Walk into almost any strategy offsite and you will meet at least one of these four frameworks on a flip chart. They are the entry-level tools of strategic analysis, and that ubiquity is exactly the problem. A framework that everyone can fill in is also a framework that almost nobody finishes: the grid gets populated, photographed, and never converted into a choice. The fix is not a better grid. It is knowing what question each framework was built to answer, and treating the output as the start of a decision rather than the decision itself.

The quick version

  • SWOT sorts what you know into four boxes (Strengths, Weaknesses, Opportunities, Threats). It is a checklist, not an analysis, its value is entirely in what you do next.
  • PESTLE scans the outside world (Political, Economic, Social, Technological, Legal, Environmental) so your plan survives contact with reality, not just your competitors.
  • Ansoff maps four growth routes by how much is new, existing vs new products, existing vs new markets, and ranks them by risk.
  • BCG looks across a portfolio of products or units and asks where to invest, hold, milk, or exit, using market growth and relative share.

The idea in depth

Group the four by the question they answer and they stop competing and start sequencing. Two look inward and outward today (SWOT, PESTLE: where are we?). Two look forward (Ansoff, BCG: where should we put our money?). Run them in that order and the second pair inherits evidence from the first.

flowchart LR
  A(["PESTLE
What's changing outside?"]) --> C(["SWOT
So where do we stand?"])
  B(["Internal scan
capabilities, results"]) --> C
  C --> D(["Ansoff
Which growth route?"])
  C --> E(["BCG
Where to back / exit?"])
  D --> F(["A choice, with
resources behind it"])
  E --> F
The four diagnostics in sequence: scan, position, then decide where the money goes. Leaders Loop

SWOT and PESTLE: the scan you mistake for the analysis

SWOT grew out of long-range-planning research at the Stanford Research Institute in the 1960s. The earliest form was an acronym, SOFT, that later got reworked into the familiar four-letter grid. Who deserves the credit is genuinely contested, Albert Humphrey's name is the one most often attached to it, but the scholarly reconstruction below points instead to Robert Stewart and his team, so treat any tidy origin story with a little suspicion. PESTLE has a cleaner pedigree: Harvard professor Francis Aguilar introduced environmental scanning in his 1967 book Scanning the Business Environment, using the cluster he labelled ETPS; over the following decades practitioners reordered it into PEST and bolted on Legal and Environmental to get PESTLE.

The two are partners. PESTLE feeds the outward-facing half of SWOT: the Opportunities and Threats in a serious SWOT should be the specific consequences of trends you found in the PESTLE scan, not a brainstorm. "Interest rates" is a PESTLE factor; "our refinancing window closes in March and rates are still climbing" is the threat it produces.

Here is the honest limitation, and it is well documented. Terry Hill and Roy Westbrook studied SWOT use across companies and concluded, in Long Range Planning (1997), that it was "time for a product recall." Their finding: teams produced long, unprioritised lists, over forty factors on average, in vague three- or four-word phrases that were never verified and, most damningly, never used in any later stage of the strategy process. The grid felt like work and substituted for it.

A SWOT that nobody prioritises is a list, not a strategy. The four boxes are the easy part; the ranking is the work.

So the move is: after the grid is full, force a conversion step. Pair items across quadrants, this is the logic of Heinz Weihrich's 1982 TOWS matrix, and ask the only question that matters: which strength can we aim at which opportunity, and which weakness does which threat expose? Then cut the list to the three pairings you will actually act on this quarter. If a factor doesn't survive that cut, it was decoration.

Ansoff: name the risk before you take it

Where SWOT and PESTLE describe the present, the Ansoff matrix points at the future and prices it. H. Igor Ansoff, a mathematician turned planner at Lockheed, laid it out in Strategies for Diversification in the Harvard Business Review in 1957. The grid crosses two axes, products and markets, each either existing or new, and the four cells are deliberately ordered by how much novelty (and therefore risk) each carries.

flowchart TB
  subgraph EM["EXISTING markets"]
    A(["Market penetration
sell more of what you have
· lowest risk"])
    B(["Product development
new products, known customers"])
  end
  subgraph NM["NEW markets"]
    C(["Market development
known products, new customers"])
    D(["Diversification
new product + new market
· highest risk"])
  end
  A --> B --> C --> D
The Ansoff matrix, read as a risk gradient from market penetration to diversification. Leaders Loop

Penetration leans on what you already know on both axes, so it is the safest bet. Diversification, a new product and a new market at once, asks you to learn two unfamiliar things simultaneously, which is why Ansoff flagged it as the riskiest of the four. The framework's quiet gift is that it makes a growth conversation honest: a board that wants "ambitious growth" usually means penetration; a founder who says "let's just expand into the US with a new line" is proposing diversification and should say so out loud.

The limitation: Ansoff measures risk by novelty, not by capability. A diversification move that sits next to your existing strengths can be safer than a "penetration" play in a market that has turned against you. Read the grid alongside your SWOT, not instead of it.

The move, then: plot your top three growth ideas on the matrix and say the risk word out loud for each. Then ask what evidence would retire the biggest unknown cheaply, a pilot, a pre-sale, a single test market, before you commit the budget. The matrix tells you where the risk lives; a small experiment tells you whether it is real.

BCG: a portfolio is a set of different jobs

The Boston Consulting Group's growth-share matrix, sketched by founder Bruce Henderson and set out in his 1970 essay The Product Portfolio, answers a different question again: not "should we grow?" but "across everything we run, where does the cash come from and where should it go?" It plots each product or unit on market growth (is the tide rising?) against relative market share (are we winning?), producing four well-worn labels.

quadrantChart
  title Market growth vs relative share
  x-axis "Low share" --> "High share"
  y-axis "Low growth" --> "High growth"
  quadrant-1 "Stars, invest"
  quadrant-2 "Question marks, pick"
  quadrant-3 "Dogs, exit / hold"
  quadrant-4 "Cash cows, milk"
The BCG growth-share matrix: stars fund question marks, cash cows fund stars, dogs fund the conversation about leaving. Leaders Loop

The insight that made it famous is that these are not grades, they are jobs. Cash cows (high share, low growth) are milked to fund the future; stars (high share, high growth) are fed to keep their lead; question marks (low share, high growth) demand a decision, back hard or let go; dogs (low share, low growth) are candidates to harvest or exit. The point is that a healthy company deliberately runs all four, with the cows quietly paying for the stars.

BCG's own retrospective is admirably candid about where this breaks. Writing in BCG Classics Revisited, the firm notes that since 1970 "conglomerates have become less prevalent, change has accelerated, and competitive advantage has become less durable", so today's question marks and cash cows can swap places faster than an annual planning cycle can track. Relative market share is also a thin proxy for advantage; a "dog" can be a profitable niche, and a high-share business in a collapsing market is not a cow but a liability.

So the move is: use the four boxes to expose your hidden cross-subsidies, then pressure-test each label. For every cash cow, ask how many quarters of milk are left before the market moves. For every dog, separate "small but profitable and self-funding" from "small and bleeding." The matrix is a prompt for that conversation, not a verdict that ends it.

A worked example

Take a fictional mid-size coffee-roasting business, call it Northbeam, to see the four run in sequence. (All figures below are illustrative, chosen to show the method, not real company data.)

PESTLE surfaces three live trends: a new packaging-waste regulation (Legal/Environmental), squeezed household budgets (Economic), and a fast-growing at-home subscription habit (Social/Technological). SWOT turns those into specifics: a strength is a loyal wholesale base of cafés; a weakness is near-zero direct-to-consumer capability; the subscription trend is the opportunity; the budget squeeze plus the packaging rules are the threats. Crucially, the team prioritises, they pick one pairing: use the strong roasting reputation (S) to chase the subscription opportunity (O).

That pairing lands on the Ansoff grid as market development with a dash of product development, known coffee, sold to new direct customers in a new channel, riskier than selling more to existing cafés, but well short of diversification. So they de-risk it: a three-month pilot to 500 existing café customers' staff before any national launch. Finally, BCG reframes the whole portfolio: wholesale is the cash cow (high share of their regional café trade, low growth) that should fund the subscription line, which is a question mark, high growth, no share yet, that the pilot will either promote toward "star" or quietly retire. One coherent story, four lenses, one funded choice.

Frequently asked questions

Which framework should I start with?

Start outside-in: PESTLE, then SWOT. The external scan stops your SWOT's opportunities and threats from being a wish-list, and a grounded SWOT then gives Ansoff and BCG real evidence to chew on. Forward-looking tools built on a weak present-day read just multiply the error.

Are these frameworks outdated?

The tools are old; the questions are not. Where they get a bad name is in misuse, treating a filled-in grid as a finished analysis. Hill and Westbrook's 1997 critique of SWOT was about exactly that failure, not the four boxes themselves. Used as conversation-starters with a forced prioritisation step, they still earn their place.

What's the difference between SWOT and PESTLE?

PESTLE is purely external and macro, forces no single company controls. SWOT mixes internal (strengths, weaknesses) with external (opportunities, threats). The clean way to use them together is to let PESTLE generate the external half of the SWOT, so the two never contradict each other.

Ansoff or BCG, when do I use which?

Ansoff is for a single growth decision: we have an idea, how risky is it? BCG is for a portfolio decision: we run several things, where should the money flow? If you have one product and a choice to make, reach for Ansoff. If you have many and a budget to allocate, reach for BCG.

How do I stop these turning into a box-filling ritual?

Add a hard conversion step every time: after the grid, the team must name the three things they will act on and the things they are choosing to ignore. A framework that doesn't end in "so we will do X and not Y" hasn't been used, it's been performed.

Related in the Toolkit

These diagnostics are the on-ramp; the modules below are where the analysis turns into durable advantage and a funded plan.

Where to go next