Analytics & chart blocks
This page is a quick summary of all thirteen chart and KPI blocks — each also has its own dedicated page with the full field list and worked examples, linked at the end of its section below.
Thirteen chart and KPI blocks for telling a data story inside the email itself — a weekly digest, a sales recap, a report your recipients don't have to click through to read. They live in the same Content section of the Insert palette as every other block; there's no separate "Analytics" category.
A few things are true of every chart in this family, so they're stated once here instead of thirteen times below:
- Every chart renders as a single server-generated image — no client-side chart library, no JavaScript — with a visible data table right alongside it. The table isn't a footnote or an accessibility afterthought: it's the same numbers as the image, in plain HTML, and it's what makes the chart actually usable in text-only clients, image-blocked inboxes, and screen readers. It's never optional.
- A chart's numbers are fixed at send or export time — the same "accurate as of send" convention as the countdown timer's deadline or a product block's inventory badge, not a live-updating widget. (The one exception: a chart bound to a campaign's own response results reads live at render time — see Binding charts to real data.)
- No dual-axis charts, ever. Two measures of different scale are two chart blocks side by side, not one chart pretending to have two y-axes.
KPI scorecard
One big number for a headline metric — revenue, signups, whatever the email is reporting on.
- Fields: a label, a value with a format (plain number, currency, or percent), an optional previous value to show a Δ against, and an optional sparkline.
- The Δ direction (up/down/flat) is something you set yourself, not inferred from whether the numbers went up — so you stay in control of what counts as good news.
- The sparkline is a small trend image; the label/value/Δ underneath it is always real HTML text, never dependent on the image rendering.
Bar chart
Vertical or horizontal bars, up to 3 authored series.
- Past around 8 categories, the chart drops direct value labels on each bar in favor of a shared axis with gridlines — reading exact numbers off a dozen labeled bars stops being readable before that, so the data table takes over as the precise source.
Line / area chart
Same row-by-row authoring shape as the bar chart — categories down the rows, up to 3 series across — plus a filled toggle for the area-chart variant.
- Point markers on each data value start thinning out past around 20 points so the line doesn't turn into a solid row of dots; the stroked line, shared axis, and data table keep carrying the trend past that.
Pie / donut chart
One value per category, drawn as a pie or (in donut mode) a ring with an optional center label for a KPI-style number in the middle.
- Past 8 slices, the smallest ones fold together into a single "Other" wedge rather than the chart getting unreadably thin — the legend and data table both show the same fold.
Scatter / bubble chart
X/Y points, with an optional series column to group points into multiple colors and an optional size column to turn it into a bubble chart.
- Series identity comes from whatever you type in the Series column — leave it blank and every point shares the block's own title as its one series.
Histogram
Bins a column of raw values into ranges automatically — hand it individual numbers, not pre-counted buckets, and it does the bucketing for you.
- You control roughly how many bins to aim for; if every value you typed is identical, it collapses to one bin instead of drawing nineteen empty ones.
Heatmap
A row × column grid of values, shaded on a sequential color ramp from light to dark.
- Cells show their exact value as text only while the grid stays small enough to read at a glance — past that the ramp itself (plus the always-visible data table) carries the numbers instead of cramming text into tiny cells. A true missing combination (no row for that row/column pair at all) renders as a distinct "no data" gray rather than looking like a zero.
Funnel
Ordered stage values — visits → signups → purchases, that kind of sequence — each shown as a percentage of the first stage.
- Because a funnel's meaning depends entirely on the order you typed the stages in, there's no folding an overflow stage into an "Other" bucket the way pie/treemap do — a stage past the cap is simply not drawn. The data table adds both "percent of first stage" and "percent of previous stage" (the step-by-step drop-off), which the image alone doesn't show.
Waterfall
A running-total chart — start value, then a sequence of increase/decrease bars, optionally landing on an authored "total" bar partway through or at the end.
- Increase/decrease/total bars always use the same three reserved colors, so a waterfall reads consistently across every email rather than however you last picked colors. Like the funnel, an overflow bar is dropped rather than folded, since the running-total math depends on the authored order.
Gauge
A single metric shown against good/warning/serious threshold bands you define — a needle-style dial, not a row-based chart.
- The only block in this family with no data table underneath it, for the same reason as the KPI scorecard: the value and its status are already real HTML text next to the image, not numbers locked away in a picture.
Radar
A multi-axis comparison — up to 3 series plotted across shared axes, useful for comparing categories across several dimensions at once (e.g. a product scored on price/quality/speed/support).
- Needs at least 3 axes to draw as a polygon at all — fewer than that shows the same "add more data" placeholder an empty chart gets. Unlike the rest of the family, this is a floor as well as a ceiling (capped at 8 axes).
Treemap
Proportional tiles sized by value, laid out to fill a rectangle (a "squarified" layout) rather than pie's angles — same one-value-per-category authoring shape as the pie chart, just a different shape for comparing share of a whole.
- Follows the same fold-past-8-slices rule as the pie chart, since it shares the same underlying data prep.
Sankey
Proportional flow between named nodes — From, To, and Value rows, e.g. traffic source → landing page → conversion.
- Nodes aren't authored directly; they're derived from whatever labels repeat across your From/To rows, and their left-to-right position is worked out automatically from the flow direction.
- Past the node or link cap, an overflow node is dropped, not folded — merging two distinct flow endpoints into one would misrepresent what the diagram is showing, unlike a pie chart's "Other" wedge. The data table still lists every row you authored regardless of whether it made it into the drawn diagram, so nothing you typed silently vanishes from view.
Binding to real data
KPI scorecard, bar chart, line chart, and pie chart — and only those four — can bind to a connected data source or a past campaign's own response results instead of numbers you type by hand, plus there's a one-click Dashboard project starter that seeds a KPI row and bar chart together. See Binding charts to real data for how.