Skip to main content

Scorecards

A Scorecard is a matrix of metrics × time periods — one row per metric, one column per period (week, month, quarter, or year). Each cell shades from cool to warm so you can spot trends across rows and seasonality across columns at a glance.

Scorecards live at the top-level Scorecards page in the nav.

Creating a scorecard​

  1. Go to the Scorecards page and click New Scorecard
  2. Give it a name (selecting a single location auto-fills the name with that location)
  3. Set the granularity (Week, Month, Quarter, or Yearly/YTD) and date range (preset windows or a custom start/end)
  4. Optionally pick one or more locations — empty selection means all locations
  5. Add rows (see below)
  6. Click Save

Adding rows​

Each row is a metric the scorecard computes once per column period. Pick a row source from the Event / Report / Computed toggle above the composer:

  • Event metrics — aggregate raw events directly. Choose a source (revenue, count, etc.) and an aggregation (sum, average, min, max). Count sources also offer Total vs Unique — e.g., total visits versus unique visitors.
  • Report metrics — point at one of your Reports, then choose which value column to track and how to aggregate it across rows. Any report SQL file with the right header annotations is selectable as a scorecard source.
  • Computed metrics — combine two existing rows with arithmetic (see Computed rows). No new query runs.

Each row can carry its own:

  • Label that appears in the leftmost column
  • Filters — category filters, rollup filters, and where applicable:
    • First-time only — limits the metric to each customer's first matching event
    • First-ever — limits to customers whose first-ever event of this type was the filtered one (true new-client acquisition by SKU)
  • Polarity — higher is better or lower is better. Polarity controls the color gradient direction (green-up or green-down) so a falling churn rate and a rising revenue both shade as wins.
  • Format — currency, percent, or number, inferred from the metric and overridable per row

Computed rows​

Computed rows do in-browser arithmetic over two rows you already have — no extra query. Use them for ratio KPIs like revenue per visit.

On the Computed tab:

  1. Pick the Top (A) row
  2. Pick an Operation — Divide (A per B), Multiply, Add, or Subtract
  3. Pick the Bottom (B) row
  4. Set the Format, Direction, and Label (the label auto-fills, e.g., "Revenue per Visit")
  5. Click Add metric

The new row computes per period from the already-loaded values of its two source rows. If a cell can't be computed — a source value is missing, or you'd divide by zero — it renders as a struck-through dash rather than a wrong number. Add at least two rows before the Computed tab has anything to combine.

Comparison rows​

Turn on the Comparisons control to show a small percent-change row beneath each metric:

  • MoM % (or WoW / QoQ, matching your granularity) — each period versus the previous one
  • YoY % — each period versus the same period a year earlier

Choose one or both. Year-over-year isn't offered at weekly granularity (weeks don't align across years), and at yearly granularity only year-over-year applies. Deltas are colored by the row's polarity (green for an improvement, red for a decline) and are blank for the in-progress period.

Year-to-date columns​

At week, month, or quarter granularity, tick Show YTD vs prior year to append three summary columns on the right: prior-year YTD, current YTD, and the year-over-year Δ%. Both years are cut at the same point in the calendar so the comparison is apples-to-apples. (Yearly granularity is already YTD, so the toggle is hidden there.)

The in-progress period​

The current, incomplete period (this month, quarter, or week) renders as a gray, month-to-date column. It's excluded from each row's color scale and from comparison deltas, so a partial value can't skew the gradient or read as a catastrophic drop.

Editing rows​

  • Reorder — drag the grip handle on the left of a row to reorder
  • Edit — click the row label to reopen the metric composer with the current config
  • Delete — hover the row and click the trash icon. A confirmation prompt prevents accidents

Computed rows reference their sources by identity, so reordering or relabeling a source keeps the computed row correct. Deleting a source leaves the computed row's cells struck through until you re-point it.

Headers and formatting​

Column headers stack a year band over the short period label (e.g., 2025 over Jan, Q1, or Dec 15). Bold dividers mark year transitions so multi-year scorecards stay legible.

Cells are auto-scaled per row using a Tukey-fence so one outlier month doesn't wash out the rest of the gradient. Values format according to the row's format setting:

  • Currency — $X.XX, dropping cents at $1,000 and above (cents are noise when scanning a row)
  • Number — comma-separated, up to two decimals, and likewise no decimals at 1,000 and above
  • Percent — one decimal, e.g., 12.4%

Exporting to Excel​

Click Export in the scorecard editor to download the matrix as an .xlsx file — every metric row, its comparison and YTD columns, and the period headers, matching what's on screen. Numbers carry real Excel number formats (currency, percent, plain) so they stay sortable and chartable in the spreadsheet.

Saving and sharing​

Scorecards are saved per organization. The scorecard URL /scorecards/<id> is shareable — anyone on your team with access to the underlying data sees the same matrix.

Save As New in the editor duplicates the current scorecard with a new name. Useful for building a per-location variant from a template.

On dashboards​

Like Saved Views and Report Looks, a scorecard can be embedded on a Dashboard as a widget — pick it from the Add Widget picker.