A consultant who runs a generic AI prompt to produce a SWOT analysis for a client engagement is producing the same artifact that every other consultant running the same prompt is producing. The framework isn't the problem — SWOT, Porter's Five Forces, and the various 2x2 matrices have lived in strategy consulting for decades because, used well, they organize messy information into actionable structure. The problem is that AI's default mode is to produce the template, not the analysis, and the template is what's lost the framework its credibility.

The fix is not to retire the framework. It's to use AI for the parts of the framework where it adds value (the structuring, the gap-finding, the formatting) and to keep the parts where the consultant's judgment is the product (the priors, the specifics, the conviction) firmly with the human.

This guide is the four moves that get AI-assisted strategic analysis to the level a client actually pays for. The Consultant Strategy Memo Generator supports each of them.

Key takeaways

Generic AI strategic frameworks produce template-filling, not analysis. The framework isn't the problem — SWOT, Porter's Five Forces, and 2×2 matrices remain useful. The problem is the absent priors, missing comparables, and papered-over tensions.
Four moves to make AI-assisted analysis useful: supply industry-specific priors before populating the framework, force named comparables, ask for internal contradictions explicitly, pull the headline into the memo (don't present the framework itself).
Named comparables require verification. AI confidently produces references that are misremembered, misattributed, or fabricated. The consultant verifies every named precedent before it touches a client deliverable.
Tensions are where the interesting analysis lives. A strength that's also a weakness, an opportunity that's also a threat — naming the tension forces the strategic question the client needs to answer.
The "specific enough that it couldn't have been written about a different client" test separates real analysis from filler. Any AI output that fails this test is filler.

Why generic AI strategic frameworks read as filler

Three patterns dominate the output of generic prompts:

The bullets are at the same level of abstraction the prompt was at. Ask for a SWOT of "Acme Corp's new product launch" and you'll get four quadrants of bullets that say "strong brand recognition," "limited budget," "growing market," "competitive threats." None of those bullets is wrong. None of them is useful, because they're at the level of generic business-school examples.
There are no named comparables. A good strategic analysis is full of named precedents: this strategy is being attempted; the closest comparable is X; that comparable resolved by year three with outcome Y. Generic prompts produce frameworks with no named comparables, which signals to the reader that no actual research happened.
There is no internal contradiction. Real strategic situations have tensions — a strength that's also a weakness, an opportunity that's also a threat in the same dimension. Generic frameworks paper over the contradictions. The contradictions are where the interesting analysis lives.

The four moves below fix each of these patterns in sequence.

Move 1 — Supply industry-specific priors before asking AI to populate the framework

Generic prompt: "Give me a SWOT analysis for our client, a regional health system entering value-based contracts."

Generic output: predictable bullets about brand, market, reimbursement risk, change management.

The fix: load the prompt with the priors before asking for the framework. The priors are the consultant's accumulated knowledge of the specific industry and the specific situation:

"Our client is a regional health system in the U.S. Midwest with strong existing fee-for-service revenue, a moderately sophisticated revenue cycle, and a payer mix dominated by two commercial payers and Medicare. They are evaluating a transition to value-based contracts, specifically considering a Medicare Shared Savings Program ACO and a commercial-payer shared-savings arrangement. Comparable transitions in similar markets have typically taken 24-36 months from contract execution to first reported savings, and the savings have averaged in the low-to-mid single digits as a percentage of attributed-population spend. The principal organizational risk in comparable transitions has been the leakage of attributed patients to out-of-network specialists. The principal financial risk has been overestimating the speed at which fee-for-service revenue can be replaced. Generate a SWOT analysis grounded in this specific situation."

The output of this prompt looks completely different. The bullets reference specific dynamics — attributed-patient leakage, fee-for-service replacement speed, payer mix concentration — that the generic prompt would never have surfaced.

The consultant is doing the work of supplying the priors. AI is doing the work of organizing them into the framework's quadrants. That's the division of labor that works.

Move 2 — Force named comparables

A SWOT (or any framework) that doesn't reference specific named comparable situations is hollow. The fix is to require them in the prompt.

"For each of the four quadrants, name at least one comparable situation — a specific named organization or a clearly described case where this dynamic played out — and describe what happened. If you cannot supply a comparable for a given bullet, omit the bullet rather than including it without one."

The "omit if no comparable" instruction is the discipline. It prevents AI from filling the framework with bullets it can't substantiate. The output will be shorter. Shorter is the goal.

A note on accuracy: AI's named-comparable output is a starting point, not a fact. The consultant verifies every comparable. AI can confidently produce a reference to a situation that doesn't exactly match the facts, or that the consultant should not be citing without independent verification. The consultant's review is the load-bearing step.

Move 3 — Ask for the internal contradictions explicitly

After the framework is populated, run a second prompt focused entirely on tensions:

"Looking at the SWOT you just produced, identify three internal contradictions or tensions — places where one quadrant's content directly complicates another quadrant's content, or where a stated strength is, viewed from another angle, a weakness. For each tension, name the dimension on which the items conflict, and propose the strategic question that the tension forces."

The output of this prompt is often where the analysis becomes interesting. A "strong existing fee-for-service revenue" that's also "an obstacle to value-based transition" is a tension worth naming explicitly. A "growing addressable population" that's also "operating at near-capacity" is a tension. The consultant who surfaces and names the tensions produces analysis that reads as serious.

This is also the move where the framework stops being a template and starts being thinking. Templates have neat quadrants. Thinking has tensions.

Move 4 — Pull the headline from the framework, not the framework into the memo

The final move is structural. A common failure mode is to present the entire framework to the client and let them draw conclusions from it. That's the consultant's job, not the client's.

Instead: use the framework as a backstage analytical tool, and present the headline conclusion in the client-facing memo. The framework can be an appendix. The memo opens with the synthesis.

"Based on the SWOT you populated and the tensions you surfaced, write a three-sentence headline statement that captures the central strategic insight, expressed as: (1) the situation in one sentence, (2) the implication in one sentence, (3) the recommended posture in one sentence. The sentences should be specific enough that they could not have been written about a different client in the same industry."

The "specific enough that they could not have been written about a different client" check is the test. A headline that could apply to any other regional health system entering value-based contracts is generic. A headline that names dynamics specific to this client is the consultant's contribution.

What this approach is not

A few honest limits:

The frameworks themselves are not magic. SWOT, Five Forces, and similar tools are organizing structures. They don't generate insight; they organize the insight the analyst already has. A SWOT done well surfaces what the analyst already knows but hadn't articulated. A SWOT done badly fills boxes.
The named comparables require verification. AI can produce comparables that sound right but aren't accurate. The consultant verifies. This is non-negotiable. A client memo that cites an incorrect comparable damages the consultant's credibility for years.
The discipline isn't about the prompt. It's about the underlying thinking the prompt represents. The priors the consultant supplies, the comparables they verify, the tensions they surface — those are the work. AI accelerates the writing layer around the work.

A practical adoption pattern

For a consultant rolling this into their workflow:

First engagement — use the four moves explicitly on a single framework deliverable. Compare to the version you would have produced without them.
Subsequent engagements — make the four moves a checklist on every framework-based analysis. The discipline becomes habit by engagement three or four.
Senior peer review — for engagements where the stakes warrant, have a senior peer pressure-test the priors and the named comparables before the memo is sent. The peer review is where errors get caught; AI doesn't do peer review.

What AI does, and what it doesn't

The Consultant Strategy Memo Generator handles the writing layer once the four moves have been worked through: producing the populated framework, drafting the tensions section, writing the headline statement, and formatting the appendix.

AI does not:

Have the priors. The consultant supplies them.
Verify the comparables. The consultant verifies them.
Form the strategic view. The consultant forms it.
Replace senior judgment. Frameworks for high-stakes decisions warrant senior-consultant or partner review. AI doesn't do that review either.

How to start

Pick the next framework-based deliverable you have queued. Before opening any AI tool, write out the priors you're bringing to the situation — a paragraph, in your own words. Then run the four moves. Compare the output to the version you'd have produced from a generic prompt. The gap is the work.

The consultants who run this discipline produce frameworks clients keep on their desks. The consultants who don't, produce frameworks clients politely set aside.

Next steps

Consultant Strategy Memo Generator — for the framework population and synthesis.
How to Write a 1-Page Strategy Memo for a Skeptical CFO — for the memo format the framework feeds into.
The Independent Consultant's Pricing Memo Stack — for the engagement-conversation memos this analysis supports.
Management Consultant Claude Plugin install guide — to run the workflow from inside Claude.

Frequently asked questions

Why do AI-generated SWOT analyses sound generic?

Three reasons: the bullets are written at the same level of abstraction the prompt was at, so they read as business-school examples rather than client-specific analysis. There are no named comparables, so the framework reads as unsourced. And the internal contradictions (the tensions between quadrants) are papered over, which is exactly where the strategic insight lives. Each is fixable by loading the prompt with specific priors before populating the framework.

How do I make AI strategic analysis sound more credible?

Four moves: (1) Supply industry-specific priors before asking AI to populate the framework — load the prompt with the accumulated knowledge of the situation. (2) Force named comparables — require AI to cite a specific organization or case for each quadrant, with the instruction to omit any bullet it can't source. (3) Ask AI to surface the internal contradictions. (4) Pull the headline finding from the framework into the memo, rather than presenting the framework itself.

Is SWOT analysis still useful in 2026?

Yes, when used as an organizing structure rather than a template. SWOT, Porter's Five Forces, the BCG growth-share matrix, and other classic frameworks remain useful organizing tools that surface what the analyst already knows but hadn't articulated. They become useless when run as fill-in-the-quadrants exercises divorced from priors, comparables, and the analyst's strategic judgment.

Should I cite comparable companies in a strategic analysis?

Yes, with verification. Named comparables ("the closest parallel to this situation is X's transition in 2018") signal serious analysis. AI's named-comparable output is a starting point, not a fact — it can produce references that sound right but are wrong about specifics. Every named comparable in a client deliverable should be verified against an independent source before inclusion.

What is a strategic tension in a SWOT analysis?

An internal contradiction between elements of the framework: a stated strength that's also a weakness on the same dimension, an opportunity that's also a threat depending on the team's response. "Strong existing fee-for-service revenue" can be a strength on cash flow and a weakness as an obstacle to value-based transition. Naming the tension forces the strategic question the team needs to answer.

How do I use AI for strategy without making it sound like ChatGPT?

The division of labor: the consultant supplies industry priors, verifies comparables, identifies tensions, and forms the headline conclusion. AI organizes the inputs into the framework's structure and drafts the writing layer. The output passes the "specific enough that it couldn't have been written about a different client" test only if the consultant has done the underlying thinking — AI accelerates the production, not the analysis.

Should I show the SWOT framework to the client or just the conclusions?

Show the conclusions. The framework is a backstage analytical tool — a way for the consultant to organize the thinking. Presenting the populated framework to the client puts the burden of synthesis on them, which is the consultant's job. The headline finding, the implications, and the recommended posture belong in the memo. The framework can be an appendix.

This article is general guidance for management consultants. AI-generated strategic content, particularly named references to comparable situations, should be independently verified before inclusion in client deliverables.

Why Your AI-Generated SWOTs Sound Like Every Other Consultant's (Fix in 4 Moves)