Positioning · 14 min read · Updated 2 Aug 2026
Competitive Differentiation Matrix Template (Free Download)
A blank competitive differentiation matrix you can fill in today, plus the guidance for what belongs in each field. This matrix scores weighted buyer decision criteria rather than features, which is what makes it a positioning decision rather than a longer comparison grid.
Copy pastes straight into Google Sheets or Excel with the columns intact. Downloads are free with a work email.
The competitive differentiation matrix 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.
Matrix scope
Fill this in first. One segment per matrix. Criteria and weights change between segments, so a single matrix covering everyone scores nobody correctly.
- Segment this matrix coversSpecific, e.g. mid-market ops teams of 25 to 100 seats
- Buying situationReplacing a tool, consolidating after an acquisition, buying for the first time
- Alternatives scoredInclude the status quo and doing nothing, not only vendors
- Who validated the criteriaNamed buyers or deals, not an internal workshop
- Owner and date completedOne named owner, with the date
- Next reviewQuarterly, or after any significant competitor launch
1. Decision criteria and weights
First row is an example, delete it. Six to nine criteria. Weights must total 1.00, which forces the trade-offs a flat list hides.
| Criterion | What the buyer is actually asking | Weight | Evidence the weight is right |
|---|---|---|---|
| ExampleTime to first useful output | How long before my team stops using the spreadsheet? | 0.20 | Named in 7 of 12 won deals as a deciding factor |
2. The scored matrix
First row is an example, delete it. Score 1 to 4, where 1 is a major weakness and 4 a major strength. Multiply by the weight to get the gap.
| Criterion | Weight | Us (1-4) | Competitor A (1-4) | Status quo (1-4) | Weighted gap vs best |
|---|---|---|---|---|---|
| ExampleTime to first useful output | 0.20 | 4 | 2 | 1 | +0.40 in our favour |
3. Evidence behind each score
First row is an example, delete it. One row per score you would struggle to defend in a room. Unsourced scores are opinions with decimals attached.
| Criterion | Whose score | The evidence | Source and date | Confidence |
|---|---|---|---|---|
| ExampleTime to first useful output | Ours, 4 | Median 19 minutes from signup to first populated pipeline | Product analytics, Q2 2026 | High, measured not estimated |
4. Where we genuinely win
First row is an example, delete it. Only criteria with real weight behind them. Winning a criterion the buyer weights at 0.02 is not a differentiator.
| Criterion | Weight | Why we win it | Can a competitor copy it, and how fast? |
|---|---|---|---|
| ExampleTime to first useful output | 0.20 | No mandatory data import before the first pipeline view | Yes, but it is an onboarding rebuild, so 2 to 3 quarters |
5. Where we genuinely lose
First row is an example, delete it. Be specific and unflattering. A matrix where you win every weighted row will not be believed by the people who most need it.
| Criterion | Weight | Who beats us and by how much | What it costs us | Evidence |
|---|---|---|---|---|
| ExampleCost for view-only stakeholders | 0.15 | Competitor A, 3 points, they charge nothing for viewers | Surfaced in 4 of 9 losses last quarter | Loss reviews, Q2 2026 |
6. Close, neutralise or concede
First row is an example, delete it. One decision per losing criterion. Most gaps should be neutralised or conceded, because closing all of them is how products become undifferentiated.
| Gap | Decision: close, neutralise or concede | What that takes | Cost and time | Rationale |
|---|---|---|---|---|
| ExampleCost for view-only stakeholders | Neutralise | Reframe on total cost at the buyer's real seat count | One enablement cycle, no engineering | Closing it means rebuilding seat pricing for a 0.15 criterion |
7. How to say it
First row is an example, delete it. The sentence a rep uses, plus the proof and the version that would be unfair to say.
| Criterion | What we say | Proof we show | What we never say |
|---|---|---|---|
| ExampleTime to first useful output | Most teams have a populated pipeline in under 20 minutes | Product analytics median, and a customer in their segment | Anything asserting their setup is slow without a measured comparison |
8. Decisions, owners and dates
First row is an example, delete it. Three to five rows. Split them between what positioning changes now and what product changes later.
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| ExampleWe win the highest-weighted criterion and never lead with it | Move time to first output into the opening of every demo | Maya R., Product Marketing | 12 Sep | It appears as a stated win reason in more closed deals |
How to fill in your competitive differentiation matrix
How to scope a competitive differentiation matrix
One segment per matrix, and this is the constraint that decides whether the exercise produces anything usable. Decision criteria and their weights change substantially between segments: a 15-seat startup and a 400-seat enterprise are not weighing the same things, and averaging them produces a matrix that describes a buyer who does not exist. If you serve three segments, build three matrices. They will disagree with each other, and the disagreement is the most useful output.
Segment this matrix covers
Specific enough that you could name five customers in it: "mid-market ops teams of 25 to 100 seats", not "mid-market". If you cannot name the customers, you cannot validate the weights, and the matrix becomes an internal opinion survey.
Buying situation
Replacing an incumbent, consolidating after an acquisition, or buying this category for the first time. These weight criteria very differently. A replacement buyer weights migration effort heavily; a first-time buyer weights it near zero because they have nothing to migrate.
Alternatives scored
Include the status quo and doing nothing alongside the named vendors. April Dunford's framing in Obviously Awesome is the useful one here: your competitive alternatives are whatever the buyer would do if you did not exist, and for many deals that is a spreadsheet and a recurring meeting rather than a competitor.
Who validated the criteria
Named buyers or specific deals, not an internal workshop. Criteria invented in a room full of your own employees reliably over-weight the things you are good at, which is the single most common way this artifact becomes self-congratulatory.
Owner and date completed
One named owner, usually product marketing. Scores drift as competitors ship, and a matrix with no owner is a document that was accurate on the day it was written and has been quietly wrong ever since.
Next review
Quarterly, plus immediately after any significant competitor launch. Weights move more slowly than scores, so a review usually means re-scoring against the same criteria rather than rebuilding the matrix.
How to choose criteria and weights in a competitive differentiation matrix
Six to nine criteria, weighted so the weights total 1.00. That total is the entire mechanism. A flat list of things buyers care about lets everything be important, which is how comparison grids grow to forty rows and stop informing any decision. Forcing the weights to sum to one converts the list into a set of trade-offs, and the argument that breaks out while you assign them is more valuable than the finished numbers.
Criterion
Framed as the buyer's decision, not as your feature area. "Time to first useful output" is a criterion. "Onboarding module" is a feature. The first can be won by several different mechanisms; the second presupposes that everyone solves it the way you do.
What the buyer is actually asking
The question in their words: "how long before my team stops using the spreadsheet?". This column catches criteria that sound important internally and turn out to have no buyer question behind them, which is a common and expensive discovery.
Weight
A decimal, and the column must total 1.00. Assign weights from evidence rather than intuition, and expect the total to force you to demote something you thought mattered. If everything lands between 0.10 and 0.12 you have not really weighted anything, you have distributed politeness.
Evidence the weight is right
Deal counts, win/loss reasons, discovery notes, review-site themes. "Named in 7 of 12 won deals" is evidence. "Everyone cares about this" is not. Weights are where bias enters this artifact, so they need the tightest sourcing in the document.
Six to nine, no more
Below six you are missing the trade-offs; above nine the weights become so small that scoring differences stop being meaningful. If a tenth criterion feels essential, it is usually a sub-part of one already in the list.
This is the Competitive Profile Matrix mechanism
The weighted-criteria approach comes from the Competitive Profile Matrix in Fred R. David's Strategic Management, where critical success factors are weighted to total 1.0 and each firm is rated 1 to 4, with weight times rating giving a weighted score. That structure is genuinely established, unlike most numbered frameworks attached to this topic, and it is worth knowing the lineage when someone asks where the method comes from.
How to score a competitive differentiation matrix
Score every alternative 1 to 4 on every criterion, then multiply by the weight to see where the differences actually sit. The four-point scale is deliberate: it has no midpoint, so you cannot score everything a 3 and avoid the judgement. The weighted gap column is where the surprises appear, and the most common one is that a criterion the team argues about constantly turns out to carry a weight of 0.05 and therefore does not matter.
Us, and each competitor
Score 1 for a major weakness, 2 for a minor weakness, 3 for a minor strength, 4 for a major strength. Score your own product the way a buyer who has used both would, which is usually one point lower than the way you would like to.
Status quo
Score it honestly, and expect it to win rows. The spreadsheet scores 4 on cost and 4 on nobody has to learn anything, which is exactly why it beats well-funded products regularly. A matrix that omits the status quo is missing the alternative that wins most often.
Weighted gap vs best
Weight multiplied by the difference between your score and the best score on that row. This is the column that ranks your problems. A two-point deficit on a 0.05 criterion is worth 0.10 and can wait; a one-point deficit on a 0.20 criterion is worth 0.20 and cannot.
Do not average the total into a single number
You can total the weighted scores, and the total is mildly interesting, but it is not the output. Two products with identical totals can have completely opposite shapes, and the shape is what positioning is built from. Teams that reduce this matrix to one number per vendor have thrown away the reason they built it.
Score against the tier a buyer would actually buy
Comparing your mid-tier against a competitor's enterprise edition, or against their free plan, produces a matrix that no buyer's experience will match. Name the tier in the scope section and hold to it.
How to source the scores in a competitive differentiation matrix
Every score you would struggle to defend in a room needs a row here. This section is what separates a differentiation matrix from a confident opinion with decimals attached, and it is the section that gets cut when time is short. Resist that, because the first time a rep is contradicted by a buyer holding better information, the entire matrix stops being used regardless of how right the other rows were.
Whose score
Yours or a named competitor's, with the number. Your own scores need evidence more urgently than theirs do, since internal generosity is the bias this artifact is most prone to and the one nobody in the room will challenge.
The evidence
Specific and measured where possible: "median 19 minutes from signup to first populated pipeline". Measured beats estimated, first-hand beats reported, and a number with a method beats a number without one.
Source and date
Product analytics, a dated teardown, a pricing page read on a stated day, win/loss interviews, verified review themes. Competitor scores decay fastest, which is why the date matters more than it appears to.
Confidence
High, medium or low with a reason. "High, measured not estimated" and "low, inferred from their marketing" belong in the same document but not in the same sentence when a rep repeats them to a buyer.
Prefer primary sources for competitor scores
Their own pricing and documentation pages, a hands-on teardown, or customers who use both. Third-party round-ups are unreliable on exactly the details that matter: we verified several in August 2026 that conflated annual and monthly pricing, producing figures nobody's invoice would match.
How to record where you genuinely win in a competitive differentiation matrix
Only criteria that carry real weight. This is a filter, not a list of everything you do well, and applying it properly is uncomfortable. Winning decisively on a criterion the buyer weights at 0.02 is not a differentiator, it is a fact about your product that no purchasing decision has ever turned on. The durability column matters just as much, because an advantage a competitor can copy in six weeks is not a position, it is a head start.
Criterion and weight
Carry the weight across from the criteria table so the reader sees immediately whether a win is consequential. Wins on criteria below roughly 0.10 belong in the product's feature list, not in its positioning.
Why we win it
The mechanism, not the outcome: "no mandatory data import before the first pipeline view", not "better onboarding". Mechanisms are defensible in a conversation and can be demonstrated; outcomes are claims that invite a counter-claim.
Can a competitor copy it, and how fast?
The most useful column in this section. Packaging decisions can be copied in a quarter. Architectural choices take years. Something rooted in your business model may be effectively uncopiable because matching it would break their economics, and those are the wins worth building a position on.
Expect two or three, not eight
A matrix showing you winning eight weighted criteria has almost certainly been scored generously. Genuine differentiation is narrow, and a document claiming otherwise gets quietly discounted by every sales rep who has met a real buyer.
How to record where you lose in a competitive differentiation matrix
Specific, unflattering, and sourced. This is the section that determines whether anyone trusts the rest of the document. A matrix where you win every weighted row will be dismissed by exactly the people whose behaviour it was meant to change, because they have been in the deals and know better. Sales teams are extremely good at detecting a document written to reassure rather than to inform, and they stop opening it.
Who beats us and by how much
Named, with the point difference: "Competitor A, 3 points, they charge nothing for viewers". Vagueness here reads as reluctance, and a rep who has lost three deals on this criterion will trust the entire matrix less for it.
What it costs us
Quantified where you can: "surfaced in 4 of 9 losses last quarter". This is what separates a gap that hurts from a gap that merely exists, and it is what determines whether the response is worth its cost.
Evidence
Loss reviews, win/loss interviews, deal notes. Sales anecdote is a starting point, not evidence. The distinction matters because the loudest gap in an internal discussion is frequently not the one appearing most often in the data.
Include the status quo's wins
Losing to a spreadsheet on cost and on nobody has to learn anything is a real loss and belongs in this table. Teams routinely record competitor losses and omit status-quo losses, which is how a category ends up competing hard with itself while the actual winner is inertia.
How to decide what to close, neutralise or concede
One explicit decision per losing criterion, and most of them should not be close. This is the section that makes the matrix a strategy document rather than a scoreboard. Deciding to close every gap is the default failure mode of competitive analysis, and it produces a product that matches competitors on everything and leads on nothing. Conceding a gap deliberately, and saying why, is a legitimate and frequently correct answer.
Close
Build or fix it. Justified when the criterion is heavily weighted, the gap is large, and the fix is achievable in a horizon that matters. Closing a 0.20 criterion where you score 2 against a 4 is usually right; closing a 0.05 criterion because it annoys people internally is usually not.
Neutralise
Change the conversation rather than the product: reframe the criterion, bundle around it, or make the gap irrelevant at the buyer's real usage. Most gaps belong here. Neutralising costs an enablement cycle rather than a roadmap quarter, which is why it is under-used relative to how well it works.
Concede
Say plainly that you do not compete on this, and mean it. Conceding well is a positioning move: it signals confidence, it saves reps from defending indefensible ground, and it often improves credibility on the criteria you do claim. The concession must be paired with a criterion you win, or it is just a weakness.
Cost and time
Honest figures, including engineering quarters where relevant. A gap analysis with no costs attached always concludes that everything should be closed, because nothing in the document argues otherwise.
Rationale
One sentence naming the trade-off: "closing it means rebuilding seat pricing for a 0.15 criterion". This is the sentence that will be quoted back to you in six months when someone reopens the decision, so it is worth writing carefully.
How to turn a competitive differentiation matrix into what reps say
Scores do not survive contact with a sales call. This section converts each conclusion into the sentence a rep actually uses, the proof they show when challenged, and the version that would be unfair to say. That last column exists because differentiation material has a strong tendency to drift into disparagement as it passes from a document to a deck to a conversation, and the drift always happens at the point where the proof runs out.
What we say
One sentence, in a buyer's language, specific enough to be checked: "most teams have a populated pipeline in under 20 minutes". Sentences that cannot be checked are also sentences a buyer cannot be persuaded by.
Proof we show
What goes on screen: the measurement and its method, a customer in their segment, a documented capability, both pricing pages side by side. Every claim in this table needs one, and a claim whose proof is "our website says so" is not proof.
What we never say
The unfair version of the same point, written out so nobody has to guess where the line is. "Anything asserting their setup is slow without a measured comparison" is a useful entry. Naming competitors is fine and often necessary; asserting things about them you cannot source is not.
Keep it factual about competitors
State what is verifiable, cite where it came from, and let the comparison do the work. Beyond being the right thing to do, disparagement is fragile: one correction from a buyer who knows the competitor better than your rep does costs more credibility than the entire matrix earned.
Push these lines into the battlecard
This table is the source, but reps do not read matrices before calls. The sentences and proof belong in the maintained battlecard where they are actually reachable, and the matrix is what justifies them.
How to fill in the decisions section of your competitive differentiation matrix
Three to five rows, split between what positioning changes now and what product changes later. The most common finding in a completed matrix is not a gap at all: it is that the team wins the highest-weighted criterion and never mentions it, because it has become so familiar internally that nobody thinks of it as remarkable. That finding costs nothing to act on and is frequently worth more than any roadmap item in the same document.
Finding
Drawn from the weighted numbers, not from the discussion: "we win the highest-weighted criterion and never lead with it". Findings that restate what everyone already believed are a sign the weights were assigned to confirm rather than to test.
What we will do
One action per finding, specific enough to schedule. Messaging and enablement changes ship in weeks; product changes take quarters. Writing both in the same table is what stops the fast ones being deferred alongside the slow ones.
Owner
Named, and split by type. Positioning and narrative rows go to product marketing, close decisions go to product, pricing responses go to whoever owns pricing. Rows defaulting to whoever built the matrix produce no change.
By when
Real dates. Scores decay as competitors ship, so a decision still unactioned at the next quarterly review is usually operating on numbers that are no longer accurate.
How we will know it worked
Observable: "it appears as a stated win reason in more closed deals". This is also the check on whether the weights were right, since acting on a heavily weighted criterion should visibly move deal outcomes, and if it does not, the weight was wrong.
Sourcing and upkeep: keeping a competitive differentiation matrix honest
These rules apply to every section above. This artifact has one characteristic failure mode and it is not inaccuracy, it is flattery. Built internally, scored by people who work on the product, and reviewed by people who want it to say something encouraging, it drifts toward a document showing you winning comfortably. Every habit below is a defence against that drift, because a flattering matrix is worse than none: it is acted upon.
Weights come from buyers, scores come from evidence
The two most corruptible parts of the document. Weights derived from an internal workshop over-value what you are good at; scores assigned from memory over-value your own product. Both need sourcing, and the weights need it more.
Score the status quo, and expect it to win rows
Doing nothing is the alternative that wins most often in most categories. A matrix listing only vendors is describing a competition the buyer is not necessarily having.
Do not reduce it to a single total per vendor
The shape is the output, not the sum. Two vendors with equal totals and opposite shapes call for completely different positioning, and the total conceals exactly that.
Most gaps should be neutralised or conceded
Closing every gap is how a product ends up matching competitors on everything and leading on nothing. If your response column reads close on every row, the matrix has been used as a roadmap wish list rather than as a positioning decision.
Re-score quarterly, and keep the old version
Competitor scores decay fastest; weights move slowly. Keeping the previous version is what lets you see whether a gap is widening or closing, which is considerably more decision-useful than the current snapshot and impossible to reconstruct after an overwrite.
Name competitors, never disparage them
Everything asserted about a competitor should trace to a source you could show a buyer. This is both the honest approach and the durable one, since an unsourced claim survives only until it meets someone better informed.
Three segments usually need three matrices
If you find yourself arguing about a weight, the argument is often two segments disagreeing inside one document. Splitting it resolves the argument and produces two more useful artifacts.
A competitive differentiation matrix example
You run product marketing at Pipedrive. Mid-market deals of 25 to 100 seats keep coming down to the same handful of criteria against HubSpot Sales Hub, so you score them properly. This is that matrix, 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.
Matrix scopeIllustrative
| Field | Example entry |
|---|---|
| Segment this matrix covers | Mid-market sales teams, 25 to 100 seats, no dedicated ops headcount |
| Buying situation | Replacing a first CRM or consolidating two after an acquisition |
| Alternatives scored | Us, HubSpot Sales Hub, and staying on spreadsheets |
| Who validated the criteria | 12 win/loss interviews from Q2 2026, 7 won and 5 lost |
| Owner and date completed | Maya R., Product Marketing, 2 Aug 2026 |
| Next review | 1 Nov 2026, or sooner if either competitor changes packaging |
1. Decision criteria and weightsIllustrative
| Criterion | What the buyer is actually asking | Weight | Evidence the weight is right |
|---|---|---|---|
| Time to first useful output | How long before my team stops using the spreadsheet? | 0.20 | Named in 7 of 12 interviews as a deciding factor |
| Total cost at our real seat count | What does this actually cost once everyone who needs it has access? | 0.20 | Raised in all 5 losses, and in 3 of 7 wins |
| Cost for view-only stakeholders | Do I have to pay for people who will never enter data? | 0.15 | Named in 4 of 5 losses |
| Depth of automation | Can we stop doing the follow-up admin by hand? | 0.15 | Raised in 6 of 12, decisive in 1 |
| Reporting and consolidation | Can I get one forecast out of two teams? | 0.15 | The stated driver in both acquisition-led deals |
| Migration effort from the current system | How much work is it to move, and who does it? | 0.10 | Raised in 5 of 12, never decisive on its own |
| Breadth of the wider platform | Will this cover marketing and service later? | 0.05 | Raised in 2 of 12, both larger accounts |
2. The scored matrixIllustrative
| Criterion | Weight | Us (1-4) | Competitor A (1-4) | Status quo (1-4) | Weighted gap vs best |
|---|---|---|---|---|---|
| Time to first useful output | 0.20 | 4 | 2 | 1 | +0.40 in our favour |
| Total cost at our real seat count | 0.20 | 3 | 3 | 4 | -0.20 against the status quo |
| Cost for view-only stakeholders | 0.15 | 1 | 4 | 4 | -0.45 against both |
| Depth of automation | 0.15 | 3 | 4 | 1 | -0.15 against HubSpot |
| Reporting and consolidation | 0.15 | 3 | 4 | 1 | -0.15 against HubSpot |
| Migration effort from the current system | 0.10 | 3 | 3 | 4 | -0.10 against the status quo |
| Breadth of the wider platform | 0.05 | 2 | 4 | 1 | -0.10 against HubSpot |
3. Evidence behind each scoreIllustrative
| Criterion | Whose score | The evidence | Source and date | Confidence |
|---|---|---|---|---|
| Time to first useful output | Ours, 4 | Median 19 minutes from signup to a populated pipeline | Product analytics, Q2 2026 | High, measured not estimated |
| Cost for view-only stakeholders | HubSpot, 4 | View-Only Seats are assignable at no cost on their published pricing | HubSpot pricing page, read 2 Aug 2026 | High, primary source and confirmed in-product |
| Total cost at our real seat count | Both, 3 | Our Growth tier is EUR 39 per seat per month billed annually; their Sales Hub Professional is USD 90 per seat per month billed annually, and both list onboarding fees separately | Both pricing pages, read 2 Aug 2026 | Medium, the two are quoted in different currencies so any comparison must state which |
| Depth of automation | HubSpot, 4 | Automation sits on Professional; we include it from Growth, so they lead on depth and we lead on where it starts | Both pricing pages, read 2 Aug 2026 | Medium, depth was not tested hands-on above entry tier |
| Breadth of the wider platform | HubSpot, 4 | Sales Hub is one hub within a broader suite | Their published product pages | High, and not contested |
4. Where we genuinely winIllustrative
| Criterion | Weight | Why we win it | Can a competitor copy it, and how fast? |
|---|---|---|---|
| Time to first useful output | 0.20 | No mandatory data import before the first pipeline view, and fewer setup decisions | Yes in principle, but it is an onboarding rebuild rather than a feature, so 2 to 3 quarters |
| Depth of automation relative to where it starts | 0.15 | Automation is included from our second tier rather than our third | Yes, and cheaply, since it is a packaging decision rather than an engineering one |
5. Where we genuinely loseIllustrative
| Criterion | Weight | Who beats us and by how much | What it costs us | Evidence |
|---|---|---|---|---|
| Cost for view-only stakeholders | 0.15 | HubSpot by 3 points, and the status quo by 3, since neither charges for viewers | Named in 4 of our 5 losses last quarter | Win/loss interviews, Q2 2026 |
| Total cost at our real seat count | 0.20 | The spreadsheet by 1 point, because it is free | The deals we never enter, which do not appear in loss data at all | Inferred, and deliberately marked as the weakest evidence in this matrix |
| Breadth of the wider platform | 0.05 | HubSpot by 2 points | Very little, this is our lowest-weighted criterion | Raised in 2 of 12 interviews, decisive in none |
6. Close, neutralise or concedeIllustrative
| Gap | Decision: close, neutralise or concede | What that takes | Cost and time | Rationale |
|---|---|---|---|---|
| Cost for view-only stakeholders | Neutralise | Ask how many of the seat count would ever edit a record, then compare totals at that split rather than per seat | One enablement cycle, no engineering | Closing it means rebuilding seat pricing for a 0.15 criterion, and it would reopen our whole pricing model |
| Depth of automation | Concede on depth, compete on where it starts | Say plainly they go deeper, and lead with automation being included a tier earlier | Messaging only | A 0.15 criterion where we are one point behind does not justify a roadmap quarter |
| Reporting and consolidation | Close | Cross-team forecast consolidation in a single view | Two engineering quarters | It is the stated driver in acquisition-led deals, which is our fastest-growing entry point |
| Breadth of the wider platform | Concede | State that we are a sales tool and not a suite, and stop defending it | Nothing | Weighted 0.05 and decisive in no deal we have reviewed |
7. How to say itIllustrative
| Criterion | What we say | Proof we show | What we never say |
|---|---|---|---|
| Time to first useful output | Most teams have a populated pipeline in under 20 minutes, with no data import first | Our measured median, and a customer in their segment | Anything asserting their setup is slow, which we have not measured |
| Cost for view-only stakeholders | They price viewers at zero, which is genuinely good if most of your seats never edit. Worth checking how many of yours do | Both pricing pages side by side, and the buyer's own seat split | Any suggestion the free seats are restricted in ways we have not verified |
| Depth of automation | They go deeper on automation. Ours is included a tier earlier, so most teams reach it for less | Both pricing pages, with the tier each capability sits on | That their automation is hard to use, which is not something we tested |
8. Decisions, owners and datesIllustrative
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| We win the highest-weighted criterion and never lead with it | Move time to first output into the opening two minutes of every demo | Maya R., Product Marketing | 12 Sep 2026 | It appears as a stated win reason in more closed deals next quarter |
| The view-only seat gap decided 4 of 5 losses and we keep arguing per seat | Retrain the team to compare totals at the buyer's real edit or view split | Dana K., Sales | 29 Aug 2026 | The objection stops appearing as a primary loss reason |
| Reporting is the only gap worth closing on the weights | Scope cross-team forecast consolidation for the roadmap | Sam L., Product | 3 Oct 2026 | A build-or-defer decision is recorded with a date |
| Our status-quo loss evidence is the weakest row in the matrix | Add a question about what they use today to the discovery script | Dana K., Sales | 29 Aug 2026 | Next quarter's matrix scores the status quo from data rather than inference |
How to roll out your competitive differentiation matrix
- 1Copy or download the blank matrix. Use Copy to paste it straight into Google Sheets or Excel with the columns intact, or download the CSV, Notion or PDF version.
- 2Pick one segment and name five customers in it. Criteria and weights change between segments. If you cannot name the customers, you cannot validate the weights.
- 3Delete the example rows. Each table ships with one example row so the pattern is obvious. Remove it before you circulate the matrix.
- 4Write six to nine criteria as buyer questions. Frame each as the decision a buyer is making, not as your feature area, then assign weights that total exactly 1.00.
- 5Score every alternative including the status quo. One to four, with no midpoint. Score the spreadsheet honestly and expect it to win rows on cost and familiarity.
- 6Multiply weight by gap to rank your problems. A one-point deficit on a 0.20 criterion outranks a two-point deficit on a 0.05 one. Do not total it into a single score.
- 7Decide close, neutralise or concede for each gap. Most should be neutralise or concede. Closing every gap is how products end up matching competitors on everything and leading on nothing.
- 8Convert the conclusions into sentences reps can use. One sentence, one proof and one thing never to say per criterion, then push them into the battlecard where reps will actually find them.
Competitive differentiation matrix FAQ
What is a competitive differentiation matrix?
A competitive differentiation matrix scores you and every alternative a buyer is considering against weighted decision criteria, where the weights total 1.00 and each alternative is rated 1 to 4. Multiplying weight by the gap shows where differences actually matter, which is usually not where internal debate concentrates. The defining property is the weighting: a comparison grid treats every row as equally important, and a differentiation matrix does not, which is what turns a comparison into a positioning decision.
What is a competitive comparison matrix?
A competitive comparison matrix lists capabilities down one axis and vendors across the other, marking who has what. It answers whether a capability exists. A differentiation matrix asks a different question: given what this buyer actually weighs, where do the meaningful differences sit? The comparison matrix is the input; the weighting is what makes it decision-useful. Both are worth having, and the common failure is building the first, discovering it has thirty-eight rows and settles nothing, and concluding the method does not work.
What is competitive differentiation?
Competitive differentiation is the set of things you do better than the alternatives, on criteria buyers actually weight heavily, that competitors cannot easily copy. All three conditions are required. Being better at something nobody weighs is not differentiation. Being better at something copied in a quarter is a head start. And being different is not the same as being differentiated: a difference only counts when it maps to a decision a buyer is making. Genuine differentiation is usually narrower than teams expect, often two or three criteria.
What is another word for competitive differentiation?
Several terms overlap. Unique selling proposition and USP describe the single strongest differentiator in advertising language. Differentiation strategy is Porter's term for competing on distinctiveness rather than cost. Competitive advantage is broader, covering structural advantages like scale that are not buyer-facing at all. A kill sheet is the sales-side name for a document built on this analysis, sometimes called a competitive kill sheet, which lays out the decision criteria for a segment and where each alternative falls short. It is essentially a differentiation matrix turned into sales material.
What is a kill sheet?
A kill sheet is a sales document, also called a competitive kill sheet, that positions your solution against the alternatives on the criteria a target segment cares about most. In practice the term is used interchangeably with battlecard, and most published examples are competitor profiles rather than anything more structured. The useful version, argued well by SellingBrew, is exactly what this template produces: the decision criteria that matter to a segment, weighted, with where each alternative falls short. If you build the matrix, you have built the kill sheet.
What is the McKinsey competitor matrix?
There is no McKinsey competitor matrix, and this query is usually reaching for the GE-McKinsey nine-box matrix. McKinsey built that for General Electric in the early 1970s, when GE had more than 150 business units and needed something more nuanced than the BCG growth-share matrix. It plots industry attractiveness against competitive strength on a three-by-three grid to decide where to invest, hold or divest. It is a corporate portfolio tool for allocating capital across your own business units, not a tool for comparing competitors. The closer fit for a competitor grid is the Competitive Profile Matrix.
What is an example of a competitive profile matrix?
A Competitive Profile Matrix lists critical success factors down the left with a weight against each, weights summing to 1.0, then rates your firm and each competitor from 1 to 4, where 1 is a major weakness and 4 a major strength. Weight multiplied by rating gives a weighted score per factor, and the column totals compare firms. It comes from Fred R. David's Strategic Management, where it sits alongside the Internal and External Factor Evaluation matrices in the input stage. Typical factors include price competitiveness, product quality, customer loyalty, market share and financial position.
How do you develop a competitive profile matrix?
Identify the critical success factors for your industry, ideally from buyer evidence rather than internal discussion. Assign each a weight reflecting its importance, and make the weights total 1.0 exactly. Rate your firm and each competitor 1 to 4 on every factor. Multiply weight by rating for each cell, then total each firm's column. The mechanics are simple; the difficulty is entirely in the weights, which is where bias enters. Weights assigned by people who work on the product reliably over-value what that product is good at, so source them from win/loss data.
How do you do a competitive matrix?
Pick one segment, since criteria and weights differ between segments and averaging them describes a buyer who does not exist. List six to nine criteria framed as buyer questions rather than as your feature areas. Weight them to total 1.00, which forces the trade-offs a flat list hides. Score every alternative 1 to 4, including the status quo. Multiply weight by gap to rank the differences. Then decide, for each gap, whether to close, neutralise or concede it. That final step is what makes it a strategy document instead of a scoreboard.
How do you build a comparison matrix?
Rows are what you compare on, columns are the alternatives, and the two decisions that matter are which rows and who validates them. Keep rows in buyer language and keep the count under ten, because a matrix with forty rows is a specification document and nobody makes a decision from one. Include the status quo as a column. If you want the matrix to inform positioning rather than just record facts, add a weight column, since without weights every row implicitly counts the same and that is never true.
What is a 2x2 matrix in competitive analysis?
A two-by-two plots competitors on two axes to show clustering and open space, which is a positioning map rather than a scoring tool. It is excellent for one job: showing at a glance that four competitors sit in the same quadrant and nobody occupies another. It is poor at anything requiring more than two dimensions, and choosing axes that flatter your position is the standard way it gets misused. A differentiation matrix and a two-by-two answer different questions, and teams usually need both. Our positioning map tool builds the two-by-two.
What are the four quadrants of competitive analysis?
This usually refers to a two-by-two positioning map, where the four quadrants are whatever your two chosen axes produce, so there is no fixed set. It sometimes refers to SWOT's four boxes of strengths, weaknesses, opportunities and threats, which is a different exercise entirely. Occasionally it means the BCG growth-share matrix quadrants of stars, cash cows, question marks and dogs, which is a portfolio tool rather than a competitor tool. Worth checking which is meant before answering, since the three lead to unrelated pieces of work.
What are the 4 types of differentiation?
No canonical four-type model exists, and the lists in circulation disagree with each other. The distinction that is genuinely established in economics is between horizontal differentiation, where products differ in ways buyers rank differently according to taste, and vertical differentiation, where products differ in quality that buyers would rank the same way given equal prices. That two-way split is real and useful. The four-item lists mixing product, price, channel and service are practitioner groupings without an attributable origin, which does not make them useless, only not authoritative.
What are the 4 competitive strategies?
This points at Porter's generic strategies, which he presented as three: cost leadership, differentiation, and focus. Focus splits into cost focus and differentiation focus, which is why the model is frequently shown as four. Some textbooks, notably Thompson, Strickland and Gamble, extend it to five by adding a best-cost provider strategy. Porter's underlying argument matters more than the count: a firm that fails to commit to one of them ends up stuck in the middle, competing on everything and winning at nothing. That is the same warning as closing every gap in this matrix.
What are the 4 pillars of competitive advantage?
There is no canonical version, and the lists that circulate share no common origin. Some name leadership, operations, marketing and finance; others name entirely different things. The frameworks in this area with real attribution are worth using instead: Porter's generic strategies for how you choose to compete, Barney's resource-based view for what makes an advantage durable, and Hamilton Helmer's 7 Powers for a modern treatment of the same question. When a numbered list has no source, treat it as one author's structure rather than as established theory.
What is the difference between competition and competitive differentiation?
Competition is the situation you are in: other options exist and buyers are weighing them. Competitive differentiation is your answer to it, meaning the specific, weighted, hard-to-copy reasons a buyer picks you. The practical consequence is that you can face intense competition and have no differentiation, which is the position most crowded categories are in and which shows up as discounting. It also means differentiation is defined by the buyer's criteria rather than by your product's distinctiveness, since a difference nobody weighs is not a differentiator.
How do you create a matrix in Excel or Google Sheets?
Put criteria in column A, weights in column B, and one column per alternative including the status quo. Score each cell 1 to 4. Add a weighted column that multiplies the weight by the difference between your score and the row's best score, which is what ranks your gaps. Freeze the top row and the first column so the labels stay visible while scrolling. Conditional formatting on the weighted column makes the shape readable at a glance. Copy the template above and it pastes in with the columns intact, so the structure is already set up.
What is a comparative matrix?
Comparative matrix is a generic term for any grid comparing options against criteria, used well beyond competitive analysis in procurement, academic literature reviews and vendor selection. In a competitive context it means the same thing as a competitive comparison matrix. If someone hands you one, the two questions worth asking are where the criteria came from and whether they are weighted, since an unweighted comparative matrix implicitly treats a criterion that decides deals and one nobody mentions as equally important.
What is the difference between a differentiation matrix and a feature comparison?
A feature comparison records presence: who has what, usually across many rows, and it is the right artifact for answering an RFP or a specific capability question. A differentiation matrix records weighted judgement: given what this segment weighs, where do the meaningful differences sit, and what should we do about each one. Two products can look nearly identical on a feature grid and be completely different to use, which is the limitation the weighting and the 1 to 4 scoring exist to address. Our feature gap analysis template covers the capability-level view.
What is an example of a competitive differentiation matrix?
Segment: mid-market sales teams of 25 to 100 seats with no dedicated ops headcount, validated against 12 win/loss interviews. Criteria and weights: time to first useful output 0.20, total cost at real seat count 0.20, cost for view-only stakeholders 0.15, depth of automation 0.15, reporting and consolidation 0.15, migration effort 0.10, platform breadth 0.05. Scored against a named competitor and against staying on spreadsheets, the shape is clear: we lead the highest-weighted criterion by two points, we are behind by three on view-only seat cost which appeared in four of five losses, and we trail by one on both automation depth and reporting. Responses: neutralise the seat-cost gap by comparing totals at the buyer's real edit-versus-view split, concede platform breadth entirely at a weight of 0.05, and close only reporting, since it is the stated driver in acquisition-led deals. The most actionable finding costs nothing: we win the heaviest criterion and never lead with it.
What are the most common mistakes in a competitive differentiation matrix?
Six recur. Weighting criteria in an internal workshop, which reliably over-values what you are already good at. Omitting the status quo, which is the alternative that wins most often. Scoring your own product generously, usually by a full point. Reducing the matrix to one total per vendor, which conceals the shape that positioning is built from. Concluding that every gap should be closed, which produces a product that matches competitors everywhere and leads nowhere. And building one matrix for three segments, which scores none of them correctly and generates arguments that are really two segments disagreeing inside one document.
Keep your competitive differentiation matrix current
Flares tracks competitor capabilities, pricing and messaging automatically, so every score reflects what shipped this week.
Discover Flares14-day free trial · 30-second setup