Identifying Cargo Cult Engagements
The cargo-cult engagement diagnostic is a tool for identifying cargo-cult engagements by recognizing their pattern from the inside.
When Everything Looks Productive
Imagine you are halfway through an AI engineering engagement. Meetings are on the calendar, documents are being written, dashboards are being built, and the team keeps shipping artifacts. From the inside, the work feels like progress. Yet the source behind this article points out that most engagements in AI engineering do not announce that they are going badly. Work continues, meetings happen, and artifacts get produced, so an engagement can feel productive even when it is not.
This is the problem the cargo-cult engagement diagnostic is meant to address. It is a tool that helps a practitioner notice this kind of trouble while there is still time to act on it. Its design is deliberately small: it is built around a short set of questions rather than a long checklist. In this article you will look at when the diagnostic runs, what it is for, why a few questions can be enough to recognize a failing engagement, and how to tell a passing result from one whose answers come back wrong.
Where the Diagnostic Sits in an Engagement
The easiest way to understand the diagnostic is to see where it fits in time. The source describes a single path: the engagement runs for some time, reaches a four-week checkpoint, and then the five questions are applied. Trace that path in the timeline below, and notice that the checkpoint is a fixed point on the calendar rather than a reaction to a problem someone has already spotted.
The source states that the five questions are run on every engagement at the four-week mark. The word every matters. The diagnostic is not reserved for engagements that already look suspicious. It is a routine checkpoint applied across the board, so that trouble which is hard to see early can still surface at a predictable point, rather than depending on someone's hunch that something is wrong.
What the Diagnostic Is Designed to Do
The cargo-cult engagement diagnostic is a tool for identifying cargo-cult engagements by recognizing their pattern from the inside.
The phrase from the inside is the key to the whole tool. An outside reviewer may have the benefit of distance, but the diagnostic is written for the person in the middle of the engagement. That person needs a way to check their own situation without waiting for a retrospective to reveal the problem. The diagnostic is therefore a self-check that runs while the engagement is still in progress, not a post-mortem.
The diagnostic is also meant to help you notice trouble, which is different from fixing it. Its job is to surface the pattern early enough that action is still possible. The source frames it as a tool for noticing kinds of trouble while there is still time to act on them, and that framing sets the scope: recognition first, response second.
Why Five Questions Can Recognize a Pattern
Why does the diagnostic use five questions instead of a long audit? The source explains that the recognizable pattern of a cargo-cult engagement is consistent across the engagements its author has seen go wrong. Because the pattern repeats, it does not have to be rediscovered from scratch each time. It can be reduced to a short list of probes. The diagnostic compresses that consistent pattern into five questions, and the source attributes this compression to the author's experience with engagements that went wrong.
A useful way to think about the reasoning is this. If the same surface features keep appearing across failing engagements, a short set of questions aimed at those features can detect the pattern without requiring a full investigation every time. The questions act as probes. Each probe checks one part of the pattern, and together they cover the recognizable shape of the failure. The source does not spell out the individual features in this excerpt, so the claim to hold onto is the structural one: a repeated pattern can be compressed into a small set of probes.
A Worked Scenario at the Four-Week Mark
Tracing an engagement through the checkpoint
A small team has been working on an AI engagement for four weeks. Their status meetings are regular, their documents are complete, and no one has raised a concern. Walk through how the diagnostic applies to this engagement and what a reader should take from the result.
- 1Recognize the timing
The engagement has reached the four-week mark. The source says the five questions are run on every engagement at this point, so this team is due for the checkpoint whether or not anything looks wrong.
- 2Do not rely on the surface
Regular meetings and complete documents are exactly the kind of activity the source says can make an engagement feel productive while it is not. The diagnostic exists because this surface can mislead, so the team's apparent progress is not evidence that the checkpoint can be skipped.
- 3Run the five questions
The team answers the five questions as written in the full section of the source. Each answer is a probe for the recognizable pattern.
- 4Read the answers as a whole
The source says answers that come back wrong point toward the cargo-cult pattern and should be read as a whole rather than one question at a time. A single odd answer is not the verdict. The overall picture is.
The checkpoint happens because of the calendar, not because of suspicion. The result depends on the full set of answers read together, and any pattern that points toward the cargo-cult shape is a reason to look more closely while there is still time to act.
Passing Versus Wrong Answers
An engagement can pass the diagnostic or have its answers come back wrong. The source does not reproduce the composite illustration that shows what an engagement looks like when its answers come back wrong, so the contrast below is described only at the level the source supports. The key idea is that the verdict comes from the full set of answers, read together, rather than from any one answer on its own.
| Aspect | Engagement that passes | Engagement whose answers come back wrong |
|---|---|---|
| Question answers | Do not point toward the cargo-cult pattern | Point toward the cargo-cult pattern |
| How the set is read | As a whole | As a whole |
| What the result signals | No recognizable cargo-cult pattern in the answers | A recognizable pattern that deserves attention |
| What to do next | Continue the engagement | Look more closely while there is still time to act |
Both engagements are judged by the same five questions read together. The divergence is in what the answers point toward.
Notice what the two columns share. Both are read through the same five questions and the same whole-set reading. The divergence happens at the level of the answers, not at the level of the procedure. A passing engagement and a failing one go through an identical checkpoint, and only the answers differ.
Mistakes When Reading the Diagnostic
Common mistake Running the diagnostic only when an engagement already looks suspicious
Why it failsThe source says the questions are run on every engagement at the four-week mark. The diagnostic is a routine checkpoint, not a reaction to suspicion.
Do this insteadSchedule the checkpoint for every engagement at the four-week mark, whether or not anything seems wrong.
Common mistake Treating one wrong answer as the verdict
Why it failsThe source says answers that come back wrong should be read as a whole rather than one question at a time.
Do this insteadLook at the full set of answers together and judge the overall pattern they point toward.
Common mistake Assuming productive activity means a healthy engagement
Why it failsMeetings, documents, and artifacts can continue while the engagement is failing, so from the inside it can feel productive even when it is not.
Do this insteadUse the checkpoint to check the engagement directly rather than inferring health from visible activity.
Common mistake Waiting for a retrospective to reveal the problem
Why it failsThe diagnostic is written for the person in the middle of the engagement, so that trouble can be noticed while there is still time to act on it.
Do this insteadRun the checkpoint during the engagement, not after it ends.
Common mistake Reconstructing the five questions from memory
Why it failsThe individual questions are not written out in the source excerpt available for this article, so a remembered version may be inaccurate.
Do this insteadTake the wording of each question from the full section of the source.
Limits of This Excerpt
Predicting the Checkpoint
What do you think happens?
An engagement looks healthy and its team sees no reason for concern. Should the five questions be run on it, and if so, when?
Reveal answer
Yes. The five questions are run on every engagement at the four-week mark, regardless of whether it looks suspicious.
The source stresses that the word every matters. The diagnostic is a routine checkpoint, so a healthy-looking engagement still reaches the four-week mark and still gets the check.
Consider an engagement whose four-week checkpoint has just been completed. Four of the five answers come back looking fine, and one answer points toward the cargo-cult pattern. Using only what this article has taught, explain why the team should not dismiss the result simply because most answers passed, and describe what reading the answers as a whole would involve.
Hints
- Recall what the source says about reading answers as a whole rather than one question at a time.
- Think about whether a single answer is the verdict or one part of the overall picture.
Summary of the Diagnostic
- 1The cargo-cult engagement diagnostic is a tool for identifying cargo-cult engagements by recognizing their pattern from the inside.
- 2The five questions are run on every engagement at the four-week mark, making the check routine rather than triggered by suspicion.
- 3The five questions compress a consistent, recognizable pattern, so a short set of probes can detect it.
- 4Answers that come back wrong point toward the cargo-cult pattern and should be read as a whole, not one question at a time.
- 5The individual questions are not reproduced in this excerpt, so their wording should be taken from the full source section.
At a glance
- The cargo-cult engagement diagnostic is a tool for identifying cargo-cult engagements by recognizing their pattern from the inside.
- The five questions are run on every engagement at the four-week mark, so the check is routine rather than triggered by suspicion.
- A short set of questions works because the recognizable pattern of failing engagements is consistent and can be compressed into probes.
- Answers that come back wrong point toward the cargo-cult pattern and should be read as a whole rather than one question at a time.
- The individual question wording is not in this excerpt, so consult the full source section for it.