Management consulting pipeline is often created through relationships that do not fit a campaign field. A partner introduces a director, a former client opens a door, a founder speaks at an event, and a content asset helps an already-known account decide. A risk register keeps those paths visible while protecting the firm from claiming that one touch caused the work.
1. Name the decision
State the decision the source report informs: partner investment, practice focus, event choice, content, hiring, or forecast. Add owner, period, minimum evidence, and cost of error.
2. Record the source fact
Capture exact source, date, person, account, channel, referral permission, and what the buyer said. Separate reported origin from inferred influence.
Use a controlled path for sensitive client and relationship details. Do not copy confidential context into a broad report.
3. Define attribution views
Keep first touch, latest touch, partner, founder, account, and influence views distinct. State unit, denominator, window, exclusions, and confidence.
Digital events can support the layer; Google Analytics key events are not proof of a consulting decision.
4. Assess qualification risk
Check problem, role, authority, urgency, service fit, geography, capacity, partner relationship, and delivery path. A source with many introductions may still generate work the practice should decline.
Record accepted, deferred, referred, disqualified, proposal, won, and lost states.
5. Assess relationship and permission risk
Mark client reference, partner permission, personal data, restricted commercial terms, and customer-facing claims. Define who can view, export, change, or publish the evidence.
If a source is uncertain, hold the claim rather than turn a relationship note into public proof.
6. Assign ownership and repair
Give each issue likelihood, impact, evidence, owner, action, due date, mitigation, and stop condition. Repairs may include source vocabulary, referral form, event code, CRM stage, or partner handoff.
Keep a rollback and a before-after sample. Do not change every report to solve one unclear record.
7. Review commercial consequence
Compare source, cycle length, effort, fee, margin, delivery fit, repeat work, and client outcome. A source may deserve investment because it creates trust, not because it wins an attribution contest.
Mark uncertainty and keep the cash or capacity decision separate from marketing activity.
8. Govern claims and content
Before a source claim enters a case study or page, check evidence, date, permission, market, and wording. Google’s people-first content guidance supports explaining a real buyer decision instead of overstating a source.
9. Use the risk register
| Field | Record | Escalate when | | — | — | — | | decision | choice, owner, period | public or irreversible action | | source | fact, date, confidence | origin is inferred | | attribution | view, unit, denominator | causal claim is proposed | | qualification | fit, stage, decline | source quality is hidden | | relationship | permission, sensitivity | client or partner risk | | repair | owner, test, rollback | change cannot be isolated |
Review monthly and after a major referral, event, loss, or partner change. The register is useful when it supports honest investment and makes weak evidence visible before it becomes a public promise.
How to review a disputed source
When two people classify an opportunity differently, do not resolve the conflict by choosing the most convenient label. Read the original conversation, proposal, partner note, and customer-confirmed next step. Record what each source proves, what it does not prove, and which definition should govern the report. Give the decision to one owner and add a review date. This process is slower than a silent edit but protects the next comparison.
Include delivery in the source conversation. A source that creates a proposal but not a workable engagement may require a different offer or qualification route. A partner that sends fewer but better-fit opportunities may deserve investment even if a last-touch report gives credit to a later event. Show the views side by side and explain the decision each one supports.
Finally, keep a hold state for any source claim that lacks permission, context, or corroboration. The hold should say what evidence is needed and who can close it. That discipline lets the firm use useful internal learning while keeping public and client-facing claims defensible.
For search context, Search Console performance can show language and page observations; it cannot prove that a consulting relationship was created by search.
Include a simple reconciliation table in the monthly review: source as reported, source as recorded, latest touch, partner or founder influence, qualification state, and delivery outcome. The purpose is not to force one answer. It is to show where the firm has evidence and where it has an interpretation. Use the table to choose the next research question or process repair. Over time, the disputed rows become a useful training set for better vocabulary and more honest reporting.
When leadership chooses a partner or channel action, record the evidence threshold and the review date. A small relationship-led route may deserve a pilot even when attribution is incomplete, but the pilot should name what will be learned: fit, response, client permission, delivery effort, or repeat work. This turns uncertainty into a bounded decision and prevents the firm from either overclaiming or ignoring useful evidence.
Keep a separate note for any source that is used in a public claim. It should identify permission, wording, date, market, proof, and expiry. Internal learning can remain directional; public language needs a stronger gate. When a client or partner corrects the source story, update the register and the asset rather than treating the correction as an isolated editorial issue.
How did this article land?
Choose one reaction. You can change it anytime.
