How to Do a Competitive Analysis

A competitive analysis is a decision tool, not a slide. Its job is to change what you do — how you position, price, build, or sell — based on an honest read of the field. If it doesn't change a decision, it was theater.

And most of it is theater, for three predictable reasons: it is self-serving (the dimensions are cherry-picked so your product looks best), anecdotal (claims with no source behind them), and stale (built once for a deck, never refreshed). The fix for all three is the same discipline: compare the dimensions that matter to the decision, tie every claim to evidence, and be honest about where rivals genuinely win.

What Is a Competitive Analysis?

It is a structured, evidence-based comparison of your offering against the alternatives a buyer is actually weighing — including the alternative of doing nothing. Two principles separate a useful one from a flattering one:

• Honesty over advocacy. The fastest way to make a competitive analysis worthless is to use it to reassure yourself. Don't trash-talk, don't exaggerate differences, and don't claim superiority without evidence. Acknowledge where a competitor is genuinely strong — that is the part that actually informs strategy.
• Evidence over assertion. Every comparison row, every market-share figure, every "they can't do X" should trace to a source. A competitive claim a stakeholder can't verify is a liability in the room, not an asset.

What Dimensions Should a Competitive Analysis Compare?

A competitive analysis is only as good as the dimensions you choose — and the temptation is always to choose the ones you win. Pick the dimensions that matter to the buyer's decision, then fill an honest "us vs. them" view across:

• Positioning — how they describe themselves and who they say they're for.
• Products and services — core capabilities, and the gaps.
• Pricing and packaging — model, tiers, and what's bundled vs. charged.
• Target customers — the segments they actually serve well.
• Go-to-market — how they reach and sell to buyers (channel, motion, distribution).
• Notable strengths — where they are genuinely better. Name them.
• Weaknesses and gaps — where they fall short, with evidence, not hope.
• Momentum — growth, funding, hiring, release cadence, market share where you can source it.

The honest version of this is a three-column table — Dimension · Us · Competitor — where the "Us" column is allowed to lose rows. If it never loses a row, you're writing marketing, not analysis.

What Are the Types of Competitive Analysis?

"Competitive analysis" is not one document. Match the type to the decision you're feeding.

Competitor Snapshot

A tight, side-by-side read of you versus one competitor — positioning, products, pricing, target customers, strengths, and gaps — for sales enablement. Short (roughly 500–1,200 words), factual, no hype. The job is to help a rep position honestly, not to script trash-talk.

Competitive Landscape

The whole field at once: who the direct, indirect, and aspirational competitors are, and how they cluster. Use it when you're entering a market or repositioning, not when you're prepping a single deal.

Feature or Vendor Comparison Matrix

A structured matrix scoring options across capability dimensions — for buyers evaluating tools, or for you to map functional parity and gaps. Evidence-required by nature: every cell should be defensible.

Win/Loss Analysis

What actually decided recent deals, from the buyer's mouth. The most honest competitive input you have, because it's grounded in outcomes rather than opinions.

Market Deep Dive

Long-form strategic analysis (think 3,000–5,000 words) combining the competitive landscape with market sizing, segments, and trends — for board, strategy, and investor audiences who need the full picture and the citations behind it.

The Competitive Analysis Frameworks That Structure It

Frameworks are scaffolding, not the analysis itself. The useful ones:

• SWOT — strengths, weaknesses, opportunities, threats. Best as an honest internal mirror; useless if the weaknesses column is empty.
• Positioning maps — plot competitors on the two axes buyers actually care about to find white space and crowding.
• Feature comparison matrices — structured parity-and-gap analysis across capabilities, priced and sourced.
• Strategic move analysis — not just where rivals are, but where they're heading, and what your response options are.

Which Competitor Analysis Framework Should You Use?

There is no single competitor analysis framework — there is a framework per question, and picking the wrong one is how a week disappears into a grid nobody acts on. Start from the decision, then choose:

• "Is this market worth being in at all?" → Porter's Five Forces. Rivalry, new entrants, substitutes, supplier power, buyer power. It analyses the structure of a market rather than any one rival, which is why it is useless for prepping a specific deal and invaluable before you commit a roadmap to a category.
• "Where are we genuinely weak?" → SWOT. An internal mirror, not a competitor profile. Fill the weaknesses and threats columns first; a SWOT that starts with strengths ends as a morale exercise.
• "Where is the white space?" → Perceptual or positioning map. Two axes the buyer actually prices on, every competitor plotted. The value is the empty quadrant — and the discipline of proving it is empty because nobody wants it, not because nobody thought of it.
• "Can we win this evaluation?" → Feature or vendor comparison matrix. Capability rows, one column per option, a source per cell. This is the framework a buyer is running on you, so run it on yourself first.
• "What will they do next?" → Strategic-move analysis. Read hiring, pricing changes, release cadence, funding and filings as intent, and write down your response options before you need them.

Two rules apply to all of them. A framework organises evidence; it does not produce any — a Five Forces diagram built on assumptions is five assumptions in a nicer shape. And you rarely need more than two: one to frame the market, one to frame the deal. Stacking four frameworks on the same thin evidence base is the most common way a competitor analysis gets long without getting better.

A Repeatable Process

1. Define the set. List direct, indirect, and aspirational competitors. Most weak analyses ignore the indirect ones — including "do nothing" — which is often the real competitor.
2. Pick the dimensions that matter to the decision. Not the ones you win. Tie them to what the buyer or the strategy call actually hinges on.
3. Gather evidence and cite it. Pricing pages, docs, filings, reviews, win/loss notes. Record the source for every claim as you go.
4. Verify the numbers. Market share, revenue, growth — a single wrong figure can sink an otherwise sound argument. Check each against its source before it ships.
5. Assess honestly. Fill the "us vs. them" view and let yourself lose rows. Name where they're genuinely stronger.
6. Synthesize into a position. Turn the comparison into a "so what" — where you win, where you're exposed, and the white space.
7. Tie it to a decision and refresh it. End with what changes. Then save the setup and re-run it when the market moves — competitive analysis is a quarterly habit, not a one-off deck.

Why Competitive Analyses Fail

• Cherry-picked dimensions. Every row is chosen so you win. The analysis reassures and informs nothing.
• No sources. Confident claims with nothing behind them — vibes dressed as intelligence. They collapse the moment a stakeholder pushes.
• Unverified numbers. A wrong market-share figure undermines the entire strategic argument it supports.
• Ignoring indirect competitors. Especially the status quo and "do nothing," which win more deals than any named rival.
• Trash-talk instead of assessment. Disparaging competitors signals insecurity and hides the strengths you actually need to plan around.
• One and done. A landscape built for one deck and never refreshed is wrong within a quarter.
• No decision attached. If the analysis doesn't end in "therefore we will…", it was theater.

The Test That Separates Signal From Spin: Evidence-Bound

There's one test that catches almost every failure mode above at once: can you trace every claim to a source, and is every number checked against one? If yes, the analysis survives the room. If no, it's spin — however confident it sounds.

This is exactly where AI-generated competitive analysis tends to fail: it produces fluent, plausible comparisons with invented figures and unsourced claims. The fix isn't to avoid the tooling — it's to insist the tooling be evidence-bound: pulling only from real source material, citing inline, checking numbers against sources, and showing you what it could not match instead of smoothing it over.

Do It Faster — Sourced, Not Vibes

Business can prepare a competitive-analysis draft from the landscape and source material you supply. It attempts source-number reconciliation and lists partial generated-draft Evidence findings for review, with model-generated citation markers where supported. Original figures can remain and references can be missing. Inspect the dates, definitions, competitor strengths and proposed interpretation before using the analysis.

What the Evidence tab can and cannot tell you: For a brief written from sources you upload or from the search results web research drew on, the Evidence tab under Checks in the brief editor lists up to five sentences or table cells whose wording or figures it could not match to those sources, for you to check. It matches wording and exact figures, not meaning: a faithful paraphrase, or $8.1M where the source says 8,100,000, can fail to match while being true, so an unmatched sentence is a place to look, not an error, and a matched one is not confirmed either, because this is a comparison with your own sources, not a fact-check. It reads only part of a long source (24 passages of up to 1,600 characters) and at most 12 sources, files first. Of web research it reads the title and snippet of each search result, not the page, so a figure taken from the page alone does not match, and if it could not read every result the list is advisory. It reads a table cell with its row and column, and a total the table works out from its own cells passes when the arithmetic holds. It checks the draft as generated, not your later edits.