Skip to content
Faisal Mahmud TusherDigital marketing, SEO & analytics
with Faisal Tusher

Comparisons

Marketing Funnel vs Flywheel

The marketing funnel and the flywheel compared as planning models: stages ending at purchase against a loop where customers feed growth, and when each helps.
Comparisonbeginner8 min readUpdated

Marketing funnel and flywheel at a glance

The funnel describes a population narrowing as it moves from awareness to purchase. The guide to the marketing funnel, explained covers its stages and the metrics at each one, and the marketing funnel framework treats it as a counting system: define each stage as a state, count who is in it, and measure the rate between stages so that a vague sense that something is wrong becomes a located constraint. Its strength is precision. Its well-known weakness is that it ends at the purchase and treats the customer as output; whatever customers do afterwards sits outside the model or in a retention stage added at the bottom.

The flywheel takes the opposite view. The metaphor comes from Jim Collins, whose summary of the flywheel effect describes a heavy wheel turned one push at a time in the same direction until momentum does much of the work. Marketing borrowed it to describe the customer relationship as a cycle: attract, engage and delight, where a delighted customer feeds attract by returning, recommending and producing Social Proof. The flywheel framework covers its stages and the vocabulary of force and friction. Its strength is that it makes what happens after the sale part of growth. Its weakness is that momentum is not a metric, and the circle can blur the hard, sequential work of turning a stranger into a first-time buyer.

The difference that matters for planning is where each model says growth comes from. Under a funnel, new customers come from filling the top, so growth tends to follow spend. Under a flywheel, a growing share of new customers comes from existing ones, so growth depends on how well the business keeps and delights the customers it already has. Which picture is closer to the truth varies by business, and that is what decides which model should lead.

Compare table

Marketing funnel vs Flywheel

Row labelMarketing funnelFlywheel
ShapeA sequence of stages that narrows towards a purchaseA cycle whose output feeds its own input
End pointThe purchase, with retention often added at the bottomNone; a delighted customer re-enters the cycle
Role of customersOutput of the processA source of force: repeat purchases, referrals and reviews
MetricsCounts at each stage and the conversion rates between themSlow proxies: repeat or renewal rate, referral share, churn, acquisition cost trend
Team alignmentClear hand-offs between marketing and sales stage by stageShared ownership across marketing, product, sales and service
Budget logicSpend at the top to fill the funnel; fix leaks lower downAdd force where it compounds and remove friction where it drains momentum
Diagnostic useStrong: locates the stage where prospects are lostWeak for single campaigns; strong for spotting friction after the sale
Best-fit businessConsidered or infrequent purchases, long sales cycles, acquisition-led growthFrequent purchase, subscriptions and categories where people recommend and review

The rows on metrics and diagnostic use explain why the two coexist. The funnel produces fast, precise numbers that can judge a campaign this month. The flywheel's measures move slowly and judge the business over quarters. A team that tries to run weekly optimisation from the flywheel has nothing to steer by, and a team that sets strategy from the funnel alone tends to treat retention as somebody else's problem.

When to choose the marketing funnel

Lead with the funnel when the planning question is about acquisition. Building a campaign, choosing what each piece of content or ad is for, setting stage-to-stage targets and finding why a launch is underperforming are all funnel questions. The model gives every stage a job and a count, which makes it easy to see whether the problem is too few people at the top, too many lost in consideration or too few converting at the end.

The funnel also fits businesses where customers buy rarely or through a long, formal process. A purchase made once every few years, or one that runs through procurement and a buying committee, stores little momentum between sales, and the work that matters is moving each prospect through the stages. Here the funnel's sequence matches reality closely enough to plan with, and the flywheel adds little beyond a reminder to collect reviews.

Use it with its limits in mind. People do not move in a straight line; they skip stages and return months later. Defining stages as states rather than steps keeps the counts valid, but the funnel will not explain why people behave as they do, and it will not show the value of customers once they have bought.

When to choose the flywheel

Lead with the flywheel when existing customers drive a large share of growth. Subscriptions, frequently bought products, marketplaces and categories that people talk about all fit: retention, repeat purchase, referrals and reviews feed new demand, so a fall in delight shows up later as dearer acquisition. Here the flywheel's main contribution is to make visible a link the funnel hides, between how the business treats customers after the sale and what it costs to win the next one.

It is also the better model for aligning teams. Because each stage's failure lands in another stage, the flywheel gives marketing, product, sales and service a shared picture of how their work connects. Rising Churn Rate or a slow fall in LTV becomes a question for everyone rather than for the retention team alone, and the biggest friction often turns out to sit outside marketing's control.

Do not expect it to run campaigns. Momentum has to be approximated with proxies that move over quarters, so the flywheel cannot judge a single month or a single ad. And it is easy to adopt as a slogan; declaring a flywheel changes nothing unless the forces and frictions are listed, owned and measured. Where the question is how a specific loop, such as a referral programme, turns one customer into the next, the growth loops in practice guide is the more precise tool.

Illustrative example — not a real result

A meal-kit company moves its planning question

Brindlecombe, a meal-kit subscription, plans entirely with a funnel: ad impressions, site visits, trial boxes and first paid orders. Its funnel rates are steady, but cost per new subscriber has risen from £38 to £52 in a year. Mapping the business as a flywheel shows why. Referrals fell from about a quarter of new subscribers to about a tenth after the referral reward was moved to the third paid box, and churn after the first month rose from 30% to 36% when delivery slots were cut. The company keeps the funnel for weekly campaign reporting, but sets the next two quarters' priorities from the flywheel: restore delivery choice and move the reward back to the first box, with referral share and first-month churn as the measures.

Decision table

SituationLead withWhy
Planning a campaign or a launchMarketing funnelEach stage gets a job, a count and a target
Finding where prospects are lost before a first purchaseMarketing funnelStage-to-stage rates locate the constraint
Acquisition cost rising while funnel rates hold steadyFlywheelThe cause may sit after the sale, in retention or referral
Subscription or frequent-purchase businessFlywheel for strategy, funnel for campaignsExisting customers feed a large share of growth
Infrequent purchase through formal procurementMarketing funnelLittle momentum is stored between purchases
Aligning marketing, product and service on growthFlywheelIt shows how each team's failure lands elsewhere
Choosing between a self-serve and a sales-assisted motionSee product-led versus sales-led growthA broader decision about who drives the sale

Using both models in one plan

The simplest arrangement gives each model the job it does best. The flywheel sits at the top of the plan, as the picture of how the business grows: which forces the year will add, which frictions it will remove, and which slow measures will show whether the wheel is turning faster. The funnel sits underneath, as the measurement system for the acquisition steps inside the attract and engage stages, with its stage definitions and conversion rates reported every week or month.

The two meet at the bottom of the funnel and the delight stage of the wheel. A funnel that ends at the first purchase should be extended with the measures the flywheel cares about, such as repeat purchase, renewal and the share of new customers who arrived by referral, so that the weekly report shows whether acquired customers are staying. And the flywheel's priorities should be translated into funnel terms where they touch acquisition: a referral programme is a force on the wheel, but its effect shows up as a new source at the top of the funnel and as a higher conversion rate for referred prospects.

Budgets follow the same split. Funnel logic decides how acquisition spend is distributed across stages and channels. Flywheel logic decides how much goes to removing friction after the sale, which is often cheaper than buying more traffic and harder to justify on a campaign report alone.

Common mistakes

  • Declaring the funnel obsolete. The flywheel replaces the funnel's picture of where customers come from, not its counting.
  • Running weekly optimisation from the flywheel. Its measures move over quarters and cannot judge a single campaign.
  • Ending the funnel at the first purchase. Without repeat and referral measures, retention problems stay invisible in campaign reports.
  • Drawing a flywheel without owners. Forces and frictions that nobody owns or measures are a diagram, not a plan.
  • Assuming delight produces referrals in every category. Some products are rarely talked about, and their wheels store little momentum.
  • Adding forces when friction is the problem. More campaigns cannot outrun a broken checkout or a slow support queue.
  • Treating the two models as a choice of identity. They answer different questions, and the plan needs both answers.

Frequently asked questions

Has the flywheel replaced the marketing funnel?

No. The flywheel changed the picture of where customers come from, but it did not replace the funnel's counting. A team still needs to know how many prospects reach each stage before a first purchase and where they drop out, and the funnel is the clearest way to do that. The flywheel adds what the funnel leaves out: what customers do after they buy and how that feeds new demand. Most teams use both.

Which model suits a business that sells something bought once?

Usually the funnel. A flywheel depends on customers coming back, recommending or producing visible proof that brings in others, and a product bought once a decade or through formal procurement stores little of that momentum between purchases. Reviews and referrals can still matter in such categories, so a light flywheel view of the delight stage can be useful, but the planning and measurement belong to the funnel.

What do I measure if I adopt the flywheel?

Momentum is a metaphor, so it has to be approximated with proxies that move slowly: repeat purchase or renewal, the share of new customers who came through referral or reviews, churn, and the trend in the cost of acquiring each new customer. Each push the team adds should be paired with the friction it is meant to remove and a measure for that pair. The funnel's stage-to-stage rates remain the measures for the acquisition steps inside the wheel.

Terms used on this page

Frameworks

Where this page is used

Also related

Sources

  1. The Flywheel Effect — Jim Collins

Related