Competitive strategy · 14 min read · Updated 2 Aug 2026
Competitor Switching Cost Analysis Template (Free)
A blank switching cost analysis you can fill in today, plus the guidance for what belongs in each field. It runs in both directions at once, because the friction stopping a competitor's customer reaching you is usually the same friction keeping yours in place, and most teams only measure one of them.
Copy pastes straight into Google Sheets or Excel with the columns intact. Downloads are free with a work email.
The switching cost analysis 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.
Analysis scope
Fill this in first. Name one competitor and one segment. Switching costs differ enormously by company size, so an all-segments answer describes nobody.
- Competitor analysedOne named competitor and the tier a buyer would actually be on
- SegmentSpecific, e.g. 25 to 100 seat teams with no dedicated ops headcount
- Direction of interestInbound, outbound, or both. Both is the useful answer
- The decision this informse.g. whether to fund migration tooling, or what to change about our own contract terms
- Evidence baseHow many real migrations you have data from, in each direction
- Owner and dateOne named owner, with the date and next review
1. Cost for their customer to switch to us
First row is an example, delete it. Estimate in days and money. Every row here is a reason a deal you should win does not close.
| Cost | What it actually involves | Time | Money or effort | Who inside the buyer bears it |
|---|---|---|---|---|
| ExampleMigrating historic deal and activity data | Export, field mapping, deduplication, then a validation pass | 3 to 10 working days | Mostly internal time, occasionally a contractor | Their ops lead, who did not ask for this project |
2. Cost for our customer to leave us
First row is an example, delete it. Fill this in as honestly as the table above, and compare the two totals when you are done.
| Cost | What it actually involves | Time | Money or effort | Who inside our customer bears it |
|---|---|---|---|---|
| ExampleRebuilding reporting and dashboards | Recreating saved views and any custom reporting logic | 2 to 5 working days | Internal time only | Whoever built the reports, usually one person |
3. Costs by type
First row is an example, delete it. Every cost from the two tables above goes in exactly one type. Expect the relational rows to be the hardest to write.
| Type | The specific cost | Direction | Severity | Evidence |
|---|---|---|---|---|
| ExampleRelational | Their admin built the current setup and is identified with it | Outbound, keeps our customers | High, and almost never discussed openly | Named in 3 of 6 renewal conversations |
4. What real migrations actually took
First row is an example, delete it. Only migrations that actually happened belong here. Leave the table empty rather than filling it with estimates.
| Account and direction | What it actually took | Elapsed days | What surprised us | Source |
|---|---|---|---|---|
| ExampleNorthwind, inbound from a competitor | Data import in a day, then three weeks of user habit change | 26 days to full use | The technical part was trivial; adoption was the whole cost | Onboarding notes and their ops lead |
5. Reducing the cost of switching to us
First row is an example, delete it. Order the rows so the barriers buyers name in real evaluations sit at the top.
| Barrier | How we reduce or absorb it | Cost to us | Expected effect | Owner |
|---|---|---|---|---|
| ExampleMigrating historic data | Run the import ourselves during the trial, at no charge | Roughly half a day of solutions time per deal | Removes the largest cited blocker in evaluations | Solutions team |
6. Retention we earn, not retention we trap
First row is an example, delete it. Two columns for the test: does this create value the customer would choose, or only cost they cannot avoid?
| Mechanism | Value it creates for the customer | Cost it creates if they leave | Would we defend this publicly? |
|---|---|---|---|
| ExampleDeep two-way integration with their helpdesk | Removes duplicate entry across two teams every day | Rebuilding the workflow elsewhere | Yes, the value came first and the cost is a consequence |
7. Where switching costs are a durable moat
First row is an example, delete it. Be sceptical. Most claimed moats are conveniences that a competitor could match in a quarter.
| Claimed moat | Why it is durable | What would erode it | Confidence |
|---|---|---|---|
| ExampleYears of historical activity data in one place | Cannot be recreated after the fact, only carried forward | A competitor building a high-fidelity importer for our format | Medium, an importer is a quarter of work for a motivated rival |
8. Decisions, owners and dates
First row is an example, delete it. Three to five rows. Split what reduces inbound friction from what earns outbound retention.
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| ExampleThe blocker buyers cite is adoption time, not data migration | Move the guided-setup offer into the evaluation, not post-sale | Sam L., Product | 3 Oct | Time from signup to team-wide use falls in new accounts |
How to fill in your switching cost analysis
How to scope a switching cost analysis
Name one competitor and one segment, and run both directions at once. Running both is the whole point: the friction stopping a competitor's customer from reaching you is usually the same friction keeping your own customers in place, and teams that examine only one direction reliably conclude that their own lock-in is healthy while the competitor's is unfair. Doing both in one document makes that inconsistency visible on the same page.
Competitor analysed
One named competitor and the tier a buyer would actually be on. Switching costs vary sharply by tier, since the customer with three integrations and the one with twelve are facing entirely different projects.
Segment
Specific, because size changes everything here. A ten-person team switches in an afternoon. A 400-seat organisation with custom fields, integrations and a compliance review needs two quarters, and averaging the two produces a number that describes neither.
Direction of interest
Inbound, outbound or both. Both is the useful answer. Inbound costs tell you why winnable deals stall; outbound costs tell you whether your retention is earned or merely structural.
The decision this informs
Usually whether to fund migration tooling, or what to change about your own terms and architecture. Say which, because the evidence you need differs: the first needs buyer-cited blockers, the second needs honest internal assessment.
Evidence base
How many real migrations you have data from, in each direction. If the answer is none, the analysis is a set of estimates and should be labelled as such until a real migration corrects it.
Owner and date
One named owner, and a review after any significant change to either product's integrations, data model or contract terms. Switching costs move when architecture moves, which is less often than pricing but with larger consequences.
How to record the cost of switching to you
Every row in this table is a reason a deal you should win does not close. This is the direction teams under-examine, because the costs fall on the buyer rather than on you and therefore never appear in any internal system. A competitor's customer who prefers your product and does not move is not a lost deal in your pipeline; they are a deal that never opened, which makes this the least visible and often the largest source of blocked growth.
Cost
Specific and concrete: "migrating historic deal and activity data", not "migration". Named costs can be attacked one at a time; a category cannot, which is why most migration programmes stall at the planning stage.
What it actually involves
The real steps: export, field mapping, deduplication, validation. Writing them out usually reveals that one step accounts for most of the pain, and that step is frequently not the one your team assumed.
Time
In working days, as a range. Elapsed time matters more than effort hours, because a project needing three days of work spread across six weeks of someone's attention is a six-week project to the person living it.
Money or effort
Distinguish cash from internal time. Internal time is the larger cost in most B2B switches and the one that kills projects, because nobody has budget for it and somebody has to do it alongside their actual job.
Who inside the buyer bears it
The most useful column here, and the one nobody fills in. The person paying the switching cost is often not the person who wanted the change, and an ops lead who did not ask for this project is a highly effective and entirely invisible blocker.
How to record the cost of leaving you
Same columns, uncomfortable answers. A low total here explains churn you have been attributing to price or to competitor features. It also tells you something more useful than either: if leaving you is easy, your retention depends entirely on the product remaining the best choice every single renewal, which is a fine position to be in but a very different one from what most teams assume they have.
Cost
The same honesty you applied to the inbound table. Teams consistently overestimate this direction, because the internal narrative that customers are deeply embedded is more comfortable than the alternative and nobody audits it.
What it actually involves
Concretely, from the customer's side. If you have never watched a departing customer leave, ask one who did. The answer is regularly that it took less time than anyone internally believed possible.
Time and money
Ranges, in working days. Compare the total against the inbound total on the previous table. If leaving you is materially easier than reaching you, you have a structural problem that no amount of competitive positioning will fix.
Who inside our customer bears it
Usually one person, often the one who built the current setup. That individual is your strongest retention asset and your largest single point of failure, and their departure from the company is one of the better predictors of churn.
Ask a churned customer what it actually took
This is the only reliable source for this table, and the exit conversation is where to get it. Internal estimates of how hard it is to leave you are the least reliable numbers in the entire document.
How to classify switching costs by type
Sort every cost into procedural, financial or relational. This taxonomy comes from Burnham, Frels and Mahajan's 2003 paper in the Journal of the Academy of Marketing Science, which is the properly attributable answer to a question that usually gets a made-up one. Procedural costs are lost time and effort. Financial costs are quantifiable resources given up. Relational costs are the psychological discomfort of breaking bonds and losing an identity, and they are the ones teams forget entirely.
Procedural
Time and effort: evaluation, learning, setup, data migration, rebuilding reports and integrations. These are the costs teams measure because they are visible and quotable, and they are frequently not the ones that decide the outcome.
Financial
Money given up: remaining contract term, lost prepaid balance, implementation fees, discounts tied to length, and the cost of running both systems in parallel. The parallel-running period is the one most often left out and it is rarely small.
Relational
The psychological and social cost: the admin who built the setup and is identified with it, the champion who advocated for the current tool and would have to explain the reversal, the relationships with your support team. Burnham and colleagues found these to be genuinely material rather than sentimental, and they explain switches that look irrational when only the first two types are counted.
Direction
Mark whether each cost blocks inbound or retains outbound. Many appear in both directions symmetrically, which is a useful thing to see written down, since it is the clearest possible evidence that your lock-in and your competitor's are the same mechanism.
Severity and evidence
High, medium or low, with where it came from. Relational costs are hardest to evidence and most often dismissed for that reason. A renewal conversation where someone defends a setup they personally built is evidence, and it should be recorded as such.
What drives how large these feel
Burnham and colleagues found perceived switching costs rise with product complexity, with how different providers are from one another, with how broadly the customer uses the product, and with how much switching experience the customer has. The last is worth noting because it runs the other way: a buyer who has migrated before finds the next one far less daunting.
How to gather evidence for a switching cost analysis
Measured, not estimated. One completed migration beats any amount of internal reasoning about how hard switching ought to be, and the finding is usually the same one: the technical part was smaller than expected and the human part was larger. Teams that have never watched a migration end to end consistently build migration tooling for the step that turns out not to matter, which is an expensive way to learn this.
Account and direction
Named internally, with which way they moved. Inbound migrations are the more useful evidence because you observed them; outbound ones require asking, which is worth doing and rarely done.
What it actually took
The real sequence and where the time went: "data import in a day, then three weeks of user habit change". This is the sentence that redirects a migration-tooling roadmap, and it usually contradicts the plan.
Elapsed days
From decision to full use, not from import to import complete. The gap between those two definitions is where the actual cost sits, and measuring the narrower one is how teams convince themselves switching is easy.
What surprised us
Keep this column. It is where the analysis learns something rather than confirming what was already believed, and a table with nothing surprising in it usually means nobody looked closely.
Source
Onboarding notes, the customer's own account, support tickets during the transition, exit conversations. Attribute it, because switching-cost estimates circulate internally for years after the migration they came from stopped being representative.
How to reduce the cost of switching to you
Rank by what buyers actually cite, not by what is easiest to build. This section is where a switching cost analysis pays for itself, because inbound friction is usually attacked with engineering when the highest-return moves are commercial or operational: absorbing the work, timing the switch to their renewal, or simply doing the migration for them during the evaluation. Those cost a day of someone's time and remove the blocker that engineering would spend a quarter on.
Barrier
Carry across the specific costs buyers name in evaluations, not the ones your team finds most interesting. If a barrier has never appeared in a real deal, reducing it is speculative work.
How we reduce or absorb it
Three broad options: remove the work through tooling, absorb it by doing it yourself, or make it unnecessary by changing what the customer has to bring across. Absorbing is under-used and frequently the cheapest per deal won.
Cost to us
Per deal, honestly. Half a day of solutions time per deal is sustainable at your current volume and might not be at ten times the volume, which is worth knowing before it becomes a promise in a sales motion.
Expected effect
Which specific blocker this removes, and in how many deals. Vague expected effects produce migration tooling nobody uses, which is the most common wasted investment in this whole area.
Time the offer to their contract, not to your quarter
The single highest-leverage move here and the cheapest. A buyer three months from renewal at their current vendor can act; one eighteen months in cannot, whatever you absorb. Knowing their renewal date is worth more than any migration feature.
How to tell earned retention from trapped retention
Two columns and one test: does this mechanism create value the customer would choose, or only cost they cannot avoid? Both retain customers and they are not equivalent. Value-based retention compounds, because the customer gets more embedded as they get more benefit. Cost-based retention without value degrades into resentment, shows up in reviews, and produces customers who leave the moment an alternative makes it easy, which a competitor will eventually do deliberately.
Mechanism
Integrations, accumulated data, workflow depth, training investment, contract terms, data portability. List them all, including the ones that exist by accident rather than by design.
Value it creates for the customer
Stated from their side. A deep integration removing duplicate entry every day is genuine value. If this column is empty for a row, you are looking at pure friction, and pure friction is a liability.
Cost it creates if they leave
The consequence of that value, which is legitimate when the value came first. Data accumulating because the product is used daily is earned; data being hard to export because export was never built is not the same thing.
Would we defend this publicly?
The clarifying question, and the reason this column exists. Anything you would be uncomfortable explaining to a customer, or seeing described accurately in a review, belongs on the fix list rather than in the retention strategy. Difficult exports, unclear data ownership and automatic long renewals all fail this test.
Earned lock-in survives a competitor attacking it
The practical argument, independent of the ethical one. A competitor can build an importer to neutralise a data barrier in a quarter. They cannot neutralise a workflow the customer actively values, because the customer would have to want to leave it, and they do not.
How to assess whether switching costs are a real moat
Be sceptical here, because this is the section where a useful analysis turns into a comfortable one. Most claimed moats are conveniences a motivated competitor could match in a quarter, and calling them moats produces exactly the complacency that lets it happen. The test is not whether switching is currently inconvenient. It is whether a well-resourced competitor who decided to attack this specific barrier could remove it, and how long that would take them.
Claimed moat
State it plainly: "years of historical activity data in one place". Vague moats like brand or relationships are usually not switching costs at all and belong in a different analysis.
Why it is durable
The mechanism that makes it hard to overcome. Things that accumulate over time and cannot be recreated retroactively are the strongest category. Things that are merely tedious are the weakest, because tedium is exactly what a competitor's engineering team can remove.
What would erode it
The most valuable column, and the one to write honestly: "a competitor building a high-fidelity importer for our format". Naming the specific attack tells you what to watch for, and it is a concrete thing to monitor rather than a vague concern.
Confidence
Low confidence is the correct default. A moat you have not seen tested is a hypothesis. If a competitor has already built an importer for a rival's format, assume yours is a quarter of work away and plan accordingly.
Switching costs are one of Porter's entry barriers
Worth knowing the lineage: Porter set switching costs out as a source of entry barriers in Competitive Strategy in 1980, alongside economies of scale, product differentiation, capital requirements, access to distribution channels, cost disadvantages independent of scale, and government policy. That is the frame most strategy readers will have, and the list is usually given as seven.
How to fill in the decisions section of your switching cost analysis
Three to five rows, splitting what reduces inbound friction from what earns outbound retention. These are different investments with different owners, and combining them is how the inbound half quietly disappears: reducing friction for a competitor's customers benefits no internal team's current numbers, so it loses to work with a nearer payoff unless somebody owns it explicitly with a date.
Finding
From the measured evidence rather than the estimates: "the blocker buyers cite is adoption time, not data migration". Findings contradicting the internal assumption are the valuable ones, and they are exactly the ones that get softened in review.
What we will do
One action per finding. Absorbing work and retiming an offer ship in weeks; tooling and architectural change take quarters. Both belong in the table so the fast ones are not deferred alongside the slow ones.
Owner
Named, and split by direction. Inbound reduction usually sits with product or solutions; outbound retention sits with product and whoever owns contract terms. An unowned inbound row is the default outcome, since nobody's quota depends on it.
By when
Real dates. Switching costs move when architecture and integrations move, so a decision left for two quarters may be solving a problem that has already changed shape.
How we will know it worked
Observable: "time from signup to team-wide use falls in new accounts". For inbound work, watch whether the barrier stops appearing as a stalling reason in evaluations, which is the direct measure of what you were trying to change.
Sourcing and upkeep: keeping a switching cost analysis honest
These rules apply to every section above. This artifact has a characteristic bias and it runs in one direction: teams overestimate how hard it is to leave them and underestimate how hard it is to reach them. Both errors point the same way, toward comfort, and together they produce a company that believes its retention is structural while its growth is blocked by friction nobody has measured. Every habit below corrects one half of that.
Run both directions in the same document
The core discipline. Examining only outbound produces a flattering account of your own lock-in; examining only inbound produces a grievance about the competitor's. Side by side, the symmetry is obvious and usually uncomfortable.
Measure at least one real migration before trusting any estimate
Internal reasoning about switching difficulty is consistently wrong in the same direction. One observed migration reorders the whole document, and the usual finding is that the technical work was small and the habit change was not.
Ask departing customers what leaving actually took
The only reliable source for the outbound table, available in the exit conversation, and almost never collected. Nobody inside your company knows how easy you are to leave.
Count relational costs
Procedural and financial costs are easy to list and are not usually decisive. The admin identified with the current setup and the champion who would have to explain a reversal decide more renewals than either, and Burnham, Frels and Mahajan's work is the citation if you need to argue the point.
Apply the public-defence test to every retention mechanism
If you would not be comfortable explaining it to a customer or seeing it accurately described in a review, it is a liability rather than a moat. Difficult exports and automatic long renewals fail this and eventually appear in your review-site themes.
Assume claimed moats are a quarter of work away
Low confidence is the right default. A barrier that is merely tedious is one motivated competitor away from disappearing, and treating it as durable is how the complacency it enables becomes the reason it falls.
Review when architecture changes, not on a calendar
Switching costs move with integrations, data models and contract terms rather than with quarters. A pricing change rarely affects them; a new export API or a competitor's importer changes them immediately.
A switching cost analysis example
You lead competitive strategy at Pipedrive. Deals against HubSpot Sales Hub keep stalling at the migration conversation, so you measure the friction in both directions rather than assuming it. This is that analysis, 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.
Analysis scopeIllustrative
| Field | Example entry |
|---|---|
| Competitor analysed | HubSpot Sales Hub, Professional tier, which is where our mid-market deals meet them |
| Segment | 25 to 100 seat sales teams with no dedicated ops headcount |
| Direction of interest | Both, deliberately, since we had only ever measured inbound |
| The decision this informs | Whether to fund a migration importer, and what our own export story should be |
| Evidence base | 4 inbound migrations observed, 2 outbound reconstructed from exit calls |
| Owner and date | Tom A., Competitive Intelligence, 2 Aug 2026, next review 1 Feb 2027 |
1. Cost for their customer to switch to usIllustrative
| Cost | What it actually involves | Time | Money or effort | Who inside the buyer bears it |
|---|---|---|---|---|
| Migrating historic deal and activity data | Export, field mapping, deduplication, then a validation pass | 3 to 10 working days | Internal time, occasionally a contractor | Their ops lead, who did not ask for this project |
| Rebuilding automation and sequences | Recreating each workflow by hand, since nothing transfers between vendors | 2 to 6 working days | Internal time | Whoever originally built them, usually one person |
| Retraining the team | New interface, new habits, a period of lower activity logging | 3 to 5 weeks to full use | Lost productivity rather than cash | Every rep, and the manager who absorbs the dip |
| Remaining contract term | Paying two vendors until their term ends | Up to 12 months | Real cash, and the single largest financial item | Their finance lead |
| Reconnecting the wider stack | Reauthorising each integration and re-testing the data flow | 1 to 3 working days | Internal time | Their ops lead again |
2. Cost for our customer to leave usIllustrative
| Cost | What it actually involves | Time | Money or effort | Who inside our customer bears it |
|---|---|---|---|---|
| Rebuilding reporting and dashboards | Recreating saved views and custom reporting logic | 2 to 5 working days | Internal time only | Whoever built the reports, usually one person |
| Exporting historic data | Standard export, no obstruction, formats documented | Under a day | None | Their ops lead |
| Retraining the team | Symmetrical with the inbound case, and just as large | 3 to 5 weeks to full use | Lost productivity | Every rep |
| Remaining contract term | Annual terms, so between zero and twelve months depending on timing | Up to 12 months | Real cash | Their finance lead |
3. Costs by typeIllustrative
| Type | The specific cost | Direction | Severity | Evidence |
|---|---|---|---|---|
| Procedural | Rebuilding automation and reporting by hand | Both directions, roughly symmetrical | Medium, tedious rather than blocking | 4 observed inbound migrations |
| Procedural | Team retraining and the activity dip during it | Both directions | High, and consistently underestimated by everyone | Onboarding data from 4 migrations |
| Financial | Remaining contract term at the incumbent | Both directions | High, and the only true blocker mid-term | Named in 6 of 9 stalled evaluations |
| Financial | Onboarding fees, which both vendors list separately from seat price | Both directions | Medium, and visible on both published pricing pages | Both pricing pages, read 2 Aug 2026 |
| Relational | Their admin built the current setup and is identified with it | Outbound, keeps our customers | High, and almost never discussed openly | Named in 3 of 6 renewal conversations |
| Relational | The champion who chose the incumbent would have to explain a reversal | Inbound, blocks us | High, and invisible in any pipeline report | Reconstructed from 2 lost evaluations |
4. What real migrations actually tookIllustrative
| Account and direction | What it actually took | Elapsed days | What surprised us | Source |
|---|---|---|---|---|
| Northwind, inbound from a competitor | Data import in a day, then three weeks of user habit change | 26 days to full use | The technical part was trivial; adoption was the whole cost | Onboarding notes and their ops lead |
| Bellweather, inbound | Stalled 7 months waiting for the incumbent contract to end | 212 days from decision to start | Timing, not difficulty, was the entire delay | Deal notes |
| Calder & Sons, outbound | Exported and left over a weekend, reporting rebuilt the following week | 9 days to full use elsewhere | Leaving us was materially faster than reaching us | Exit conversation, Feb 2026 |
5. Reducing the cost of switching to usIllustrative
| Barrier | How we reduce or absorb it | Cost to us | Expected effect | Owner |
|---|---|---|---|---|
| Remaining contract term at the incumbent | Track renewal dates in the CRM and time the approach to three months before | No build cost, process change only | Addresses the blocker named in 6 of 9 stalled evaluations | Dana K., Sales |
| Team retraining and the activity dip | Move guided setup into the evaluation rather than after signature | Roughly a day of solutions time per deal | Attacks the largest measured cost rather than the assumed one | Sam L., Product |
| Migrating historic data | Run the import ourselves during the trial, at no charge | Half a day of solutions time per deal | Removes a cited blocker, though not the largest one | Solutions team |
| Rebuilding automation | Do not build an importer this year | Nothing | Evidence says this is tedious rather than blocking | Sam L., Product |
6. Retention we earn, not retention we trapIllustrative
| Mechanism | Value it creates for the customer | Cost it creates if they leave | Would we defend this publicly? |
|---|---|---|---|
| Deep two-way integration with their helpdesk | Removes duplicate entry across two teams every day | Rebuilding the workflow elsewhere | Yes, the value came first and the cost is a consequence |
| Accumulated historic activity data | Multi-year reporting nobody could reconstruct after the fact | Carried forward but not recreated | Yes, and our export is documented and unobstructed |
| Annual contract terms | A lower rate than monthly billing | Up to 12 months of overlap if they leave mid-term | Yes, provided renewals are not automatic and silent |
| The admin identified with the setup they built | Real, they get a system shaped to their team | Personal and social rather than technical | Yes, but it is fragile: it leaves when they do |
7. Where switching costs are a durable moatIllustrative
| Claimed moat | Why it is durable | What would erode it | Confidence |
|---|---|---|---|
| Years of historical activity data in one place | Cannot be recreated after the fact, only carried forward | A competitor building a high-fidelity importer for our format | Medium, an importer is about a quarter of work for a motivated rival |
| Deep helpdesk integration | The customer actively values it, so removing the cost does not make them want to leave | The helpdesk vendor building an equivalent integration with a competitor | Medium to high, and the most durable item here |
| Difficulty of rebuilding automation | It is not durable, and calling it a moat would be self-deception | Any competitor deciding to build a workflow importer | Low, this is tedium rather than a barrier |
8. Decisions, owners and datesIllustrative
| Finding | What we will do | Owner | By when | How we will know it worked |
|---|---|---|---|---|
| The measured blocker is adoption time and contract timing, not data migration | Move guided setup into the evaluation and stop scoping the importer | Sam L., Product | 3 Oct 2026 | Time from signup to team-wide use falls in new accounts |
| Leaving us took 9 days where reaching us took 26 | Treat retention as earned rather than structural and brief the leadership team accordingly | Tom A., Competitive Intelligence | 29 Aug 2026 | The asymmetry is stated in the next QBR rather than assumed away |
| Contract timing decided 6 of 9 stalled evaluations | Capture incumbent renewal dates in discovery and work the calendar | Dana K., Sales | 29 Aug 2026 | Renewal date is populated on new opportunities |
| One retention mechanism depends on a single person at the customer | Flag accounts where only one admin has built anything, as a churn risk signal | Tom A., Competitive Intelligence | 1 Feb 2027 | Admin departure appears as a tracked risk rather than a surprise |
How to roll out your switching cost analysis
- 1Copy or download the blank analysis. 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 competitor and one segment. Switching costs differ enormously between a ten-person team and a 400-seat organisation, so an all-segments answer describes nobody.
- 3Delete the example rows. Each table ships with one example row so the pattern is obvious. Remove it before you circulate the analysis.
- 4Fill in both directions before drawing any conclusion. What it costs their customer to reach you, and what it costs yours to leave. The comparison is the finding.
- 5Sort every cost into procedural, financial or relational. The third is the one teams forget: the admin who built the setup, the champion who would have to explain a reversal.
- 6Replace estimates with one real migration. Measured elapsed days from decision to full use. The technical part is usually smaller than assumed and the habit change larger.
- 7Test each retention mechanism against public defence. Would you be comfortable explaining it to a customer? Anything that fails belongs on the fix list, not in the retention strategy.
- 8Time your migration offer to the buyer's renewal. A buyer three months from renewal can act; one eighteen months in cannot, whatever you absorb for them.
Switching cost analysis FAQ
What is switching cost analysis?
Switching cost analysis measures what it actually costs a customer to move between providers, in both directions: what a competitor's customer would spend in time, money and disruption to reach you, and what your own customer would spend to leave. Running both directions is the point, because the friction blocking your growth is usually the same mechanism protecting your retention, and teams that measure only one conclude that their own lock-in is healthy while the competitor's is unfair.
What is a switching cost?
A switching cost is anything a customer gives up or has to do in order to change providers, beyond the price of the new product. It covers migrating data, rebuilding configurations, retraining people, paying out a remaining contract, running two systems in parallel, and the social cost of reversing a decision someone advocated for. The defining feature is that it is paid by the customer rather than by either vendor, which is why it rarely appears in any internal system and is so consistently underestimated.
What is a switching cost example?
A concrete B2B software example: a 60-seat team moving CRM spends three to ten days exporting, mapping and deduplicating historic data, two to six days rebuilding automation that does not transfer between vendors, three to five weeks reaching full team adoption with an activity dip throughout, and up to twelve months paying out the remaining term at the incumbent. In practice the last item blocks more deals than all the others combined, because it is the only one that cannot be reduced by effort. In consumer terms, the equivalent is porting a phone number or moving a bank's direct debits.
What are the three types of switching costs?
Procedural, financial and relational. This taxonomy comes from Burnham, Frels and Mahajan's paper in the Journal of the Academy of Marketing Science in 2003, which is worth citing because most numbered frameworks in this area have no source and this one does. Procedural costs are lost time and effort: evaluation, learning, setup and migration. Financial costs are quantifiable resources given up: remaining contract, lost discounts, parallel running. Relational costs are the psychological discomfort of breaking bonds and losing an identity, and they are both the least measured and frequently the most decisive.
What are the two kinds of consumer switching costs?
The two-way splits in circulation vary, with the most common being transactional costs paid once at the moment of switching versus learning costs paid gradually as the customer becomes proficient. Another common split is financial versus non-financial. Neither has the standing of the three-type taxonomy from Burnham, Frels and Mahajan, which was built specifically because the earlier two-way divisions were too coarse to guide anything. If you need a framework to actually classify costs with, use the three types.
How do you measure switching costs?
Measure one real migration rather than estimating. Record elapsed days from decision to full use, not from import start to import finish, because the gap between those definitions is where nearly all the cost sits. Separate cash from internal time, since internal time is larger in most B2B switches and has no budget line. Then ask a departing customer what leaving actually took, which is the only reliable source for the outbound direction and is almost never collected. Internal estimates of switching difficulty are consistently wrong in the same direction.
What does a high switching cost mean?
For the incumbent it means retention that does not depend on winning every renewal on merit, which is comfortable and can mask a deteriorating product. For a challenger it means winnable deals that never open, since a customer who prefers your product and cannot justify the move is not visible anywhere in your pipeline. For the customer it means reduced bargaining power, which is why high switching costs eventually attract both competitor attacks and, in some markets, regulatory attention around data portability.
What do low switching costs mean?
That your retention is earned every renewal rather than structural. This is a more demanding position and a more honest one: the product has to remain the best choice continuously, because nothing else is holding the customer. It also cuts the other way and favourably, since low switching costs across a category make competitors' customers reachable too. The dangerous case is asymmetry, where leaving you is easy and reaching you is hard. That combination shows up as churn attributed to price and growth attributed to competition, when the real cause is friction.
Who benefits from switching costs?
Incumbents with market share benefit most, since switching costs convert an existing customer base into recurring revenue that competitors must overcome rather than merely outperform. New entrants are harmed, which is exactly why Porter treated switching costs as a barrier to entry. Customers are generally disadvantaged, because reduced ability to leave weakens their negotiating position. The nuance worth holding is that some switching costs are a by-product of genuine value, such as accumulated data from daily use, and those benefit both sides.
Is switching cost a barrier to entry?
Yes, and it is one of the classic ones. Porter set switching costs out as a source of entry barriers in Competitive Strategy in 1980, alongside economies of scale, product differentiation, capital requirements, access to distribution channels, cost disadvantages independent of scale, and government policy. The mechanism is that an entrant must not merely be better, it must be better by more than the cost of moving, which is a substantially higher bar and the reason many technically superior products fail to displace incumbents.
What are the five barriers to entry?
Porter's own treatment is usually given as seven rather than five: economies of scale, product differentiation, capital requirements, switching costs, access to distribution channels, cost disadvantages independent of scale, and government policy. Five-item versions in circulation are abridgements that drop whichever two the author found least relevant, and they do not correspond to a distinct model. If you want the canonical list, use Competitive Strategy from 1980, and note that switching costs appear in it explicitly.
How do high switching costs serve as a barrier to entry?
By raising the threshold a new entrant must clear. If moving costs a customer roughly three weeks of disrupted productivity plus a contract payout, the entrant has to be better by more than that, not simply better. This is why entrants frequently attack switching costs directly rather than competing on capability: building high-fidelity importers, absorbing migration work, or offering to cover the remaining term. Those moves are usually cheaper than out-featuring the incumbent and considerably more effective.
What is a moat in switching costs?
Moat is investing language, popularised by Warren Buffett and formalised by Morningstar, for a durable structural advantage protecting returns, and switching costs are one recognised source of one. The important word is durable. Most claimed switching-cost moats are conveniences a motivated competitor could remove in a quarter by building an importer, which makes them a head start rather than a moat. The genuinely durable kind involves something that accumulates over time and cannot be recreated retroactively, or value the customer actively wants and would not abandon even if leaving were free.
What is the difference between a switching cost and a network effect?
A switching cost makes leaving expensive for the individual customer, and it exists whether or not anyone else uses the product. A network effect makes the product more valuable as more people use it, so the pull comes from other users rather than from the difficulty of exit. They often appear together and reinforce each other, but they behave differently under attack: a competitor can neutralise a switching cost unilaterally by absorbing the work, whereas neutralising a network effect requires moving other people, which is much harder. Network effects are the stronger position.
What are switching costs in strategic management?
In strategic management, switching costs appear in two places. In Porter's five forces they shape the threat of new entrants and the bargaining power of buyers, since a buyer who cannot leave has less leverage. In the resource-based view they are one mechanism by which an advantage becomes durable rather than temporary. The practical implication in both frames is the same: switching costs determine how much of your market position is defended by structure rather than by continuing to be the better product.
How would you implement a switching cost?
The framing of this question is worth pushing back on, because deliberately engineering friction and building value that happens to be hard to replace are different strategies with different outcomes. Value-based retention compounds, since the customer becomes more embedded as they get more benefit. Friction-based retention without value degrades into resentment, appears in your review-site themes, and collapses the moment a competitor makes leaving easy. A useful test: would you be comfortable explaining the mechanism to a customer, or seeing it described accurately in a public review? Difficult exports and silent automatic renewals fail that test.
What are the 4 types of costs?
This query belongs to accounting rather than to competitive strategy, and it usually means fixed, variable, direct and indirect costs. That is a different taxonomy from switching costs entirely and the two get mixed in search results. For switching costs specifically, the established classification is the three-type model from Burnham, Frels and Mahajan: procedural, financial and relational. If your question is how to budget a migration, the accounting categories are relevant; if it is why customers do or do not move, the three-type model is the one that explains behaviour.
How do you reduce switching costs for a competitor's customers?
Three levers, in rough order of return. Time the approach to their contract renewal, which costs nothing and addresses the one barrier effort cannot overcome. Absorb the work by running the migration yourself during the evaluation, which is usually a day of solutions time and removes the most-cited blocker. Then, only if the evidence supports it, build tooling. Teams reliably do these in reverse order, funding an importer for a step that turns out to be tedious rather than blocking, while never recording the incumbent's renewal date in discovery.
What is the difference between switching cost analysis and win-back analysis?
Switching cost analysis measures friction: what moving costs, in both directions, for customers in general. Win-back analysis examines specific customers who already left and asks which are recoverable. They connect at exactly one point, and it is the most practically useful one in either document: a former customer's contract at their new vendor is a switching cost, so the renewal date determines when a win-back attempt can succeed. Our win-back analysis template covers the account-level exercise.
What is an example of a switching cost analysis?
Scope: one named competitor at the tier mid-market deals meet them, 25 to 100 seat teams, both directions, with four inbound migrations observed and two outbound reconstructed from exit calls. Inbound costs: three to ten days of data migration, two to six days rebuilding automation, three to five weeks to full team adoption, and up to twelve months of remaining contract term. Outbound costs: reporting rebuilt in two to five days, export in under a day, the same adoption period, the same contract exposure. By type: retraining and contract term are the heavy items, while the relational costs, an admin identified with the setup they built and a champion who would have to explain a reversal, appear on both sides and are invisible in any pipeline report. Measured evidence: one inbound migration took 26 days to full use with the import itself taking one day, one stalled 212 days purely waiting for a contract to end, and one outbound departure took 9 days. The conclusion contradicts the assumption: the blocker is adoption time and contract timing, not data migration, so the importer is not funded and guided setup moves into the evaluation instead.
What are the most common mistakes in a switching cost analysis?
Five recur. Measuring only one direction, which produces a flattering account of your own lock-in or a grievance about the competitor's. Estimating rather than measuring, when internal reasoning about switching difficulty is consistently wrong in the same direction. Omitting relational costs, which decide more renewals than procedural ones. Treating tedium as a moat, when anything merely tedious is one motivated competitor away from disappearing. And building migration tooling for the step nobody named, which is the most common wasted investment in this area and is avoided entirely by watching one real migration first.
Switching cost analysis on live competitor data
Flares tracks competitor integrations, packaging and terms automatically, so you can see what switching really involves today.
Discover Flares14-day free trial · 30-second setup