How to Create an Animated Funnel Chart

Every team has a funnel, and almost every team shows it the same forgettable way: a list of stages with a number next to each, or a grey set of stacked bars pasted into a slide. The data is all there, 18,000 visits, 5,400 signups, and so on down the line, and yet the single most important thing about a funnel, where you are bleeding people, somehow never lands. People nod, the slide moves on, and the leak that is quietly costing you the most money goes unexamined for another quarter.

The problem is that a funnel is fundamentally a story about loss, about people falling out between one step and the next, and a static list of counts makes the reader do the subtraction to find where the loss is worst. Most people will not do that subtraction. They will glance, register "numbers go down, I guess," and move on.

This is a practical guide to building a funnel chart that does the work for them, and specifically to making an animated one, where each stage visibly narrows so the biggest drop-off is impossible to miss. I will skip the spreadsheet-formula trivia and spend the time on the decisions that decide whether your funnel actually changes a conversation: clean stage data, the right number of stages, the conversion math that matters, and an export that survives being seen on a phone.

In this article

1. What a funnel chart actually shows
2. Why a funnel chart should move
3. Step 1: Get your stage data right first
4. Step 2: Choose your stages, and how many
5. Step 3: Decide what to measure at each step
6. Step 4: Build it and export the animated version
7. The mistakes that make a funnel misleading
8. A few funnels worth copying
9. The two-minute version

What a funnel chart actually shows

A funnel chart shows how a population shrinks as it moves through a sequence of stages, visitors becoming signups becoming paying customers, so you can see both how many survive to the end and, more importantly, where the steepest losses happen along the way.

There is a distinction worth getting straight before you build anything, because mixing it up is the most common reason funnels confuse people. A funnel can show two different things at each stage:

• Absolute counts. The raw number of people at each step: 18,000 visits, 5,400 signups, 2,060 activated, 680 paid. This answers "how big is each stage?"
• Conversion rate. The percentage that makes it from one step to the next, or from the very top. This answers "how leaky is each step?", which is usually the real question.

The most useful funnels show both: the count so the scale is clear, and the conversion rate so the leak is obvious. The point of the chart is rarely the final number on its own; it is the shape of the decline, where the funnel pinches hardest. Decide that the drop-off is your story, and the rest of the design follows.

Why a funnel chart should move

Here is the case for animation, and for a funnel it is unusually strong. Of every chart type, a funnel benefits most from motion because its entire subject is narrowing, people being lost between steps, and animation lets you enact that loss rather than just report it. When each stage shrinks into the next in sequence, the reader watches the population collapse, and the stage where it collapses hardest becomes a genuine gut-punch instead of a number they have to compute.

There is a sharper reason too. A funnel is the single most relatable chart you can put in front of an operator, because everyone in your audience has one of their own. A static funnel gets a polite nod. An animated one that dramatises the drop between signup and activation makes every viewer instinctively compare it to their own activation rate, and that is what earns a comment, a question, or a reply. We made exactly this point about why funnels travel so well in the guide to animated charts for LinkedIn.

The old objection, that animating a chart is expensive and fiddly, no longer holds. Producing the moving version now takes about as long as the static one, which is the whole assumption behind the rest of this guide.

Step 1: Get your stage data right first

A funnel is only as honest as the population you put into it, and this is where most funnels quietly go wrong before anyone reaches the design.

Make sure every stage counts the same people. The cardinal rule of funnels is that each stage must be a subset of the one before it. The 5,400 signups have to come from within the 18,000 visits, not from a different time window, a different campaign, or a different definition. If your "visits" are this month but your "signups" include last month's stragglers, the funnel lies before you have drawn a thing.

Pick one consistent time window. Decide whether you are showing a single cohort moving through the stages over time, or a snapshot of everyone in each stage right now. Both are valid; mixing them is not. A cohort funnel ("of the people who visited in May, how many eventually paid?") is usually the most honest, because it actually follows the same people down the steps.

Order the stages by sequence, not by size. Stages go in the order people actually move through them, visit, then signup, then activation, then paid, even if that is not the order of magnitude. The whole point is the path, so the path has to be the axis.

Calculate the conversion at each step. For each stage, the step conversion is simply that stage divided by the one above it. From our example: 5,400 / 18,000 is a 30% visit-to-signup rate; 2,060 / 5,400 is about 38% signup-to-activation; 680 / 2,060 is about 33% activation-to-paid. The overall conversion, top to bottom, is 680 / 18,000, or about 3.8%. Those step numbers are where your story lives, because they tell you the leakiest step is not always the one with the biggest raw drop.

Step 2: Choose your stages, and how many

The number of stages is a real design decision, and more is emphatically not better. The sweet spot for a funnel chart is four or five stages. Fewer than three and it is not really a funnel; more than six and the chart turns into a thin stack of slivers where no single drop-off stands out, which defeats the entire purpose.

The discipline is to choose the stages that represent genuine, meaningful transitions, the moments where people make a decision and some of them say no. Visit, signup, activation, and paid are meaningful gates. "Viewed pricing page," "scrolled to mid-page," and "hovered the button" are not stages; they are noise dressed up as rigour.

If you have a longer process, say a sales pipeline with eight steps, resist the urge to plot all eight. Group the minor steps into the major transitions, or build a focused funnel around the two or three steps where you actually lose deals. A funnel chart is a story about your biggest leaks, not an audit trail of every micro-step. When the comparison you care about is between a few discrete values rather than a flow, a bar chart is sometimes the cleaner choice; reach for a funnel chart specifically when the narrowing is the message.

Step 3: Decide what to measure at each step

Once your stages are set, decide which numbers actually appear on the chart, because a funnel can carry three different figures at each stage and showing all three at once is a mess.

The three candidates are the absolute count (5,400), the step conversion from the previous stage (30%), and the overall conversion from the top (also 30% here, since signups is the second stage). They answer different questions, and the right pick depends on your audience.

For an operator trying to find the leak, step conversion is the most actionable number, because it isolates each transition. A 38% signup-to-activation rate tells you exactly how that one step performs, independent of how good or bad the steps around it are. For an executive or investor who wants the headline, overall conversion and the final count matter more, because they care about the end result, not the internal mechanics.

A clean default that works for most audiences: label each stage with its count and its step-conversion percentage, and let the biggest drop be visually obvious. That gives the operator the diagnostic and the executive the scale, without clutter. Whatever you choose, be explicit, an unlabeled percentage on a funnel is ambiguous, because the reader cannot tell whether 33% means "of the previous step" or "of the very top."

Step 4: Build it and export the animated version

With the thinking done, building is quick. The workflow that takes you from a list of stages to a finished animated funnel in a couple of minutes:

1. Open the funnel chart generator and enter your stages in order with their counts, visits, signups, activated, paid.
2. Turn on the step-conversion labels so each transition shows its percentage, and add a title that states the funnel's subject and time window ("Q2 signup funnel, May cohort").
3. Pick a theme and watch the preview animate as each stage narrows into the next, so you can see where the pinch lands before you commit.
4. Export, choosing the format by where the funnel is going.

The destination decides the format, and getting this right is what separates your funnel from the static screenshot everyone else pastes in:

• A slide deck or report? A crisp PNG is fine if the deck is read on paper or static, but if you are presenting live, an MP4 that plays each stage narrowing as you reach the slide makes the leak land far harder.
• A social feed (LinkedIn, X)? Upload a native MP4 video, square or portrait, and keep the animation to three to six seconds so it loops while people read. The LinkedIn chart generator is tuned for this, and the animated LinkedIn guide covers the platform specifics.
• A quick share in Slack or email? A GIF loops automatically without anyone pressing play.
• A page or dashboard? An SVG stays crisp at any size.

The animated export is not decoration here. A funnel's whole job is to make a drop-off visceral, and a stage that visibly collapses does that in a way a frozen image never will, at no extra effort now that the tooling does the work.

The mistakes that make a funnel misleading

Most weak funnels are not ugly; they make one of a handful of credibility errors that a sharp audience catches immediately.

• Stages that are not subsets. Counting different populations at different steps, different time windows, different campaigns, so the "funnel" is really several unrelated numbers stacked up. Every stage must come from the one above it.
• Too many stages. Eight or ten thin slivers where no single drop stands out. Collapse to the four or five transitions that matter.
• Vanity top-of-funnel. Starting the funnel at "impressions" or "reach", a huge, cheap number, to make the conversion look more dramatic than it is. Start at a stage that represents real intent.
• Unlabeled or ambiguous percentages. A number that could be step conversion or overall conversion, leaving the reader to guess. Always say which.
• Hiding the biggest leak. Designing the chart so the worst step blends in with the rest. The largest drop-off is the entire point; make it the most obvious thing in the frame.
• A static export when the story is loss. Sending a frozen funnel when an animated one would make the leak undeniable. If the narrowing is the message, let it narrow.

A few funnels worth copying

Steal these formats and drop in your own numbers.

The marketing or product funnel

The classic. Visits, signups, activation, paid, four stages narrowing toward the customers you actually keep, with each step's conversion labelled, 18,000 to 5,400 to 2,060 to 680. Animate the stages collapsing in sequence so the leakiest step lands as the headline, and you have the most relatable chart you can put in a deck or a post. Build it with the funnel chart generator and let the biggest drop be the punchline.

The sales pipeline

When the story is deals rather than users, the same shape works for a pipeline: leads, qualified, demo, proposal, closed-won. Keep it to the transitions where deals actually die, usually qualification and proposal, rather than every CRM stage. Label the step conversion so the team can see exactly where the pipeline leaks, and animate it for the QBR or investor update where it will do the most good.

The onboarding or activation funnel

When your concern is the early-life journey, account created, profile completed, first action taken, "aha" moment reached, habit formed, a tight funnel of the onboarding steps shows precisely where new users stall. This is often the single most valuable funnel a product team can build, because the steepest drop usually points straight at the fix. Animate each stage narrowing and the stall becomes impossible to ignore in the next planning meeting.

The two-minute version

You do not need a designer or a motion tool for any of this. The whole method:

1. Make sure every stage is a subset of the one above it, drawn from one consistent population and time window.
2. Choose four or five meaningful stages, the real decision points, not micro-steps.
3. Decide what to show at each step, count plus step-conversion is a safe default, and label it unambiguously.
4. Make the biggest drop-off the most obvious thing in the frame, because the leak is the story.
5. Build it, then export the animated version, an MP4 for decks and feeds, a GIF for quick shares.
6. Let the funnel narrow on screen, so the place you are losing people lands as something felt, not calculated.

Your funnel already knows where you are losing people. Stop hiding that answer in a list of numbers nobody subtracts, and start letting the chart show it, one narrowing stage at a time.