How to do a competitor teardown (with a free template)

Most competitor analysis dies in a spreadsheet. Someone records logos, feature lists, social followers, and prices; the document looks substantial but never changes a decision. A competitor teardown is narrower. It asks: how does each alternative win a buyer, where does its promise become weak, and what can we do differently?

You can complete a first pass in two focused hours. Use public evidence, label your interpretations, and end with one or two actions. The process below works for software, professional services, local businesses, and many direct-to-consumer offers.

1. Write the decision before the competitor list

Start with one sentence: “We are doing this teardown to decide whether to ___.” Examples include narrowing your target customer, setting an entry price, rewriting the homepage promise, or choosing which feature to build next. If you cannot finish that sentence, the research scope is not ready.

Then define what counts as a competitor. Include three types:

• Direct: sells a similar outcome to the same customer.
• Indirect: solves the same job differently, such as a consultant instead of software.
• Status quo: the spreadsheet, manual process, internal hire, or “do nothing” option that absorbs the budget today.

Five to eight alternatives is usually enough for pattern recognition without creating a research swamp. If the market is crowded, choose representatives: the category leader, a low-price option, a specialist, a newer challenger, and the most common non-purchase alternative.

2. Capture evidence, not impressions

Create one row per claim and save the page URL, the exact wording, and the date you saw it. Screenshots help when pages change. Separate observation (“the homepage says setup takes ten minutes”) from interpretation (“they are targeting buyers anxious about implementation”). That separation keeps a confident guess from turning into a fake fact.

For each competitor, inspect the homepage, pricing page, key product or service page, case studies, FAQ, onboarding or booking flow, and a small sample of recent public reviews. Do not scrape hundreds of reviews. Read enough positive, mixed, and negative examples to identify repeated language, then record the actual excerpts or themes that support your conclusion.

3. Compare the parts that affect a purchase

Feature grids are useful only when a feature changes the buyer’s choice. Focus first on the buying system:

• Audience: Who is explicitly addressed? What stage, size, industry, or situation?
• Promise: What outcome leads the page? How fast, easy, safe, or complete is it claimed to be?
• Mechanism: How do they explain why their approach works?
• Offer: What is included, excluded, packaged, guaranteed, or limited?
• Price and commitment: Entry point, contract, trial, setup fee, minimum, and cancellation terms.
• Proof: Named customers, quantified cases, demonstrations, reviews, credentials, or only generic assertions.
• Friction: What must a buyer do before seeing value—book a call, migrate data, wait for a quote, or learn a complex tool?
• Objections: Which concerns does the FAQ answer? Which concerns appear in reviews but remain unanswered?

4. Use this free competitor teardown template

If several teammates are scoring the same evidence, use a simple 1–3 scale rather than pretend precision: 1 means weak or absent, 2 means adequate, and 3 means unusually strong. Score only dimensions relevant to your original decision.

5. Turn patterns into a decision

Read across competitors, not down them. Repeated claims reveal category conventions: perhaps everyone promises “one platform,” “personal service,” or “AI-powered efficiency.” These phrases may be necessary for category recognition, but repetition makes them poor differentiators.

Look for a gap at the intersection of three tests: buyers appear to care, competitors handle it weakly, and you can deliver it credibly. A blank cell alone is not an opportunity. Nobody may offer a feature because nobody wants it. Validate a promising gap through customer conversations, sales-call notes, search behaviour, support questions, or a small offer test.

Finish with no more than three recommendations, each tied to evidence and an owner. For example: “Rewrite the homepage for operations managers at 20–100 person agencies because four leading tools speak to enterprise IT, while recent reviews repeatedly praise fast setup. Test the new page against the current one before changing the product.” That is more useful than “differentiate on ease of use.”

Common teardown mistakes

• Copying the leader: their choices fit their brand, cost structure, and customer base—not necessarily yours.
• Treating positioning as reality: “instant,” “premium,” and “complete” are claims until evidence supports them.
• Ignoring the status quo: your hardest competitor may be a familiar manual process with no new budget approval.
• Using stale evidence: record dates and revisit the few facts your decision depends on.
• Ending without a test: a teardown should produce an action, a person responsible, and a way to learn.