Competitor research · 14 min read · Updated 2 Aug 2026
Competitor Teardown Template (Free Product Teardown)
A blank competitor teardown you can fill in today, plus the guidance for what belongs in each field. A teardown is primary research: you use the product and record what you observed with dates and screenshots, which is what separates it from a profile assembled out of their marketing.
Copy pastes straight into Google Sheets or Excel with the columns intact. Downloads are free with a work email.
The competitor teardown template
This is exactly what you get when you copy or download. Blank fields are yours to fill in; each table ships with one example row to show the pattern, which you delete.
Teardown scope
Fill this in first. One product per teardown, and one question you are trying to answer about it.
- Competitor and productThe specific product and the plan or tier examined
- Question this answersThe one thing you want to know, e.g. why buyers say setup is fast
- Version or date of the buildProducts change; record what you actually saw
- Who performed the teardownOne named person, with their role
- Time spentHonest figure, so the depth of the findings is readable
- Date performed and next teardownWhen it was done, and when it gets repeated
1. Access and method
First row is an example, delete it. Record how you actually got hands on it. This determines how much of the teardown is observation and how much is inference.
| Source of access | What it let us see | What it did not | Constraints or terms noted |
|---|---|---|---|
| ExamplePublic self-serve trial, 14 days | Signup, setup, core workflow, in-product upgrade prompts | Enterprise features, admin controls, integrations behind sales | Trial capped at 3 tracked competitors |
2. First-run experience
First row is an example, delete it. Time each step. This is the part buyers experience before anyone talks to them, and the part competitors optimise least.
| Step | Time taken | What happened | Friction or delight | Screenshot reference |
|---|---|---|---|---|
| ExampleSignup to first competitor added | 6 minutes | Email verification, then a four-field setup wizard | Delight, no sales call required | shot-01 |
3. Core workflow walkthrough
First row is an example, delete it. Trace the job the product exists to do from start to finish, the way a customer would.
| Step in the job | How their product does it | How ours does it | Who does it better, and why | Evidence |
|---|---|---|---|---|
| ExampleGetting a competitor change to a rep | Digest email, manual battlecard edit afterwards | Change pushed into the battlecard automatically | Us, theirs needs a human in the loop | shot-04, their docs |
4. Capability observations
First row is an example, delete it. Only what you saw with your own eyes. Anything you did not verify goes in the not-verified column, not in the observed one.
| Capability | Observed behaviour | Claimed but not verified | Depth | Screenshot reference |
|---|---|---|---|---|
| ExampleAutomated competitor alerts | Fires on any page change, including cosmetic edits | "AI-filtered for relevance" per their site | Shallow, no severity classification | shot-07 |
5. Where it is genuinely good
First row is an example, delete it. Three to five rows. A teardown that finds nothing impressive was not performed carefully.
| Strength | Why it works | Would our buyers notice? | Could we adopt it? |
|---|---|---|---|
| ExampleSetup requires no sales contact | Four fields, sensible defaults, no data import | Yes, evaluators short-list what they can try | Partly, our onboarding needs an import step |
6. Where it breaks down
First row is an example, delete it. What failed, at what point, and whether a real buyer would hit it. Reproduce it before you record it.
| What breaks | At what point | Would a real buyer hit it? | Reproduced? | Screenshot reference |
|---|---|---|---|---|
| ExampleAlert volume becomes unusable | Past roughly five tracked competitors | Yes, most teams pass five within a quarter | Yes, twice | shot-11 |
7. What it reveals about their strategy
First row is an example, delete it. The analysis layer. Product decisions are strategy made visible, and this is where a teardown outperforms any amount of marketing analysis.
| Observation | What it suggests about their priorities | Confidence | What we would expect next if true |
|---|---|---|---|
| ExampleEvery workflow ends in a battlecard | They are built for sales enablement, not product or strategy teams | High, consistent across the whole product | Deeper sales-tool integrations before analytics |
8. Decisions, owners and dates
First row is an example, delete it. Three to five rows. Split them: what sales says now, and what product does later.
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| ExampleTheir alerts have no severity classification | Add the noise question to discovery and a proof point to the battlecard | Priya, PMM | 30 Apr | Alert relevance appears in more competitive win reasons |
How to fill in your competitor teardown
How to scope a competitor teardown
A teardown is primary research: you use the product and record what you observed. That is the whole distinction from a competitor profile, which is assembled from what a competitor publishes about themselves. Scope it to one product and one question, because an open-ended teardown produces forty screenshots and no conclusion, which is the most common outcome for this exercise.
Competitor and product
Name the tier as well as the product. Trials usually expose the entry tier only, and a teardown that silently generalises from a trial to their enterprise offering will be contradicted by the first buyer who has seen both.
Question this answers
One question: "why do buyers say their setup is fast?" or "how do they actually get a competitor change to a rep?". A teardown with a question finishes in an afternoon with findings; one without becomes an open-ended exploration that gets abandoned.
Version or date of the build
Software changes weekly. A teardown without a date describes a product that may no longer exist, and this is the single most common reason teardown findings get challenged internally six months later.
Who performed the teardown
Named, with their role. Findings from a product manager and from a sales engineer differ in useful ways, and knowing who looked tells a reader which blind spots to expect.
Time spent
Record it honestly. Two hours produces a usable read on the first-run experience; two days is needed for depth. Stating it lets a reader calibrate how much weight the conclusions carry, which is more useful than pretending to a completeness you did not achieve.
Date performed and next teardown
Twice a year for competitors you meet often, plus after any significant launch. The comparison between two teardowns is where the trajectory shows up, and trajectory beats any single snapshot.
How to record access and method in a competitor teardown
This section is unusual and it belongs at the front, because everything downstream depends on it. How you got access determines what you were able to see, and being explicit about the boundary is what separates a credible teardown from a document that quietly presents inference as observation. It also keeps the exercise honest about what remains unknown.
Source of access
A public self-serve trial, a recorded demo, their documentation, a sandbox, a review-site video walkthrough, or a customer who uses both products and walked you through it. The last one is the most underused and frequently the best, because you get the product and the buyer's reasoning together.
What it let us see
The genuine scope of observation: "signup, setup, core workflow, in-product upgrade prompts". Be specific, since this is the list a reader uses to judge which findings are first-hand.
What it did not
Just as important. Trials typically hide enterprise features, admin controls and anything gated behind a sales conversation. Naming the gaps prevents the teardown from being read as comprehensive when it covered the entry tier.
Constraints or terms noted
Trial caps, usage limits and anything in their terms that restricts what you may do. Record what you noticed rather than interpreting it, and if their terms restrict competitor access, that is a fact for the team to consider rather than something to work around quietly.
Prefer public and permissioned routes
Public trials, published documentation, recorded demos, review-site walkthroughs and conversations with customers who use both are all legitimate and sufficient for nearly every question worth asking. Misrepresenting who you are to obtain access is a poor idea on its own terms and it also produces a document nobody will circulate. If a question genuinely requires access you cannot get honestly, answer it from win/loss interviews instead.
How to record the first-run experience in a competitor teardown
Time every step. The first run is the part of a competitor's product that buyers experience before anyone from their sales team speaks to them, and it is disproportionately what evaluators judge on. It is also the part most teams never examine, because they read the competitor's marketing and then compare feature lists. Two hours here produces more usable competitive insight than a week of reading their website.
Step
Break it into the real steps: signup, verification, setup, first meaningful action, first output. The step where a product asks for something the user cannot supply yet is almost always where evaluations quietly die.
Time taken
Actual minutes, measured rather than estimated. "Six minutes from signup to first competitor added" is a competitive fact you can use in a sales conversation. "Fast onboarding" is not. Time your own product through the same steps for a comparison that means something.
What happened
Plainly, in sequence. Include the things that are easy to overlook: a required credit card, a mandatory call booking, an email verification delay, a data import before anything works.
Friction or delight
Mark both, and be honest about the delight. A competitor doing something better than you is the most valuable finding a teardown can produce, and a teardown recording only friction is a document written to reassure rather than to inform.
Screenshot reference
Number every screenshot and reference it here. Products change without notice, and a finding with no image is unverifiable within a quarter. This is also what lets someone else check a claim before it reaches a battlecard.
How to walk through the core workflow in a competitor teardown
Follow the job their product exists to do, end to end, as a real user would rather than as a reviewer sampling features. The point of this section is not to catalogue what the product has but to understand the path it puts a user on, because that path is where their product philosophy is visible and where the genuine differences between two products live.
Step in the job
Frame steps as the user's job, not as the product's features: "getting a competitor change to a rep", not "the alerts module". Feature framing makes two products look similar; job framing makes the differences obvious.
How their product does it
Describe the actual path including the manual steps. The number of human interventions required is frequently the most consequential difference between two products and is invisible in any feature comparison.
How ours does it
Fill this in side by side, honestly, and run the same walkthrough on your own product rather than describing it from memory. Teams are consistently generous about their own workflows until they time them.
Who does it better, and why
A judgement with a reason. Expect them to win some rows, and record it plainly. A teardown where you win every row will not be believed by the people who most need to read it.
Evidence
Screenshot references and documentation links. This section produces the claims most likely to end up in sales conversations, so it needs the tightest sourcing in the document.
How to record capability observations in a competitor teardown
The discipline in this section is the separation between what you observed and what they claim. A teardown's entire value is that it is first-hand, and that value evaporates the moment observed behaviour and marketing claims are recorded in the same column. Keep them apart visibly, because the gap between the two is often the most interesting finding available.
Observed behaviour
What the product actually did, in operational detail: "fires on any page change, including cosmetic edits". This level of specificity is what makes a teardown quotable and what no competitor analysis assembled from websites can produce.
Claimed but not verified
Their stated capability, in quotation marks, marked clearly as unverified: "AI-filtered for relevance, per their site". Sometimes the claim is true and you simply could not reach it on a trial. Sometimes the gap between the claim and the behaviour is the headline finding. Recording both is the only way to tell.
Depth
Deep, adequate or shallow, with the reason: "shallow, no severity classification". Feature checklists mark presence, which is why two products can look identical on a grid and feel completely different in use. Depth is the column that captures the difference.
Screenshot reference
One per row where possible. Capability claims about competitors decay faster than anything else and are the ones most likely to be repeated to a buyer, which makes them the ones most in need of a dated image.
Do not fill gaps with assumptions
If you could not reach a feature, the entry is "not observed", not an inference from their documentation. An unmarked assumption in this section becomes a stated fact in a battlecard within two weeks and gets corrected by a buyer shortly after.
How to record genuine strengths in a competitor teardown
Three to five rows, and this section is a quality check on the whole exercise. A teardown that finds nothing impressive was not performed carefully, and everyone reading it knows that. More usefully, the strengths column is where the transferable ideas are: a teardown that produces one adoptable improvement to your own product has more than paid for the afternoon it took.
Strength
Specific and observed: "setup requires no sales contact". Vague admiration is not useful, and neither is a strength you inferred from their marketing rather than experienced.
Why it works
The mechanism, not the outcome: "four fields, sensible defaults, no data import". The mechanism is the part you can learn from; the outcome is just the thing you envy.
Would our buyers notice?
The filter that separates an interesting design choice from a competitive threat. A strength your buyers never encounter during an evaluation is not costing you deals, however good it is.
Could we adopt it?
Honestly, including partly and no. This column is what turns a competitive exercise into a product input, and it routes findings to product rather than only to sales, which is where most teardown value gets lost.
How to record where a competitor's product breaks down
Reproduce before you record. This is the section whose contents will end up in front of buyers, and a weakness you saw once and could not repeat is an anecdote rather than a finding. The two columns that matter most are whether a real buyer would encounter it and whether you managed to reproduce it, and both are routinely skipped in favour of a longer list.
What breaks
Precisely, with the conditions: "alert volume becomes unusable past roughly five tracked competitors". A weakness stated without its threshold is unusable in a sales conversation, because the buyer's first question is when.
At what point
The threshold or trigger. This is what makes the weakness predictive rather than anecdotal, and it lets a rep judge whether the buyer in front of them will hit it.
Would a real buyer hit it?
Answer honestly, including no. Breaking a product by using it in a way no customer would is not a competitive finding, and a page of those destroys the credibility of the genuine ones next to them.
Reproduced?
Yes or no, with how many times. Unreproduced observations stay in the document as observations and never graduate into sales material. This single column prevents most of the errors this artifact is prone to.
Corroborate with review sites
A weakness you observed once, which also appears repeatedly in their public reviews, is a pattern rather than a glitch. Our study of 500 verified G2 reviews across the four leading competitive intelligence tools exists partly for this: it found alert noise material in all four, at 3.3% of reviews for Klue, 10.0% for Crayon, 5.0% for Kompyte and 6.0% for Contify. Independent corroboration turns a teardown observation into a citable claim.
How to read strategy from a competitor teardown
This is the analysis layer and the reason a teardown is worth more than a feature comparison. Product decisions are strategy made visible: what a company builds carefully, what it leaves rough, what it puts behind a paywall and which user it optimises the defaults for all reveal priorities that no marketing page will state. Read the pattern across the whole product rather than drawing conclusions from single features.
Observation
A pattern, not an instance: "every workflow ends in a battlecard". Single features are weak evidence of strategy; a consistent shape across the product is strong evidence.
What it suggests about their priorities
The inference, stated as an inference: "built for sales enablement, not for product or strategy teams". Where they invest polish tells you who they are trying to win, and where they tolerate roughness tells you who they have decided not to serve.
Confidence
High, medium or low with the reason. "High, consistent across the whole product" carries very different weight from a read based on one screen, and unlabelled inferences in this section have a habit of becoming stated facts about the competitor's roadmap.
What we would expect next if true
The most valuable column in the template, because it is falsifiable: "deeper sales-tool integrations before analytics". Write the prediction down and check it at the next teardown. Predictions that hold make the analysis trustworthy; predictions that fail tell you the read was wrong, which is equally useful and almost never discovered otherwise.
Read the defaults and the gates
Default settings reveal the user they optimise for. What sits behind the paywall reveals what they believe is worth paying for. Both are more reliable signals of strategy than anything in their messaging.
How to fill in the decisions section of your competitor teardown
Three to five rows, and split them deliberately between what sales can use now and what product should consider later. Teardowns generate both kinds of finding and teams routinely act on only one: sales-facing organisations extract the talk tracks and discard the product insight, while product-led ones do the reverse. Naming both in the same table is what stops half the value being lost.
Finding
Observed and specific: "their alerts have no severity classification". Findings phrased as impressions rather than observations do not survive the first challenge in a review.
What we will do
Usually two actions per teardown: a discovery question or proof point for sales, and a product observation for the roadmap conversation. The sales action ships in a week and the product one takes a quarter, which is a reason to write both down rather than to prioritise one.
Owner
Named, and split by type. Sales enablement rows go to product marketing, adoptable strengths go to product, and a teardown whose findings all default to the person who performed it will produce no change at all.
By when
Short dates for the enablement rows. Teardown findings decay because the product you tore down keeps changing, so a finding still unactioned after two months may no longer be true.
How we will know it worked
Observable where possible: "alert relevance appears in more competitive win reasons". This also tells you whether the teardown was worth repeating, which is the question a second round has to answer.
Sourcing and upkeep: keeping a competitor teardown credible
These rules apply to every section above. A teardown's entire value rests on being first-hand, dated and reproducible. Every habit below defends one of those three properties, because once a teardown contains a single confidently-stated inference that turns out to be wrong, the accurate observations around it stop being believed too.
Record what you observed, mark what you inferred
The separation is the discipline that makes this artifact worth more than a competitor profile. Observed behaviour and marketing claims never share a column.
Screenshot and date everything
Number the images and reference them per row. Software changes weekly, and an undated claim about a competitor's product is unverifiable within a quarter and indefensible in front of a buyer.
Reproduce before you record a weakness
Once is an anecdote. Twice under stated conditions is a finding. Only the second kind should ever reach a battlecard.
Use access routes you would be comfortable describing
Public trials, documentation, recorded demos, review walkthroughs and customers who use both products answer nearly every question worth asking. If a finding could only have been obtained by misrepresenting who you are, it is not a finding you can circulate, cite or defend.
Record genuine strengths and adoptable ideas
A teardown that finds nothing impressive was not done carefully, and it routes zero value to product. The transferable mechanism behind a competitor's strength is frequently worth more than any weakness you found.
Write down a falsifiable prediction
The strategy section should end in something you can check next time. Predictions that hold earn the analysis credibility; predictions that fail correct your model, which no amount of additional observation will do on its own.
Repeat it and keep the old version
Twice a year, plus after significant launches. The comparison between two teardowns shows trajectory, which is more decision-useful than any single snapshot, and it is impossible to reconstruct once the first version is overwritten.
A competitor teardown example
You work in product at Pipedrive. Two competitive analyses have ended with rows nobody could resolve from published pages, so you spend an afternoon inside HubSpot Sales Hub. This is that teardown, filled in.
Published pricing and packaging verified 2 August 2026, from the companies’ own pages rather than third-party round-ups, which frequently conflate annual and monthly prices. Pricing changes without notice, so re-check before quoting any of it.
Sections marked illustrative are invented for this example. Win rates, deal counts, discounting behaviour, customer quotes, owners and internal dates are not published by HubSpot, Pipedrive or anyone else, so those rows are a plausible fictional scenario rather than reported fact, and should not be read as claims about how either company performs or negotiates. Everything else comes from the two pricing pages linked below, read on the date shown.
Teardown scopeIllustrative
| Field | Example entry |
|---|---|
| Competitor and product | HubSpot Sales Hub, Starter tier |
| Question this answers | How far can a sales team get before hitting a paywall or a sales call? |
| Version or date of the build | As seen on 2 Aug 2026 |
| Who performed the teardown | Sam L., Product, with Tom A., Competitive Intelligence |
| Time spent | 3 hours |
| Date performed and next teardown | 2 Aug 2026, next teardown 1 Feb 2027 |
1. Access and methodIllustrative
| Source of access | What it let us see | What it did not | Constraints or terms noted |
|---|---|---|---|
| Public free tier under our own company name | Signup, core CRM, seat administration, in-product upgrade prompts | Professional and Enterprise capabilities, and anything behind a sales conversation | Free tier caps at 2 users, which limits any multi-seat observation |
| Their published documentation and pricing page | Tier gating, seat types, credit allowances, onboarding fees | How any of it behaves in practice above Starter | None noted |
2. First-run experienceIllustrative
| Step | Time taken | What happened | Friction or delight | Screenshot reference |
|---|---|---|---|---|
| Signup to account created | 4 minutes | Email verification, then a short setup questionnaire | Delight: no credit card and no call required | shot-01 |
| Account created to first deal in a pipeline | 11 minutes | Default pipeline provided, first deal entered manually | Neutral: sensible defaults, more configuration options than needed at this stage | shot-03 |
| First deal to first useful report | 9 minutes | Report library is broad; finding the simple pipeline view took longer than building it | Friction: breadth works against speed here | shot-06 |
3. Core workflow walkthroughIllustrative
| Step in the job | How their product does it | How ours does it | Who does it better, and why | Evidence |
|---|---|---|---|---|
| Getting a new rep to their first logged activity | Setup questionnaire, then a guided checklist | Straight into a pipeline view with a sample deal | Us, marginally; fewer decisions before the first useful action | shot-03, shot-04 |
| Giving a manager visibility without granting edit rights | Assign a View-Only Seat at no cost | Not possible; every user consumes a paid seat | Them, clearly, and this confirms the pricing-page reading | shot-08, their pricing page |
| Automating a follow-up sequence | Upgrade prompt to Professional | Included from Growth at €39/seat/mo annual | Us, on where it sits in the ladder | shot-11, both pricing pages |
4. Capability observationsIllustrative
| Capability | Observed behaviour | Claimed but not verified | Depth | Screenshot reference |
|---|---|---|---|---|
| Seat types | View-Only Seats assignable at no cost, confirmed in the account | Nothing outstanding; this matched the pricing page exactly | Deep, and cleanly implemented | shot-08 |
| Automation | Not reachable on the free tier; an upgrade prompt appears at the point of use | "Automation" capability on Professional per their pricing page | Not observed, do not claim either way | shot-11 |
| Credit allowances | Not exercised at free-tier volumes | 500 / 3,000 / 5,000 across the paid tiers per their pricing page | Not observed | Their pricing page |
5. Where it is genuinely goodIllustrative
| Strength | Why it works | Would our buyers notice? | Could we adopt it? |
|---|---|---|---|
| Free View-Only Seats | Viewing is unbundled from editing at the seat level, not offered as a discount | Yes, and it decided three of our losses last quarter | Yes, this is packaging rather than engineering |
| No credit card at signup | The evaluation starts before any commercial conversation | Yes, though we do the same | Already have it |
6. Where it breaks downIllustrative
| What breaks | At what point | Would a real buyer hit it? | Reproduced? | Screenshot reference |
|---|---|---|---|---|
| Finding a simple pipeline report | First session, before any customisation | Yes, most evaluators build a pipeline view first | Yes, twice | shot-06 |
| Two-user cap on the free tier | As soon as a third person needs access | Yes, but it is a stated limit rather than a defect | Yes | shot-02 |
7. What it reveals about their strategyIllustrative
| Observation | What it suggests about their priorities | Confidence | What we would expect next if true |
|---|---|---|---|
| Viewing is free and unbundled at the seat level | They are optimising for organisation-wide spread rather than for seat revenue | High, it is consistent between the product and the pricing page | More capability aimed at non-sales stakeholders, not fewer free seats |
| Automation is gated at Professional, two tiers above entry | Automation is their main upgrade trigger | High, the upgrade prompt appears exactly at the point of use | Any discounting to focus on onboarding fees rather than the seat rate |
| Report breadth slows the first useful output | They are built for a broad audience rather than for a rep in a hurry | Medium, based on one session at one tier | Guided setups rather than fewer reports |
8. Decisions, owners and datesIllustrative
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| Free View-Only Seats are confirmed in-product, not just on the pricing page | Scope a read-only option with a dilution model | Sam L., Product | 3 Oct 2026 | A build-or-accept decision is recorded |
| Time to first useful report is our clearest observed advantage | Time both products through the same task and put the number in the battlecard | Maya R., Product Marketing | 12 Sep 2026 | Setup speed appears as a stated win reason next quarter |
| Two capability rows remain unobserved above Starter | Decide whether a paid tier evaluation is justified before the next teardown | Tom A., Competitive Intelligence | 1 Feb 2027 | Next teardown either closes the rows or records the limit explicitly |
How to roll out your competitor teardown
- 1Copy or download the blank teardown. Use Copy to paste it straight into Google Sheets or Excel with the columns intact, or download the CSV, Notion or PDF version.
- 2Name one product and one question before you start. An open-ended teardown produces forty screenshots and no conclusion. A question finishes in an afternoon with findings.
- 3Delete the example rows. Each table ships with one example row so the pattern is obvious. Remove it before you circulate the teardown.
- 4Record how you got access, and what it did not let you see. Everything downstream depends on it, and naming the boundary is what stops the teardown being read as comprehensive.
- 5Time the first run, step by step. Six minutes from signup to first value is a competitive fact. Fast onboarding is not. Time your own product through the same steps.
- 6Keep observed behaviour and marketing claims in separate columns. The gap between what a product does and what its website says is frequently the most valuable finding in the document.
- 7Reproduce every weakness before recording it. Once is an anecdote, twice under stated conditions is a finding, and only the second kind should reach a battlecard.
- 8End with a falsifiable prediction and three to five owned decisions. Split them between what sales uses now and what product considers later, or half the value gets discarded.
Competitor teardown FAQ
What is a competitor teardown?
A competitor teardown is primary research on a competitor's product: you use it, record what you observed with dates and screenshots, and analyse what those observations reveal about their priorities. The distinction from a competitor profile is the source of the evidence. A profile is assembled from what a competitor publishes about themselves and what third parties say. A teardown is what you saw with your own eyes, which is why its findings are both harder to obtain and considerably more defensible.
What does teardown mean?
Worth disambiguating, because this search term spans two unrelated worlds. In construction and property, a teardown means demolishing a building, and questions about teardown cost, permits and duration belong to that field. In product and go-to-market work, a teardown means disassembling something to understand how it was built and why. The term was borrowed from hardware engineering, where taking apart a competitor's device to examine its components is long-standing practice. Nothing beyond the metaphor transfers between the two meanings.
What is a product teardown?
A product teardown examines how a product actually works by using it: the onboarding, the core workflows, the depth behind each capability, where it is polished and where it is rough. In software the term covers examining a competitor's application, and an app teardown is the same exercise applied to a mobile app, sometimes extending to inspecting the app bundle for unreleased features. A competitor teardown is a product teardown performed on a competitor specifically, with the added step of asking what the product decisions reveal about their strategy.
What is the teardown process?
Eight steps, in this order. Scope it to one product and one question. Record how you got access and what that access does not cover. Time the first-run experience step by step. Walk the core workflow end to end as a real user would. Record capability observations, keeping what you saw separate from what they claim. Note genuine strengths and where the product breaks down, reproducing each weakness. Read the pattern across the product for what it reveals about their priorities. Finish with owned decisions split between sales enablement and product.
What is a teardown plan?
A teardown plan is what you decide before opening the product: which competitor and tier, the single question you want answered, how you will get access, how long you are spending, and what you will record. It matters more than it sounds, because the failure mode of this exercise is wandering. Without a plan you finish with a large number of screenshots, no conclusion and no appetite to repeat it. The scope section of this template is the plan, and it takes about ten minutes to fill in.
How do you create a product teardown?
Get honest access, usually a public self-serve trial, their documentation, a recorded demo, or a customer who uses both products. Set a timer and record the first run in real steps with real minutes. Then follow the job the product exists to do, from start to finish, writing down what the product does rather than what features it has. Keep observed behaviour in one column and their marketing claims in another. Screenshot everything and number the images. Finish by asking what the pattern of their product decisions says about who they are building for.
How do you break down a competitor's product?
By job, not by feature. Feature-by-feature comparison makes two products look similar because both have alerts, dashboards and reports. Following the job end to end reveals that one requires four manual steps where the other requires none, which is the difference a buyer actually experiences. Count the human interventions each product needs to complete the job, note which steps are gated behind a paid tier, and observe which defaults each product chooses, since defaults reveal the user they optimised for.
How long does a competitor teardown take?
Two to three hours gives you a genuine read on the first-run experience and the core workflow, which is enough for most sales enablement purposes. A full teardown covering capability depth, breakage thresholds and strategic reads takes one to two days. Repeating an existing teardown is much faster, usually two hours, since you are looking for what changed. Record the time spent in the document, because it lets a reader calibrate how much weight the conclusions deserve, and an honest two-hour teardown is more useful than a two-hour teardown presented as comprehensive.
What is a teardown analysis?
Teardown analysis is the interpretation layer that sits on top of the observations: not what the product does, but what the fact that it does that reveals. If every workflow terminates in a battlecard, they are built for sales enablement rather than for product teams. If setup requires no sales contact, they are optimising for self-serve evaluation. If a capability is present but shallow, they built it to answer a checklist rather than to serve a job. The test of good teardown analysis is that it produces a prediction you can check at the next teardown.
What is the difference between a competitor teardown and a competitor profile?
The evidence. A competitor profile is a standing document covering the company: positioning, pricing, customers, funding, strengths and weaknesses, assembled from published sources and updated quarterly. A teardown is a point-in-time investigation of the product itself, based on first-hand use, producing observations nobody else has. They complement each other, and the sequence that works is a profile for the standing picture with teardown findings folded into it as they are produced. Our competitor profile template covers the first.
What is the difference between a competitor teardown and a pricing teardown?
A competitor teardown examines the product: how it works, what it does well, where it fails. A pricing teardown examines the commercial model: the value metric, packaging, gates and discounting behaviour. They are best done together, because packaging only makes sense once you have seen which capabilities the product treats as central, and the in-product upgrade prompts you encounter during a teardown are one of the better sources of information about their gates. Our pricing teardown template covers the commercial side.
Is it legal or ethical to sign up for a competitor's product?
Using a publicly available trial under your real identity is normal practice, and many software companies expect it. Two things genuinely matter. First, read their terms: some explicitly restrict competitor access, and if they do, that is a fact for your team to weigh rather than an obstacle to route around. Second, do not misrepresent who you are to obtain access. Beyond the ethics, a finding you cannot explain the provenance of is one you cannot circulate internally, cite in enablement or defend if challenged. Documentation, recorded demos, review walkthroughs and customers who use both products answer nearly every question worth asking without any of this difficulty.
How do you format a competitor analysis?
Match the format to the reader and the moment. A teardown is a working document with screenshots and dated observations, for the person doing the analysis and the small group acting on it. A one-pager is a three-minute briefing for a rep before a call. A battlecard is in-call ammunition. An executive brief is the strategic implications with the detail stripped out. The mistake is producing one document and circulating it to all four audiences, which serves none of them. Write the teardown, then derive the shorter formats from it.
How do you write a competitor analysis from teardown findings?
Lead with the conclusion, not the method. Open with the two or three findings that change a decision, then the evidence supporting them, then the method and its limits at the end. State clearly what you observed versus what you inferred, and include the access boundary so a reader knows what was not examined. Include the competitor's genuine strengths early, since a document that finds no merit in them will not be believed. Finish with owned decisions and dates rather than with a summary, which is the part that turns analysis into change.
What are the 5 steps of a competitive analysis?
The sequence most versions give is: identify competitors, gather data on each, analyse their strengths and weaknesses against yours, identify the resulting opportunities and threats, and act. That is a fair summary of the work. Worth being direct though: there is no canonical five-step model behind this question, and three-step, five-step and seven-step versions all circulate with equal confidence and no attributable origin. What determines whether competitive analysis works is not the step count but whether the evidence is dated and first-hand, and whether anything is owned at the end.
What are the 4 types of competitors, and which are worth tearing down?
Direct competitors solve the same problem for the same buyer. Indirect competitors meet the same need differently. Replacement competitors are what the buyer does instead of buying, most often a spreadsheet, an agency or an internal build. Potential competitors are adjacent players who could enter. Teardowns are worth the effort for direct competitors you meet regularly, and occasionally for a potential entrant whose product you want to understand before they arrive. Tearing down a replacement is a different exercise entirely, and it is usually better answered by asking customers how they currently do the job.
Can you use ChatGPT for a competitor teardown?
It cannot do the part that matters. A teardown's value is first-hand observation, and an assistant has not used the product. What it does well is the work around the observation: structuring your notes, turning a screenshot-heavy working document into a readable summary, drafting the strategic inference from a pattern you identified, or summarising a competitor's documentation before you start so you know what to look for. Never let it supply a capability claim, because a fabricated observation in a teardown destroys the credibility of the real ones sitting next to it.
How often should you run a competitor teardown?
Twice a year for competitors you meet regularly, plus an immediate teardown after any significant launch or redesign. Repeating one is much faster than the first pass because you are looking for change, and the comparison between two teardowns is where the genuinely valuable finding lives: whether their product is improving faster than yours, and whether the prediction you wrote down last time held. Keep every version rather than overwriting, since trajectory cannot be reconstructed from the current state alone.
Who should perform a competitor teardown?
Product marketing usually owns the exercise, but the most useful teardowns are done in pairs: someone who understands the product deeply and someone who understands what buyers ask. A product manager notices architectural decisions a marketer will miss, and a sales engineer notices exactly where an evaluation would stall. Whoever performs it, record their name and role in the document, since it tells the reader which blind spots to expect. The one arrangement that works badly is assigning it to whoever has spare time, since teardowns are largely judgement.
What is an example of a competitor teardown?
Scope: Northwind's Growth tier, answering one question, why buyers say their setup is fast. Access: a public 14-day self-serve trial under our own name, which showed signup, setup and the core workflow but not enterprise admin controls, and capped us at three tracked competitors. First run: six minutes from signup to first competitor added, email verification then a four-field wizard, recorded as delight because no sales call was required. Workflow: getting a change to a rep takes them a digest email plus a manual battlecard edit, where ours pushes it automatically, so we lead on that step. Capability: alerts fire on any page change including cosmetic edits, while their site claims AI filtering for relevance, which the trial did not demonstrate, so it stays in the unverified column. Breakage: alert volume becomes unusable past roughly five competitors, reproduced twice. Strategy read: every workflow ends in a battlecard, so they are built for sales enablement rather than product teams, high confidence, and we would expect deeper sales-tool integrations before analytics. That last sentence is a prediction to check at the next teardown.
What are the most common mistakes in a competitor teardown?
Six recur. Starting without a question, which produces screenshots and no conclusion. Mixing observed behaviour with marketing claims in the same column, which destroys the first-hand advantage the whole exercise exists for. Recording weaknesses without reproducing them, so anecdotes reach battlecards and get corrected by buyers. Finding no genuine strengths, which makes the document unbelievable and routes nothing to product. Undated observations of software that changes weekly. And drawing strategic conclusions from single features rather than from the pattern across the product.
Run every competitor teardown on current product data
Flares tracks what competitors ship and change, so each teardown starts from what their product does today.
Discover Flares14-day free trial · 30-second setup