> For the complete documentation index, see [llms.txt](https://docs.fullsession.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.fullsession.io/12.-funnels.md).
# 12. Funnels
A **funnel** is an ordered sequence of steps you want visitors to complete — *View product → Add to cart → Checkout → Purchase* — and the analysis of how many make it through each one. Funnels show you exactly **where people drop off**, **why**, and let you jump straight into the sessions behind every lost conversion. They build on segments (\[Chapter 11 — Segments]), with which they share their underlying model.
> Funnels require the funnel feature on your plan and the **view funnels** permission. If you don't see Funnels, it isn't enabled for your account or role.
***
### 12.1 Building a Funnel
A funnel is a named set of **ordered steps**. Each step is a condition a visitor must satisfy, in sequence, to keep advancing through the funnel.
#### Step types
A step can be any of four types:
| Step type | Matches when a visitor… | Defined by |
| ----------------- | ----------------------------------------- | ---------------------------------------------------- |
| **Visit page** | Loads a particular page/URL | A URL (with a match operator like *is* / *contains*) |
| **Click** | Clicks a specific element | A **CSS selector** or the element's **text** |
| **Changed input** | Changes an input field | The field's selector/text |
| **Custom event** | Fires one of your `FUS.event(...)` events | The event name |
Each step has a **name**, a **value** (the URL, selector, text, or event name), a **match operator**, and an **order** in the sequence.
#### Defining steps
1. Open **Funnels** and create a new funnel.
2. Add steps in order — for example:
* Step 1: *Visit page* `/product`
* Step 2: *Click* the *Add to cart* button (by selector or text)
* Step 3: *Visit page* `/checkout`
* Step 4: *Custom event* `purchase`
3. Name the funnel and save.
#### Scoping to a segment
A funnel is analyzed within a **segment**. By default that's **Everyone** (\[Chapter 11, section 11.1]), but you can select any saved segment to analyze the funnel for a specific audience — *"the checkout funnel, for mobile users in Germany."*
> **Processing** — when a funnel is created or its definition changes, FullSession may need to compute it. You might briefly see a **processing** state (in queue / processing) before the results are ready.
> **Funnels are linear.** Steps run in a single ordered sequence — there's no branching or alternative paths. Model distinct paths as separate funnels.
***
### 12.2 Reading the Results
Once computed, a funnel shows headline metrics plus a step-by-step breakdown. The view is organized into tabs — **Funnel Overview**, **Funnel Steps**, and (where enabled) **Funnel Trends**.
#### Headline metrics
| Metric | Meaning |
| ------------------------------ | ------------------------------------------------------------------------------------------- |
| **Conversion rate** | The share of visitors who completed the whole funnel |
| **Time to convert** | The **median** time from the first step to completion (shown as days/hours/minutes/seconds) |
| **Users completed the funnel** | How many visitors finished all steps |
#### Step-by-step breakdown
The chart shows the count reaching each step and the drop between steps, so the biggest leak is obvious at a glance. FullSession also highlights the **largest single drop-off** to point you at the step most worth fixing.
#### Sessions vs. unique users
Funnels can count **sessions** or **unique users** at each step — switch depending on whether you care about per-visit behavior or per-person reach.
#### Breakdown dimension
Add a **Breakdown** to split the funnel by a dimension and see how conversion differs across groups:
**Browser · Operating System · Country · Device · Region · UTM Source · UTM Medium · UTM Campaign · Referrer · Browser Version**
#### Funnel Trends
Where the trends feature is enabled, the **Funnel Trends** tab plots funnel performance **over time**, so you can see whether conversion is improving or degrading across days, weeks, or months.
#### Filtering
Funnel analysis respects the **date range** (the global time filter), the selected **segment**, and your **breakdown** and **step-scope** choices (you can focus on "all steps" or a single transition like *Step 2 → Step 3*).
> **No conversion window.** There's **no setting** that requires visitors to complete the funnel within a fixed time. FullSession reports the *median* time to convert, but it doesn't constrain conversions to a window.
***
### 12.3 Effort Analysis
Beyond raw conversion, FullSession measures the **effort** visitors expend between steps — a strong signal of friction even where people eventually convert.
For each transition between steps, effort analysis reports (as medians):
| Metric | Meaning |
| ---------------- | --------------------------------------------------------- |
| **Conversion** | The share of visitors who advanced to the next step |
| **Interactions** | How many interactions visitors made between the two steps |
| **Time Engaged** | How long visitors spent between the two steps |
| **Revisit Rate** | The share who left before advancing |
High interactions or time engaged between two steps — even with decent conversion — usually means that part of the flow is harder than it should be. Effort analysis surfaces that hidden friction.
***
### 12.4 Drop-off Analysis & Watching Sessions
The most actionable part of a funnel is understanding *why* people drop off — and watching it happen.
#### Conversion analysis
For a given transition, FullSession surfaces the **issues correlated with drop-off** — for example **dead clicks**, **rage clicks**, **network errors**, or specific **custom events** — each with:
| Column | Meaning |
| ----------------------------- | -------------------------------------------------- |
| **Event / issue** | What happened (e.g. a rage click, a network error) |
| **Location** | Where it occurred (e.g. a CSS selector) |
| **Lost conversions** | Conversions estimated lost to this issue |
| **% of lost conversions** | Its share of the drop-off |
| **Impact** | Its overall impact on the funnel |
| **Affected sessions / users** | How many were affected |
This turns "people drop off at checkout" into "people who hit this failing request at checkout were far less likely to convert."
#### Watch the sessions behind a drop-off
Click into an issue (or a step) to open the **sessions** behind it, split into:
* **Dropped off** — visitors who did *not* complete the funnel.
* **Converted** — visitors who did.
Watching a handful of dropped-off sessions usually reveals the concrete cause — a confusing form, an unresponsive button, a slow request — which you can confirm against the **player's** console and event panels (\[Chapter 6 — The Session Player]).
***
### 12.5 Top Issues and Conversion Analysis
Funnels include several higher-level reports to help you interpret a funnel without reading every session.
#### Top Issues
The **Top Issues** report ranks the problems most affecting the funnel's conversion — each with its affected sessions/users, estimated **lost conversions**, **impact**, the **step** it relates to, and a likely **cause**.
#### AI failed-conversion analysis
Funnels also offer an **AI analysis of failed conversions**: FullSession examines sessions of visitors who dropped off and produces a written explanation of **why** they likely didn't convert — a fast, plain-language read on a step's problems.
> **Funnels vs. Lift AI.** The AI explanation here is part of **Funnels** — it interprets drop-off for a step on demand. The deeper, ongoing AI capability — **AI goals**, scheduled **AI runs**, and tracked **AI issues** linked to a funnel over time — is the separate **Lift AI** feature, covered in \[Chapter 14 — Lift AI].
***
### 12.6 Comparing Across Time & Segments
Unlike standalone segments, funnels support **comparison** — putting two or more views of the funnel side by side. This is how you measure whether something changed or whether two audiences behave differently.
#### Time-period comparison
Compare the funnel against a **previous period**: previous **day, week, month, quarter, or year**. Ideal for answering *"did our checkout conversion improve after the redesign?"*
#### Segment comparison
Compare the funnel across **up to three other segments** at once — for example *Free vs. Pro vs. Enterprise* users moving through onboarding. This reveals where different audiences diverge in the same flow.
> **Comparison is time- and segment-based.** You can compare against earlier periods and against other segments; there's no separate cohort-comparison mechanism beyond these.
***
### 12.7 Managing & Exporting Funnels
Funnels live in their own list, separate from segments.
#### Managing funnels
| Action | Notes |
| ------------------- | -------------------------------------------------------------------------------------- |
| **Create** | Requires the **create funnels** permission |
| **Edit** | Update steps, breakdown, comparison, and saved selections (**Save funnel selections**) |
| **Favorite (star)** | Pin frequently used funnels to the top |
| **Delete** | Remove a funnel (requires the **delete** permission) |
Like segments, funnels are owned by their creator. Conversion analysis is gated by its own permission.
#### Exporting
Funnels can be **exported**: FullSession produces the funnel's data — the **overall** numbers plus the lists of **converted** and **dropped-off** sessions — as **CSV files in a downloadable archive**. This is useful for sharing results or cross-referencing converters and non-converters in another tool.
#### As a dashboard card
A funnel can also appear as a **funnel card** on a dashboard (\[Chapter 9, section 9.5]), so its conversion rate and step progress sit alongside your other metrics.
#### What funnels don't have
To set expectations: there's **no funnel duplication**, **no templates/library**, **no branching** (steps are linear), **no version history**, and **no built-in goals/targets** (targets live in **Lift AI**, \[Chapter 14]).
> **The big picture** — define an ordered funnel from **page / click / input / custom-event** steps, scope it to a **segment**, and read **conversion rate, median time-to-convert, and completions** with a step-by-step breakdown (sessions or unique users, optionally split by a **breakdown dimension** or over time in **Trends**). **Effort analysis** exposes hidden friction, **drop-off/conversion analysis** ranks the issues and opens the **sessions** behind them, **Top Issues/Events** and an **AI explanation** interpret the leaks, and **time/segment comparison** plus **CSV export** round it out. There's no conversion window, duplication, branching, or templates.
***
> **Next up:** \[Chapter 13 — Feedback & Surveys] covers the in-page feedback widget and survey responses — and how feedback connects back to the sessions and funnels in this and earlier chapters.
---
# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.
## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.
Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:
```
GET https://docs.fullsession.io/12.-funnels.md?ask=&goal=
```
`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.
Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.