SWOT Analysis

💪Strengths (S) #

Internal, controllable, the basis of an edge

A strength is not 'what we are good at' but 'what makes customers pick us over the competition'. A strength with no reference point is just an adjective, so attach a comparison to every item: ahead of whom, and by how much. The features the team is most attached to are the ones that slide into the strengths box uncritically, and this is where you filter them out.

Is it really a strength? #

Could a competitor copy it within six months

Ask three questions in order. Does it connect directly to why customers pay? Could a competitor replicate it quickly? If it disappeared, would revenue actually wobble? Anything that fails the test is not a strength, just an asset you happen to hold. Marking copyable strengths separately as 'current lead' keeps overconfidence in check.

Common misread #

Mistaking pride, size, or tenure for strengths

Items like 'ten years of experience' or 'first in the industry' are not strengths unless they connect to a customer's choice. Collecting strengths through internal surveys produces a list of pride; working backwards from the real reasons behind won and lost deals is far more accurate. Revenue scale is a result of strengths, not a strength itself, and mixing the two is a frequent error.

🩹Weaknesses (W) #

Internal, controllable, what loses you deals

A weakness is not every shortcoming, only the flaws that actually cost you deals and customers. Work backwards from churned-customer interviews, the causes of lost deals, and recurring complaints, and the list comes out short and sharp. Structural weaknesses you cannot fix, capital or geography for instance, are classified not as improvement projects but as inputs to an avoidance strategy.

Is it really a weakness? #

Does it connect to churn or a lost deal

Start by asking whether you can name even one customer or deal lost because of this flaw. If you cannot, it is probably not a weakness but a team gripe or perfectionism. Next, check whether competitors fail at it too; an industry-wide limitation belongs in threats, or is simply the environment, not your weakness box.

Common misread #

Fatal and ignorable weaknesses on the same line

Once the weakness box passes ten lines, nothing gets fixed. Separate the top two or three reasons customers actually left from the rest, and boldly mark the rest as 'accepted'. Conversely, the flaws the team has grown numb to tend to drop off the list; the first-30-days feedback of a new hire or a new customer brings them back into view.

🌊Opportunities (O) #

Outside changes we can ride

An opportunity is not my plan; it is a change happening outside. Only currents I did not create but that flow in my favor belong here: a regulation change, a competitor's withdrawal, a shift in customer behavior. Attach 'valid until when' to each item and the priorities set themselves, because most opportunities have expiry dates.

Is it really an opportunity? #

A market shift, or my wishful thinking

First ask whether this change is happening even without me. If the sentence only holds when my action is part of it, that is a plan, not an opportunity. Then cross-check whether the strength needed to seize it actually exists in the left column; an opportunity you have no means to grab is someone else's opportunity.

Common misread #

Writing 'launch new product' in the O box

The moment your own roadmap items enter the opportunity box, the SWOT becomes a self-introduction. 'Expanding into Asia' is a plan; 'competitor A withdrawing from Asia' is an opportunity. One more: a big addressable market (TAM) is not by itself an opportunity; you need evidence of change showing that market is moving toward you.

⛈️Threats (T) #

Outside changes that work against us

Threats are external too; internal risks like key people leaving go to the weaknesses box. You cannot prepare for every threat, so score them on two axes, probability and damage, and keep only the top ones. A good threat entry also records the early signal: what observation would mean it has become real.

Is it really a threat? #

Probability times damage, plus an early signal

Estimate in numbers what gets hurt if it happens: revenue, cost, or assets, and by roughly how much. An item you cannot even estimate is vague anxiety, not a manageable threat. For the top threats, define trigger indicators, written as observable events like 'competitor cuts prices by 20 percent', so the response is not late.

Common misread #

Listing the competitor's existence as the threat

An entry like 'intensifying competition' is always true, so it changes no decision. A threat must be written as a specific actor and a specific change; the threat is not the competitor but the competitor's particular move. Meanwhile substitutes and shifts in customer behavior are invisible and routinely omitted, yet what kills a market is usually these, not the same-category rival.

🧩Crossover strategy (TOWS) #

Strategy comes from multiplying the boxes

The output of a SWOT is not four lists but the products of the boxes. Sweep the four combinations: SO (attack opportunities with strengths), WO (enter opportunities by patching weaknesses), ST (defend against threats with strengths), WT (retreat from or avoid where weaknesses and threats overlap). In practice the fastest route is to fill the two extremes, SO and WT, first; once you know where to attack and what to abandon, the rest is adjustment.

SO strategy #

Ride the biggest wave with your strongest hand

Pair your most proven strength with the opportunity whose window is closing soonest, and pick only one or two offensive plays. More than three SO strategies and resources splinter until everything is mediocre. Each SO strategy is complete only as one sentence, 'based on this strength, before this window closes, we do this', and only then can it move into an execution plan.

WT strategy #

Do not fight where weakness meets threat

The zone where a threat lands squarely on a weakness is your first defensive priority. The options here are three: reinforce, partner, or withdraw, and choosing 'reinforce' out of pride every time is the most common failure. The moment withdrawal is removed from the menu, WT analysis becomes a formality. If there is no overlap at all, that itself is a signal of a healthy portfolio.

📏Operating rules #

Three to five per box, one line of evidence, an expiry date

Forcing three to five items per box ends the priority debate at the writing stage. Every item gets one line of supporting data (a number, a case, a source), and anything without evidence is quarantined with a 'hypothesis' tag. A SWOT is a snapshot, so stamp it with the date written and the next revision date; the most dangerous decision is one made on a SWOT six months stale.