents a convenient tracker description from quietly replacing a director's question and lets reviewers distinguish faithful routing from a later management decision about response scope. [1]

Separate accountability from delivery

Name one executive accountable for interpreting and accepting the outcome, while allowing several delivery owners and dependencies. A coordinator can chase dates and evidence but should not become the owner merely because the register needs a value. Record acknowledgement from both the accountable executive and the person expected to produce the work. When ownership is disputed, hold the item as unresolved and escalate through the approved governance path rather than assigning it to the most responsive participant. [2][3]

Define response evidence at intake

Specify what would answer the question before work begins: an approved policy, current report, completed filing, signed decision, control record, or verified explanation from a source owner. Evidence should identify its source, observation date, reviewer, and access class. A message saying answered is status, not necessarily evidence. Conversely, do not copy an entire confidential file into the tracker when a controlled reference and approved response are sufficient. The standard should be proportionate to the request and established by the accountable owner. [1][4]

Test follow-up and correction

Review every response that generated a follow-up and code why: missing approval, wrong scope, stale evidence, changed question, ambiguous explanation, or reporting error. Preserve the prior response rather than overwriting it. Compare follow-up cases with a sample that resolved the request, looking for differences in scope definition and source-owner acceptance. Do not infer that every follow-up is failure; a director may legitimately extend the inquiry after seeing current evidence. [1]

Build a minimum-disclosure reporting view

The operational register may contain dependencies, working notes, or sensitive evidence that should not appear in a board packet. Create a board-facing view containing only the approved action, accountable owner, agreed status, due date, material exception, and controlled evidence reference. Apply role-based access to the underlying record and test removal for former participants. A redacted view must not imply that omitted legal, security, personnel, or transaction detail is absent from management's analysis. [4][5]

Governance recommendation

Use support to keep question lineage, acknowledgements, deadlines, evidence references, and approved responses coherent between meetings. Require the accountable executive to define response scope and authorize release; require counsel or other specialists to handle questions within their remit. Review unsupported answers, unresolved source ownership, and access exceptions before delivery. The useful outcome is not a perfect green dashboard, but a board response whose claims can be traced without widening access to sensitive working material. [2][3][5]

Define the routing cohort

Select a fixed set of board or committee information requests received between ordinary packet cycles. Include requests resolved informally, redirected, withdrawn, corrected, overdue, and still open at cutoff. Preserve the authorized source and received time. Do not limit the cohort to items that entered the tracker, because that would exclude precisely the informal messages the study needs to understand. Record exclusions for privileged or unusually sensitive matters and have the appropriate owner approve them. [1][4]

Map the response chain

For every case, reconstruct the path from director question to accountable executive, source owner, reviewer, release authority, and delivery. A request may need several contributors without losing one accountable interpreter. Mark transfers and rejected ownership rather than overwriting the assignee. The chain should show where the exact question changed, where a due date was agreed, and which person accepted the final response. This makes silent handoffs and accidental management answers visible without evaluating a director's motive. [1]

Classify questions before comparing time

Distinguish factual retrieval, policy explanation, financial or operating analysis, legal or governance interpretation, and requests requiring a new management decision. The appropriate evidence and review path differ. Comparing all response times together would reward easy retrieval and obscure requests deliberately held for counsel or committee process. Freeze definitions before reviewers see outcomes, test a varied sample independently, and retain an uncertain category when a question legitimately spans classes. [1][5]

Test source-owner acceptance

A name entered by support is not ownership. Require the proposed source owner to accept, redirect, or identify missing scope. Measure the time to accepted ownership separately from production time. Review requests that bounced between functions and ask whether the question, audience, evidence standard, or accountability was unclear. Do not penalize a source owner for refusing to certify analysis outside the function's remit. A correct refusal can prevent an unsupported board response. [1][2]

Audit claim-level lineage

Select material claims from completed responses and trace each to a current source, observation period, qualified owner, and approval. A polished narrative may combine several sources with different dates. Record qualifications and contradictions rather than choosing the convenient number. Where a controlled source can be referenced, avoid copying it into the routing register. The study should identify unsupported or stale claims without creating a second unofficial board archive. [1][4]

Study corrections as a separate outcome

For every corrected answer, preserve the initial release, changed claim, reason, approving owner, audience notified, and elapsed time to correction. Differentiate an expanded question from an error in the original response. Check whether the same source or routing failure recurs. Correction frequency alone can mislead: a transparent process may record more corrections than an informal process that silently replaces files. Pair counts with materiality and evidence of audience receipt. [1]

Model access by audience

The director, committee, management team, counsel, source function, and board support role may each need different parts of the record. Test access with representative accounts and verify removal for former participants. Keep privileged, personnel, transaction, and security material in its authorized source. A board-facing response can link to a controlled appendix without exposing working comments. Report access exceptions and workarounds as harm signals rather than solving retrieval by widening permissions. [4][5]

Account for off-register communication

Interview a small set of authorized participants about request channels, then reconcile those descriptions with the cohort. Directors and executives may use calls or direct messages that never enter the formal route. The study should not collect private content merely to improve completeness. Instead, code the existence of an off-register path, whether an authorized owner later preserved the request, and whether the final response entered the proper record. Report the blind spot and avoid presenting the register as a complete map of board deliberation. [1][4]

Pilot a closed-loop receipt

Test a compact receipt that preserves the question, intended audience, accountable executive, accepted source owners, evidence standard, due date, review state, and approved delivery. Define who can pause the pilot and what happens when ownership remains disputed. Compare reconstruction, overdue requests, unsupported claims, access exceptions, and correction handling with the baseline cohort. The pilot should not change formal board authority or replace minutes, resolutions, committee charters, counsel channels, or other governing records. [1][5]

Interpret latency without ranking people

A long response interval can reflect an unclear question, unavailable evidence, deliberate legal review, multi-owner analysis, or an avoidable routing gap. Break the interval into received-to-ownership, ownership-to-evidence, review, and approved delivery where records allow. Pair timing with reason codes and case review. Do not publish individual league tables or infer diligence from speed. The operating question is whether the path made responsibility and constraints visible early enough for the accountable executive to manage the request. [1][2]

Set the adoption decision

Continue only if question lineage survives handoffs, owners accept responsibility earlier, evidence can be traced, approved audiences receive the response, and confidentiality does not weaken. Revise the receipt if it becomes a second packet workflow or collects sensitive working detail no routing decision needs. Stop if management treats the tracker as authority to interpret directors or bypass specialist review. Preserve the protocol, field definitions, aggregate results, limitations, and sponsor decision for later replication. [1][4]

Sources

  1. Assessing Data Reliability, U.S. Government Accountability Office.
  2. Executive Secretaries and Executive Administrative Assistants, O*NET OnLine.
  3. Secretaries and Administrative Assistants, U.S. Bureau of Labor Statistics, Occupational Outlook Handbook.
  4. NIST Privacy Framework, National Institute of Standards and Technology.
  5. Security and Privacy Controls for Information Systems and Organizations, National Institute of Standards and Technology.
  6. Executive Secretaries and Executive Administrative Assistants, U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics.
  7. Business Formation Statistics, U.S. Census Bureau.
  8. Nonfarm Business Sector: Labor Productivity, Federal Reserve Economic Data.
  9. Productivity Statistics, OECD Data Explorer.
  10. World Development Indicators, World Bank DataBank.
  11. ILOSTAT Labour Statistics, International Labour Organization.
  12. 2024 Work Trend Index Annual Report, Microsoft and LinkedIn.
  13. The economic potential of generative AI, McKinsey Global Institute.
  14. Creating helpful, reliable, people first content, Google Search Central.
  15. Search Engine Optimization Starter Guide, Google Search Central.
  16. Dear Manager, You Are Holding Too Many Meetings, Harvard Business Review.

Related research