The phantom engagement

Why this matters to you right now

Cal Newport, in Slow Productivity, articulates a rule most senior consultants learn by accident: a practice's productivity is inversely proportional to the number of simultaneously open commitments. Not the number of active engagements, the number of open commitments. The distinction matters. An engagement that moves consumes the attention it deserves. An engagement that no longer moves consumes the same attention with no return. That is the phantom engagement. It no longer fills your calendar but it fills your head. Atul Gawande, in The Checklist Manifesto, documents the value of protocols in high-stakes situations. The phantom engagement belongs to that category: weak signals ignored, sudden acceleration, late irreversibility. We learn to sign. We never learn to close.

It is May 5, 2026. Tax season has just ended. Most SMEs are redirecting attention to strategic projects. Consultants who signed in 2025 and have been carrying an engagement since Q4 are entering the window where the client is about to ask "where are we?" again. The window to act is between now and end of June, before H2 planning locks the second-half commitments.

Most senior consultants live through one or two phantom engagements per year. No major publication addresses the topic head-on. It is the operational blind spot of consulting, because nobody signs an engagement intending to see it linger. The client acts in good faith. The consultant too. And yet the drift sets in. Not at signing, in execution. The skill of recognizing the five signals that announce the drift is what separates the methodical practice from the intuitive one. As in edition #7 on the engagement to decline, upstream prevention matters. But once the signature is in, the playing field changes.

If you cannot name a single phantom engagement in your practice right now, you are probably carrying one without knowing it.

Pricing: write the indicators into the engagement letter

The action: Inscribe five quantitative indicators as contractual clauses in your service offer and engagement letter, with a negotiated reactivation mechanism. The client and you read the same dashboard. The phantom engagement is prevented by contractual design, not by individual vigilance.

Why it works: Blair Enns, in Pricing Creativity, argues that the proposal is the contract and that the quality of the conditions predicts the quality of the engagement. Ron Baker, in Implementing Value Pricing, pushes further: scope traceability is mutual protection, not a mistrust mechanism. The five indicators to insert in your next offer:

1. 1.Timeline replanning beyond two weeks: a new milestone documented and signed by both parties.
2. 2.Change of more than 30% among client-side decision-makers: a rescoping meeting within 15 days.
3. 3.Out-of-scope hours beyond 15% of the initial budget: a written change order before execution.
4. 4.Two consecutive meetings without a documented decision: a formal reset.
5. 5.Client-to-consultant response-time ratio above 3 to 1 over a rolling two-week window: a calibration call.

The trap: Believing that asking for these clauses costs you the engagement. In practice, they filter clients who are not truly committed and they harden the commitment of those who are. A client who refuses the five indicators is a phantom signal before the signature. A second, more structural trap: if your pricing model is hourly, the indicators are less effective because the drift bills itself without flagging itself. See edition #10 on the pricing model as liability. The fixed-fee with milestones is the pricing structure that makes the five indicators truly operative.

This week: Draft the five clauses in your usual contractual voice. Thirty minutes for a first version, then validate with your usual legal advisor before insertion into a real offer. The engagement you sign this month will not become next spring's phantom.

Sales and business development: transparency as a commercial argument

The action: Present the five indicators during discovery, before the signature, not buried in a contract appendix. Make operational rigour a commercial differentiator. The cover page of your next proposal could read "Our shared success framework" and list the five indicators as a promise to the client, not as a protection mechanism against them.

Why now: David Maister, in The Trusted Advisor, frames trust as the product of transparency about failure conditions. The consultant who talks about deadlock conditions before they occur takes the ethical high ground over competitors who don't. Liz Wiseman, in Impact Players, documents that candour is the distinctive mark of the senior contributor. Not flattery. Not diplomatic caution. Operational candour.

The hidden benefit: A client who validates the five indicators during discovery feels treated as a partner, not as a commercial target. They sign with a higher level of commitment because they have already mentally accepted the discipline of the engagement before the first invoice. Same logic as edition #8 on selling without selling: demonstration converts, argument doesn't. While preparing this edition, I tested the framing with three peers. All three said the same thing: they had never thought of the closure conditions as a signing argument.

The trap: The wording. "Here is how we give ourselves the means to succeed" works. "Here is how I protect myself from you" fails. The words describing the indicators must put the client and the consultant on the same side of the table. If the client reads the document and feels surveilled, the mechanism backfires.

This week: Build the cover page of your next proposal that frames the five indicators as a promise to the client. Fifteen minutes of writing, three reads aloud, one revision to fine-tune the tone. Use it in your next discovery meeting.

Collaboration networks: your peer is your phantom detector

The action: Set up a quarterly peer audit. A trusted peer reviews your active engagement list with the five indicators in hand and flags what you no longer see. You return the favour the following week. The solo practice has a blind spot on its own engagements. Another consultant sees what you have stopped seeing.

The mechanism: Patrick Lencioni, in Death by Meeting, shows that the leader's blind spots are systemic, not accidental. The more daily attention is captured by delivery, the less structural diagnosis is possible from inside. Dorie Clark, in The Long Game, runs in the same direction: peer communities among senior consultants are a longevity condition, not a social nicety. The peer is not a confidant, the peer is a detection instrument. This strategic use of the peer extends the work of edition #2 on the network you neglect: your best network is not the one that feeds you engagements, it is the one that protects you from your blind spots.

The format: A 60-minute peer audit, structured in three steps. The active engagement list is shared 24 hours ahead, with the five indicators already calculated. Three mandatory questions guide the conversation. Which engagement looks the most at risk of becoming a phantom? Which one did you hesitate to add to the list? If you had to close one this week, which one? The third question is the one that forces the call. Without it, the audit stays descriptive.

The trap: Choosing a peer who validates you instead of a peer who challenges you. The good auditor asks the awkward question. If the audit feels good every single time, it isn't an audit, it's therapy. An effective peer will make you slightly uncomfortable at least once per session. That is the sign the mechanism is working.

This week: Identify a trusted peer who challenges you. Propose a first cross-audit. Schedule the session before May 30.

Value creation: the honourable closure protocol

The action: Build a four-step procedure (shared diagnosis, options on the table, executed closure, conditional restart) that closes a phantom without burning the client relationship. Honourable closure is not a firing. It is an explicit transition that preserves the relationship and frees up capacity.

Why it changes everything: Joey Coleman, in Never Lose a Customer Again, frames the exit phase as the decisive moment of the client relationship, more important than signing. Atul Gawande, in The Checklist Manifesto, documents the power of protocols in emotionally charged situations. Patrick Lencioni, in Getting Naked, goes to the bone: the consultant's vulnerability when saying "this is not working anymore" is what separates the advisor from the vendor. The phantom engagement is a charged situation. The protocol protects both parties from emotional improvisation.

The four-step protocol:

1. 1.Shared diagnosis. A structured meeting where both parties lay down the facts: the five indicators, the history, the blockers. No interpretation, no blame.
2. 2.Options on the table. At least three options proposed by the consultant: reset with a new frame, transition to a closure deliverable, stop immediately. Each option carries a cost, a timeline, and a deliverable.
3. 3.Executed closure. If the closure option is chosen, a transition deliverable (15 to 30 days) that synthesizes what was produced, what remains, and what the client can do alone. No grey zone.
4. 4.Conditional restart. A future re-engagement option with explicit triggers ("if X happens before date Y, here is what we do"). The door stays open without an active commitment.

The test: At the end of the procedure, the client should be able to refer you to a peer without embarrassment. If the exit creates shame, the protocol failed. The contrast with firing matters. See edition #12 on the clients you should fire. Firing severs a toxic relationship. Honourable closure preserves a neutral relationship that has gone stagnant. The distinction is structural, not semantic. And after closure, the four post-engagement anchors of edition #20 on the client you abandon take over to maintain the connection.

This week: Draft your honourable-closure email template. You should be able to send it in under 30 minutes the day you need it. Four paragraphs: shared diagnosis, options proposed, transition deliverable, conditional restart. Keep it in an accessible folder. The phantom engagement arrives without warning.

AI: the automated dashboard for weak signals

The action: Set up a Claude Project (or a Custom GPT) with the context of your active engagements and the five indicators as a weekly analysis grid. Thirty minutes on Friday morning, by hand, no automation. That's version 1, accessible to any solo consultant this week. The target version (calendar, email, and CRM connected to an agent that computes continuously) comes later. AI does the monitoring nobody does by hand, because the daily attention cost is precisely what makes humans blind to cumulative weak signals.

Ethan Mollick, at One Useful Thing, has insisted for two years on a consultant-underused application: AI as a continuous monitoring system, not as a content generator. Pattern detection in operational flows is the highest impact-over-effort use case in 2026. Eric Ries, in The Lean Startup, documents the value of short feedback loops to detect weak signals before they accumulate into irreversible drift. Rita McGrath, in Seeing Around Corners, formalizes the concept of strategic inflection points. The phantom engagement is an operational inflection point. It obeys the same mechanics: ignored weak signals, sudden acceleration, late irreversibility. Exactly the kind of use edition #14 on AI for senior consultants advocated: AI as a judgment amplifier, not as a substitute.

Suggested configuration. Inputs: Calendar exports, Gmail or Outlook threads per client, CRM exports, meeting minutes. Cadence: weekly Friday-morning analysis. Outputs: per-engagement risk score on a 0 to 100 scale, the most active signal, an action recommendation. Guardrails: tunable thresholds, mandatory human alert, never an automatic action on the client side. You stay in the decision loop at all times.

The warning: The agent can alert, it cannot decide. The final call (reset, transition, close) remains a human judgment. The real risk isn't that the AI gets it wrong. The real risk is that it gives you a pretext to postpone the difficult conversation. "The score is only 67 out of 100, I'll keep watching" becomes the new "let's revisit that." AI buys you detection time, it does not buy you courage.

While writing this edition, I set up the dashboard on three of my active engagements. Forty-five minutes to bootstrap. One engagement came out at 58 out of 100, higher than my gut would have placed it. Not a phantom, but one more signal than I had been seeing with the naked eye.

This week: Block 30 minutes Friday for the first analysis. If an engagement comes out with a score above 70, schedule the peer audit from the previous section before the end of the month.