How to Optimise Betting Funnels for Better Value

How to Optimise Betting Funnels for Better Value

A betting funnel rarely fails because one advert, landing page or welcome offer is obviously poor.

More often, commercial value leaks between stages.

The acquisition message attracts one type of intent, but the landing page presents something different. Registration creates unnecessary friction. Verification is confusing. A player reaches the payment stage but cannot use their preferred method. First-deposit messaging arrives too late. CRM then places every new player into the same generic journey regardless of how they arrived or what they have already done.

Each individual problem may appear relatively small.

Together, they can turn an apparently efficient acquisition campaign into an expensive source of weak player value.

That is why betting funnel optimisation should not be treated as a collection of isolated conversion-rate tactics. Operators need to manage the journey as one commercial system, connecting acquisition, landing pages, registration, verification, payments, first product use, CRM and retention.

The objective is not simply to increase the percentage of users completing the next step.

It is to increase the number of compliant, engaged and commercially valuable players progressing through the funnel.

In short: Betting funnel optimisation means identifying where value is lost between acquisition and long-term player activity, then improving the journey without sacrificing player quality or appropriate controls. Operators should align acquisition intent with landing pages, reduce avoidable registration and payment friction, assess offers on downstream value, connect acquisition data to CRM and measure conversion at every meaningful hand-off.

What is a betting funnel?

A betting funnel describes the stages a potential customer moves through from initial exposure to becoming an active player.

A simplified sportsbook funnel might look like:

Impression
→ click
→ landing-page engagement
→ registration start
→ registration completion
→ verification
→ first deposit
→ first wager
→ repeat activity
→ retention

Not every player will move through the journey in exactly the same order.

Different channels also create different starting points.

Someone searching for a sportsbook immediately before a major football match may already have strong intent.

Someone watching a paid social video may still be comparing brands or learning about the product.

An affiliate visitor may arrive after reviewing several operators.

The funnel therefore needs to reflect both the player's stage and the acquisition context.

Betting funnel optimisation is not just conversion-rate optimisation

Traditional conversion-rate optimisation often focuses on increasing the percentage of users completing an action.

That matters.

But in betting, the quality of the resulting player matters just as much.

Suppose an operator changes a landing page and improves registration conversion from 12% to 16%.

On the surface, the test is successful.

But the newly acquired cohort then shows:

• lower verification completion;

• weaker first-deposit conversion;

• greater bonus dependency;

• poorer Day 30 retention.

The change increased registrations.

It may not have improved the funnel.

Betting funnel optimisation therefore requires a broader question:

Did the change create more commercially useful players?

Start with player-quality economics

Before redesigning a landing page or changing an offer, define what a valuable acquisition actually means.

First-time depositors are important.

They are not all equal.

A stronger acquisition framework can consider factors such as:

• verification completion;

• deposit behaviour;

• product engagement;

• second deposit;

• early retention;

• early net gaming revenue;

• bonus cost;

• relevant responsible gambling indicators.

The exact measures will vary by operator, market and product.

The important point is that funnel optimisation should work backwards from value rather than forwards from clicks.

Treat every acquisition channel differently

Different channels naturally produce different player profiles.

Paid search

Paid search can capture users with relatively high intent.

Someone searching for a sportsbook before a major event may already know what they want.

This can produce:

• stronger registration conversion;

• faster time to deposit;

• lower need for product education.

Paid social

Paid social can create demand earlier in the decision process.

These players may require:

• more context;

• clearer product explanation;

• stronger continuity between creative and landing page.

Affiliates

Affiliate traffic can arrive with strong intent, particularly from high-value comparison content.

Quality can still vary significantly by:

• publisher;

• placement;

• market;

• promotional mechanic.

Treating all FTDs as interchangeable hides those differences.

Build betting funnel reporting around cohorts

Campaign-level performance is useful.

Cohort reporting makes it more commercially meaningful.

Group players according to factors such as:

• acquisition week;

• source;

• campaign;

• creative;

• market;

• device;

• offer.

Then compare their behaviour at consistent maturity points.

Useful checkpoints may include:

• Day 7;

• Day 30;

• Day 90.

For example:

CohortFTDsDay 30 retained90-day contributionPaid Search A600228£44,000Paid Social B820205£31,000Affiliate C470216£48,000

Paid Social B creates the most first-time depositors.

Affiliate C produces the strongest retained commercial value in this hypothetical example.

That is the type of information funnel optimisation should feed into future acquisition decisions.

Optimise betting funnels by matching intent to journey

One of the biggest sources of funnel leakage is a mismatch between the acquisition message and the experience after the click.

Consider two users.

User A

Searches for football betting immediately before a major fixture.

User B

Engages with a paid social video introducing a sportsbook product.

They are at different stages of the decision process.

Sending both to an identical generic landing page ignores that difference.

Preserve message continuity after the click

The strongest landing pages continue the proposition introduced in the acquisition channel.

If the advert focuses on football betting, the destination can continue that context with relevant product information and a clear route forward.

If the campaign focuses on a welcome offer, the user should be able to understand the proposition before beginning registration.

If an affiliate recommendation discusses a particular sportsbook strength, landing on an unrelated generic page creates friction.

Continuity supports both:

• clarity;

• conversion.

It also reduces the sense that the user clicked one proposition and received another.

Do not build hundreds of landing pages unnecessarily

Intent-based routing does not mean creating a separate page for every keyword or advert.

That quickly becomes operationally unmanageable.

Instead, identify the major intent groups that genuinely warrant a different experience.

These might include:

• brand search;

• generic sportsbook search;

• major-event traffic;

• affiliate referrals;

• paid social prospecting.

Each route can then be tested around the areas most likely to influence the decision.

Test landing-page message hierarchy

The order of information matters.

Useful tests might compare:

• product benefit first versus offer first;

• event context versus generic sportsbook positioning;

• product proof versus promotional emphasis;

• different registration prompts.

The strongest result should be evaluated beyond page conversion.

If one version creates more registrations but weaker depositors, the funnel may not have improved commercially.

Use acquisition parameters intelligently

Campaign parameters and source data can help route users into the most relevant experience.

Useful context may include:

• channel;

• campaign;

• creative theme;

• market;

• offer.

That does not mean exposing every piece of tracked data to the player.

It means using known context to avoid restarting the journey from zero after the click.

Reduce betting registration friction without reducing control

Registration is one of the most common points where otherwise efficient acquisition funnels lose momentum.

Potential issues include:

• long forms;

• confusing validation;

• difficult mobile inputs;

• unclear password requirements;

• late-stage eligibility failures;

• repeated data entry.

But betting operators cannot simply remove required controls.

Depending on the jurisdiction and product, registration can involve important processes around:

• age verification;

• KYC;

• affordability;

• relevant disclosures.

The objective is therefore not minimal registration at any cost.

It is to make necessary steps as clear and usable as possible.

Explain why information is required

Users are more likely to experience friction when a form unexpectedly asks for information without explaining why.

Where appropriate, short explanations can make the journey clearer.

The principle is simple:

If a required step creates friction, reduce uncertainty around it.

That may involve:

• clearer field labels;

• useful error messages;

• visible progress;

• explanation of verification steps.

Use progressive disclosure where appropriate

Not every piece of information needs to appear simultaneously.

Where the relevant requirements and implementation permit, forms can reveal information in logical stages.

This can reduce visual overload without removing necessary steps.

The objective should be a controlled sequence rather than hiding material requirements.

Retain form data after errors

Few experiences are more frustrating than completing a long registration form, encountering one validation problem and losing previously entered information.

Where technically appropriate, retaining valid information can reduce unnecessary abandonment.

Likewise, errors should explain:

• what went wrong;

• what needs changing.

“Invalid entry” is less useful than a specific explanation.

Prioritise mobile registration

A substantial share of betting acquisition occurs on mobile.

Registration should therefore be designed for mobile completion rather than treated as a desktop form that happens to shrink.

Review:

• input types;

• keyboard behaviour;

• field spacing;

• dropdowns;

• document upload;

• page speed.

Small usability problems become more significant on a smaller screen.

Measure field-level abandonment

Overall registration conversion tells you that users are leaving.

Field-level analysis can help explain where.

Suppose one stage produces a disproportionate drop-off.

That may indicate:

• unclear language;

• technical failure;

• unexpected information requirements;

• poor mobile execution.

The next step should be investigation, not automatically assuming those users lacked intent.

Verification is part of the funnel

Registration completion should not automatically be treated as the end of conversion.

If a large percentage of registered users fail to complete required verification, the funnel is still leaking.

Track:

Registration completed
→ verification started
→ verification completed

This can reveal whether the issue lies in:

• traffic quality;

• communication;

• document experience;

• operational delays.

Payment is one of the most commercially important funnel stages

A registered and verified player who cannot deposit has not converted in a meaningful commercial sense.

Payment performance therefore deserves the same attention as:

• creative;

• landing pages;

• registration.

Track factors such as:

• payment-method availability;

• decline rates;

• processing issues;

• device performance.

Local payment relevance can outweigh another creative test

In some markets, the availability of suitable payment methods can have a larger effect on deposit conversion than another round of advertising optimisation.

That is why betting funnel optimisation needs cross-functional involvement.

A media team cannot solve a payment problem through better targeting.

A CRM team cannot solve a broken payment flow by sending more reminders.

The funnel must be managed as one system.

Track first-deposit timing

Time to deposit can also reveal friction.

Compare:

• players depositing immediately;

• within 24 hours;

• within seven days.

A long delay between registration and deposit may suggest:

• payment friction;

• verification delay;

• weak post-registration messaging;

• insufficient product motivation.

Different acquisition sources may also have naturally different timings.

The value lies in comparing similar cohorts rather than forcing every player into one expectation.

Treat betting offers as an economic lever

Welcome offers can influence conversion significantly.

They also have a direct cost.

More importantly, they can change the type of player entering the funnel.

A large incentive may attract:

• genuine prospective customers;

• promotion-led players;

• users comparing bonuses across several brands.

That makes offer strategy a player-quality decision, not simply a conversion tactic.

Do not assume every segment needs the same offer

Different acquisition routes may respond differently to incentives.

Brand search

Players already searching for the operator may have substantial existing intent.

A large incentive may be less necessary if:

• the product proposition is strong;

• the brand is established.

Generic search

Competition can be stronger.

An offer may help the operator gain consideration against several alternatives.

Paid social

Promotional messaging may generate attention, but it can also influence the profile of users responding.

The right decision needs to be tested rather than assumed.

Measure offer economics beyond opt-in rate

A successful offer should not be judged only by how many players claim it.

Measure:

• registration conversion;

• FTD conversion;

• qualifying activity;

• bonus completion;

• bonus cost;

• net revenue after bonus cost;

• repeat deposits;

• early churn.

Suppose:

Offer A

• higher FTD conversion;

• high bonus cost;

• weak second deposits.

Offer B

• lower initial conversion;

• lower incentive cost;

• much stronger retention.

Offer A wins if the team stops measuring at FTD.

Offer B may produce better economics.

Higher promotional cost can still be justified

A larger incentive is not automatically inefficient.

If it acquires a cohort that:

• retains well;

• generates strong value;

• reaches payback;

the additional cost may be worthwhile.

Likewise, a small bonus is not automatically efficient if it produces little incremental conversion.

The decision should be based on contribution, not headline generosity.

Communicate betting offers clearly

Promotional communication should receive the same level of attention as the size of the offer itself.

Where relevant, players should be able to understand important elements such as:

• eligibility;

• stake requirements;

• expiry;

• maximum returns.

The exact requirements depend on jurisdiction and promotion.

The broader principle is clear:

Do not rely on surprise after registration.

Clarity helps users make a more informed decision and can reduce unnecessary disputes or customer-support demand.

Connect acquisition data to CRM from day one

One of the biggest weaknesses in betting funnels is the hand-off between acquisition and CRM.

The media team knows:

• source;

• campaign;

• creative;

• offer.

The CRM team receives:

New player.

That wastes valuable context.

The first 72 hours after registration can be especially important because the player has recently demonstrated intent but may still be deciding whether the product is useful to them.

CRM should continue that journey rather than restart it.

Pass useful campaign context into CRM

Where appropriate and permitted, CRM can receive useful acquisition information such as:

• source;

• campaign;

• broad proposition;

• offer;

• relevant product interest.

Cognaix's guide to gambling CRM integrations covers the wider connection between player data and retention.

This information can help make early-life communication more relevant.

Use source context without stereotyping the player

A football-led acquisition campaign provides a useful initial clue.

It should not permanently define the player.

Actual behaviour should take priority once it becomes available.

A good hierarchy is:

Acquisition intent provides the starting context.
Live player behaviour determines the next step.

That avoids both extremes:

• generic journeys;

• over-personalised journeys based entirely on one initial click.

Separate behavioural states

Useful early-life states can include:

Registered but not deposited

Potential needs may involve:

• payment confidence;

• verification;

• product understanding.

Deposited but not wagered

The barrier may now be:

• navigation;

• market understanding;

• product uncertainty.

First bet completed

The player has progressed further and should no longer receive first-action reminders.

Lapsed after initial activity

This player requires a different approach from someone who never used the product.

Early repeat depositor

This may indicate a stronger early-value cohort.

Each state needs different treatment.

Use event-based CRM rather than generic schedules

A generic welcome email three days after registration may be too late or irrelevant.

Event-based automation can react to:

• registration;

• verification;

• deposit;

• first bet.

This allows communication to follow actual player behaviour.

The value is not simply automation.

It is better timing.

Exit players once the goal is complete

Every lifecycle journey should define exit criteria.

For example:

A player in a first-deposit journey should leave once the first deposit is recorded.

Otherwise, the system can continue sending irrelevant communication.

This sounds simple, but poor exit logic is a frequent cause of weak CRM experiences.

Control frequency across channels

CRM should not maximise the number of opportunities to contact a player.

Several automated journeys can overlap:

• onboarding;

• event messaging;

• promotion;

• product communication.

Frequency caps help control the total experience rather than evaluating every campaign independently.

Relevant contact preferences and applicable suppression rules should also form part of the workflow.

Responsible gambling indicators belong inside funnel logic

Short-term conversion gains should not override appropriate player-protection controls.

A mature funnel should be capable of reducing or suppressing marketing where relevant statuses or indicators require it.

The exact treatment depends on:

• operator framework;

• jurisdiction;

• player state.

The objective is sustainable value rather than maximum contact intensity.

Measure the gaps between platforms

Funnel optimisation becomes difficult when the underlying numbers cannot be reconciled.

Different systems can report different stages:

Media platforms

• impressions;

• clicks;

• attributed conversions.

Analytics

• sessions;

• landing-page behaviour.

Affiliate platforms

• referrals;

• attributed registrations;

• partner conversions.

Operator data

• registrations;

• verification;

• deposits;

• revenue.

If those systems cannot be connected consistently, teams spend too much time arguing about numbers and too little time fixing the funnel.

Build a consistent betting funnel taxonomy

Create standard naming across:

• channel;

• campaign;

• creative;

• market;

• device;

• landing page;

• offer.

The same campaign should not appear under several names across different systems.

Consistent taxonomy makes cohort analysis dramatically faster.

Monitor each meaningful hand-off

A practical funnel scorecard might include:

Funnel stageExample measureImpression → clickCTRClick → landing-page engagementEngagement rateEngagement → registration startRegistration-start rateRegistration start → completionRegistration completionRegistration → verificationVerification completionVerification → first depositDeposit conversionFirst deposit → qualifying actionFirst product conversionFTD → repeat activityEarly retention

Each metric answers a different diagnostic question.

Overall CPA tells you what changed, not why

Suppose CPA increases from £50 to £65.

That is important.

But it does not explain the cause.

The funnel might show:

• CPC unchanged;

• landing-page conversion unchanged;

• registration completion unchanged;

• KYC-to-deposit conversion down sharply.

Now the team knows where to investigate.

That is the difference between reporting and funnel management.

Use server-side and first-party data appropriately

Browser tracking can miss parts of the journey.

Where appropriate, server-side data and first-party event infrastructure can improve the connection between:

• acquisition;

• deposits;

• later player outcomes.

Any implementation should still respect relevant privacy, consent and platform requirements.

The aim is measurement resilience rather than unnecessary data collection.

Automated reporting can accelerate diagnosis

Once the funnel taxonomy is reliable, automation can help identify unusual movements.

Examples:

• paid social registration volume is stable but FTD conversion fell 18%;

• one affiliate cohort has materially weaker Day 30 retention;

• mobile registration abandonment increased after a release;

• payment declines rose in one market.

That allows specialists to investigate high-value problems earlier.

AI can assist, but context remains human

AI can help:

• flag anomalies;

• summarise cohort changes;

• group creative themes;

• identify unusual funnel stages.

It cannot automatically know whether every movement is commercially important.

A fall in conversion might reflect:

• technical failure;

• market conditions;

• campaign mix;

• stronger qualification.

Experienced teams still need to interpret the context.

Build a betting funnel testing programme

Funnel optimisation can generate hundreds of potential tests.

Not all deserve equal priority.

Testing should reflect:

• commercial impact;

• number of users affected;

• implementation effort;

• confidence in the problem.

Fix high-volume leaks first

Suppose:

• button colour might improve registration by 0.2%;

• one payment method fails for 12% of users in a major market.

The payment issue deserves attention first.

This sounds obvious.

In practice, channel teams can become focused on the parts of the funnel they directly control while larger problems sit elsewhere.

A shared funnel view helps prioritise correctly.

Write a hypothesis

A useful experiment should state what is expected to change.

For example:

Clarifying verification requirements before registration will reduce abandonment during the KYC stage without reducing qualified registration volume.

That is stronger than:

Test new registration page.

Choose the right success metric

If the business cares about funded retained players, registration rate should not automatically determine the winner.

A landing-page test may produce:

• more registrations;

• fewer deposits.

The decision needs to reflect the real commercial objective.

Give tests enough volume

Short tests can be misleading.

Sportsbook demand is affected by:

• sporting calendars;

• day of week;

• major events;

• competitor activity.

Testing over one unusual weekend can produce a result that does not translate to normal operating conditions.

Where conversion volume is low, teams can combine quantitative evidence with:

• session analysis;

• customer feedback;

• operational insight.

Document the decision

Record:

• hypothesis;

• audience;

• dates;

• primary metric;

• result;

• decision.

This creates an organisational record of what has been learned.

Without it, teams risk repeating the same tests six months later.

Not every betting market should use the same funnel

A global template can provide consistency.

It should not eliminate local relevance.

Markets can differ in:

• regulation;

• payment preferences;

• sports calendars;

• language;

• competitor activity;

• product expectations.

That means the strongest funnel in one market may not be the strongest elsewhere.

Localisation should extend beyond changing:

• currency;

• country name.

The route itself may need adjustment.

Competitor intelligence can provide funnel context

Competitive monitoring can help explain market-level changes.

For example:

• a rival introduces a stronger local payment experience;

• several competitors begin emphasising one product benefit;

• affiliate visibility changes sharply.

Cognaix's work on sportsbook competitor analysis shows how those signals can support acquisition decisions.

Competitor activity should create hypotheses rather than automatic imitation.

Common betting funnel optimisation mistakes

Optimising only the top of the funnel

More clicks or registrations do not guarantee more valuable players.

Sending every channel to the same landing page

Different acquisition intents often need different levels of context.

Removing friction without considering player quality

Not every reduction in friction creates better acquisition.

Treating registration as the end of conversion

Verification and first deposit remain critical stages.

Ignoring payment performance

Payment friction can destroy otherwise strong acquisition economics.

Using offers as the default solution

Incentives are a commercial lever, not a universal fix.

Treating all FTDs as equal

Different sources can create very different downstream cohorts.

Disconnecting acquisition from CRM

The player journey should continue after registration rather than restart.

Measuring overall CPA without stage-level conversion

Headline movement tells teams little about the underlying cause.

Testing low-impact changes first

Prioritise the leaks with the greatest commercial consequence.

Build a betting funnel optimisation scorecard

A practical scorecard might contain:

StagePrimary KPIQuality checkAcquisitionCost per qualified clickSource intentLanding pageRegistration-start rateDownstream FTD qualityRegistrationCompletion rateVerification rateVerificationKYC completionQualified-player ratePaymentFirst-deposit conversionDecline rateFirst activityFirst wager/product actionEarly engagementEarly lifecycleSecond depositDay 7/30 retentionMature cohortNGR/contributionPayback/value

The purpose is not to maximise every metric independently.

It is to identify which stages constrain overall player value.

How Cognaix approaches betting funnel optimisation

Cognaix approaches betting funnels as connected acquisition and retention systems rather than separate channel projects.

Paid media

Paid search and paid social should be evaluated through the complete acquisition journey.

A campaign that generates inexpensive registrations but weak depositors should not be scaled solely because the platform KPI looks efficient.

Affiliate marketing

Affiliate traffic should be evaluated by:

• partner;

• placement;

• offer;

• downstream cohort quality.

This helps distinguish efficient volume from valuable acquisition.

CRM

Source and campaign context can support more relevant onboarding where appropriate.

Behavioural states should then determine the next communication.

Data and attribution

Consistent naming and event definitions make it possible to connect:

• media;

• registrations;

• deposits;

• player value.

Without reliable taxonomy, funnel optimisation quickly becomes manual and reactive.

Reporting

Reporting should show:

• where users are leaving;

• which cohorts are strongest;

• what changed;

• where action is required.

The objective is decision-making rather than dashboard volume.

Automation

Automated monitoring can flag:

• unusual conversion changes;

• weak cohort quality;

• tracking issues.

This allows teams to focus manual analysis on the problems most likely to affect commercial performance.

Competitor intelligence

Market monitoring helps explain whether funnel performance changes reflect:

• internal execution;

• wider competitive conditions.

Player quality

Player value remains the final test.

The goal is not simply a smoother funnel.

It is a funnel that produces more sustainable, commercially valuable customers.

Final thoughts

Betting funnel optimisation works best when operators stop treating acquisition, registration, payments and CRM as separate optimisation problems.

The player experiences one journey.

The business should measure it the same way.

Start by defining what a valuable player looks like.

Then map the funnel backwards.

Which channels create the strongest cohorts?

Does the landing page preserve the promise made in the advert?

Where are users abandoning registration?

Are verification requirements clear?

Can players deposit through suitable methods?

Does the offer create additional value after promotional cost?

Does CRM continue the acquisition journey or restart it with generic messaging?

Can teams see where conversion changes occur without spending days reconciling systems?

Those questions reveal far more than a headline CPA.

The most effective next step is also rarely a complete funnel rebuild.

Find one high-volume leak.

Connect it with player-quality data.

Form a hypothesis.

Make a controlled improvement.

Measure the result beyond the immediate conversion event.

Then repeat the process.

That turns betting funnel optimisation from a collection of tactical conversion tests into a repeatable system for creating more valuable acquisition.

Frequently asked questions

What is betting funnel optimisation?

Betting funnel optimisation is the process of improving the journey from initial acquisition through registration, verification, deposit, product activity and retention.

The objective is not simply to increase conversion at each stage. It is to improve the number and quality of players progressing through the complete journey.

What are the main stages of a sportsbook conversion funnel?

A typical sportsbook funnel can include impression, click, landing-page engagement, registration start, registration completion, verification, first deposit, first wager, repeat activity and retention.

The exact stages will depend on the operator, market and product.

How can sportsbooks improve registration conversion?

Operators can investigate form length, mobile usability, validation errors, field-level abandonment and how clearly required information is explained.

Necessary verification or market-specific controls should not simply be removed to increase conversion. The aim is to reduce avoidable friction while maintaining appropriate safeguards.

Why is payment conversion important in betting funnels?

A registered player who cannot deposit has not completed the commercially important acquisition journey.

Operators should therefore monitor payment availability, declines, processing issues and device-level performance alongside registration conversion.

In some markets, payment improvements may create greater value than further advertising optimisation.

Should every sportsbook landing page use the same welcome offer?

Not necessarily.

Different acquisition sources can have different levels of intent and may respond differently to incentives.

Operators can test whether offers are required for particular segments and compare downstream outcomes such as repeat deposits, bonus cost and retention rather than relying solely on initial conversion.

How should acquisition and CRM work together?

Useful acquisition context can be passed into CRM where appropriate, including source, campaign and broad proposition.

This can improve early-life journeys.

Once the player begins interacting with the product, actual behaviour should determine the next communication rather than relying indefinitely on the original acquisition source.

Which metrics should be used to optimise a betting funnel?

Useful measures can include click-through rate, landing-page engagement, registration completion, KYC completion, first-deposit conversion, second-deposit rate, Day 7 and Day 30 retention and player value.

The correct KPI depends on the funnel stage being investigated.

How should betting funnel tests be prioritised?

Prioritise issues according to commercial impact, number of users affected and implementation effort.

A significant payment or registration problem usually deserves attention before marginal changes to creative or button copy.

Every material test should have a clear hypothesis and success metric.

Can AI improve betting funnel optimisation?

AI can help identify anomalies, summarise cohort differences, group creative performance and highlight funnel stages that behave unusually.

It should support rather than replace experienced interpretation, particularly where performance changes may be influenced by market conditions, compliance requirements or player behaviour.