BACK TO BLOG
TOOL HOW-TOOctober 26, 20266 min read

How to Use Lucky Orange to Find Where a Funnel Leaks

JH

By Joris van Huët

Enterprise Interim CMO & Marketing Leader · 15+ years · 50+ orgs

Published

2026-10-26

The short answer: on the Build plan, $32 a month billed annually for 3,500 sessions (pricing page, checked on 26 October 2026), Lucky Orange can show a small site which step of a signup or checkout loses people, as long as each step is a URL or an event: install the code, block your own IP, set the Primary Conversion Event, build the funnel, open the leaking step's frustrated visitors, watch their recordings, check the form, ask a one-question survey on that page, have Discovery AI summarise, then fix one thing and refresh the funnel.

Disclosure: I may earn a commission through some links on this site. It does not change what this page says.

A funnel tells you where. Recordings, form analytics and a survey tell you why. Lucky Orange keeps all four on the same visitor record, which is the reason to use it for this job instead of four separate tools. For plans and rivals, see the Lucky Orange glossary entry.

Before you start:

  • A funnel you can write as URLs or events. Two steps at least: product page, cart, checkout started, order, or landing page, form, thank-you page.
  • History inside your storage. Funnels fill in past visitors as far back as your data storage, 60 days for most plans, and an event step counts only people who triggered it after the event was created, per the funnels guide.
  • The Discovery add-on, for the AI step. On the Shopify listing it adds $10 a month to Build and $20 to Grow.

1. Install the tracking code

Copy the code from Overview, or from Settings, Sites, View Details, and paste it just before the closing head tag, per the installation guide. A shared page template needs it once; hand-coded static pages need it on each page. Shopify's integration places it for you.

2. Keep your own visits out

Open Settings, then Blocked Visitors, click Rules and Add Block Rule, and choose Block by IP for your office addresses, as the blocking article shows. One side effect: blocked addresses no longer see chat, surveys or announcements, so test those from another connection.

3. Mark the Primary Conversion Event

Open the event behind your goal, click view details and switch it on as the Primary Conversion Event, with a monetary value if it has one. A site gets one primary and any number of secondary events, and Discovery uses the primary to analyse conversion trends, per the events guide.

4. Build the funnel

Go to Analytics, then Funnels, and click New Funnel. Name it, click Next and define each step by page URL, with Contains or Exact, or by event; add steps, drag them into order and click Done (funnels guide). For paid traffic, make step one a URL that contains utm_source=google. Visitor data can take a few minutes to appear.

5. Find the step that leaks, and who left there

The step with the steepest drop is your page. Filter its visitors with the ready-made segments: Frustrated Visitors covers rage clicks, super rage clicks, repeated fields, shaky mouse movement and U-turns, and Confused Visitors catches anyone who filled the same field more than once, per the segments guide. The events guide defines a rage click as three or more clicks on one element, or within 50 pixels, inside a second, with carousels and quantity buttons excluded.

6. Watch the recordings

Open recordings for that segment on that page. Rage clicks now appear on the recording timeline just before they start, per the 2026 updates, so you can jump to the moment instead of watching the whole visit.

7. Check the form, if the step is a form

Form analytics shows how people move through each field. It cannot read a form inside an iFrame; if a recording shows a grey or white box where the form should be, that is the sign, per the form analytics FAQ. Press Shift to switch the cursor between using the page and reading the field boxes.

8. Ask the people who stop

Put a survey on the leaking page with a time or scroll delay. The survey tips are plain about the trade: multiple choice for more answers, open questions for better ones. Each response links back to that visitor's recording, per the homepage.

9. Have Discovery AI draft the diagnosis

Run the Quick Actions Conversion Changes and Bounced Session Analysis, then Analyze a Heatmap on the leaking page (2026 updates). Discovery answers from your live data and links each claim to the heatmap, funnel or recording behind it, per the Discovery AI page.

The human check: Discovery writes the hypothesis, a person watches the recordings it links and reads the survey answers before anyone briefs a change. A summary that the recordings do not show is a summary, not a finding.

10. Fix one thing, then refresh the funnel

Change one element on the leaking step. Funnel data updates when you open the funnel, and a Refresh button forces it, per the funnels guide; compare the step's drop-off before and after, over the same number of days.

What the help centre warns about

  • A quota that runs out mid-month. When sessions hit the plan limit, tracking, surveys and chat stop until the next cycle, per the billing FAQ.
  • Standard Shopify checkout. Only Shopify Plus allows checkout recordings; other stores get checkout events such as Checkout Started and Ordered, which can serve as event steps in a funnel (checkout policy).
  • A short memory. Funnels look back only as far as storage, 60 days on every paid plan by default, which is how a year-on-year question turns into a storage upgrade.

FAQ

Why does an event step show fewer people than the URL steps?

An event step counts only visitors who triggered the event after you created it, while URL steps reach back through your storage window, per the funnels guide.

Which surveys get the most answers?

Multiple choice, like/dislike and rating surveys bring more responses; open questions bring fewer but richer ones, per the survey tips.

Does Discovery AI need anything set up first?

A Primary Conversion Event, for its conversion analysis, and the Discovery add-on on a paid plan; the free plan excludes add-ons.

TAGS
Lucky OrangeConversion FunnelsSession RecordingsForm AnalyticsWebsite SurveysDiscovery AI

ABOUT THE AUTHOR

I am Joris van Huët, an enterprise interim CMO. 15+ years, 50+ organizations, including ING, P&G, Nestlé, BNP Paribas, WeTransfer and Vinted. I take on innovation projects (venture building, design sprints), agentic marketing (setting up and running AI agents) and hands-on multi-channel work. See the track record.

I wrote and published this with AI assistance, and I answer for it. Claims about my own experience are limited to the track record above, and a statistic links to its source or is labelled as an example. I sell interim and fractional CMO work, which is why this site exists. How this site is written.

Senior marketing execution

A senior operator doing the work: campaigns, channels, CRM and AI agents, live from day one.