How dashboardSMASHBOARD Calculates Your Card's Signals
~6 min readHow dashboardSMASHBOARD Calculates Your Card's Signals
This section explains what each signal on your card means and how it behaves — written for anyone who wants to understand what they're looking at, not for reverse-engineering the math behind it.
Status — Where You Stand Right Now
What Status answers: "Where does this metric stand right now, relative to its target?"
Status is a snapshot, refreshed at most once per hour. It does not project the future — that's Forecast.
Status shows one of these labels, from best to worst:
- Exceeding — comfortably ahead of target
- Goal Reached — target met, on a standard (hard) goal
- On Track — progressing well, on a relaxed (soft) stretch goal
- Needs Attention — modestly behind target
- At Risk — meaningfully behind target
- Critical — significantly behind target
For metrics where lower is better (refund rate, return rate, cost per click, churn), the comparison direction flips automatically — a refund rate well under its goal still shows as Exceeding.
Soft goals (dashed progress bar) get more forgiving Status bands than hard goals (solid progress bar) — a soft/stretch goal treats "still progressing" as On Track for longer before flagging it as Needs Attention.
Threshold goals (ratio metrics with a pass/fail line, e.g. an uptime SLA) skip the labeled bands entirely — you just see Above Threshold (green) or Below Threshold (red). If a card carries both a growth goal and a threshold floor, breaching the floor always forces the card to Critical, regardless of how the growth goal is doing.
A few special states you may see instead of a label:
- Early Window — too little of the period has elapsed to assess fairly (roughly the first fifth of the period); this is expected behavior, not an error, and clears on its own as the period progresses.
- Building History — a relative (grow-by-%) goal was set, but there's no prior-period data yet to compare against; resolves once a full comparison period exists, or switch to a fixed target if you'd rather not wait.
- Paused — the card was manually paused, or data is temporarily unavailable.
- Achieved — the period's goal was met and locked in.
Forecast — Where You'll Likely Finish
What Forecast answers: "If this metric keeps up its current pace, will it hit the goal by the end of the period?"
Forecast only applies to calendar-period windows (This month, This quarter) — rolling windows (Last 7 days) have no fixed end date, so Forecast shows "—" for those.
For metrics that accumulate over the period (revenue, orders, items sold), Forecast projects your current pace forward across the rest of the period. For metrics that represent a rate or a snapshot (AOV, refund rate, inventory levels), your current value already is the estimate — there's nothing to project.
Forecast shows a label just like Status does — Will Exceed, On Pace, Slipping, Off Track, Will Miss — or Insufficient Data early in the period, for the same reason Status shows Early Window: too little of the period has elapsed for a pace-based projection to mean anything yet.
The projection assumes the rest of the period looks like the average day so far — it doesn't know about day-of-week patterns, seasonality, promotions, or planned ad spend changes. When your daily numbers have been especially choppy, the Forecast cell shows a ⚠ to flag that the projection is less certain than usual.
Why Status and Forecast can disagree: Status only looks at what's happened so far; Forecast also accounts for how much time is left. Early in a period, Status often looks worse (you've only earned a fraction of the goal) while Forecast can look better (there's still plenty of time to catch up). Late in a period, the two converge. Watch Forecast early in the period; watch Status late.
Probability — Your Statistical Odds
What Probability answers: "Given the pace so far and how much this metric typically bounces around day to day, what are the actual odds of hitting the goal?"
Probability appears as a "% chance of hitting goal" pill next to Forecast. It starts from the same projection Forecast uses, then factors in how volatile your daily numbers have been — a metric that moves like clockwork gets a tight, confident estimate; a metric that swings wildly gets a wider, less certain one. A range (e.g. "42%–81%") is shown alongside the headline percentage to make that uncertainty visible rather than hiding it behind one falsely-precise number.
| Your daily data pattern | Effect on the probability range |
|---|---|
| Smooth, consistent day to day | Narrow range — high-confidence estimate |
| Typical ecommerce ups and downs | Moderate range |
| Highly variable (flash sales, irregular days) | Wide range — flagged with a ⚠ on Forecast too |
Reading Status and Forecast Together
| Status | Forecast | Interpretation |
|---|---|---|
| At Risk / Critical | Off Track / Will Miss | Both present and future signals are negative — act now |
| Needs Attention | Slipping | Trending the wrong way — close monitoring needed |
| Needs Attention | On Pace / Will Exceed | Behind today, but pace is fine — likely to recover |
| Critical | On Pace | Looks alarming, but pace is actually on track — typical early in a period |
The Alert pulse: when the ALERT flag is on, the card border glows a color derived from the Status band — green for the healthy labels, amber for Needs Attention, red for At Risk/Critical/Below Threshold.
Gap — How Far Ahead or Behind
Gap is the plain-language distance between where you are and your goal right now — shown as a signed amount ("$3,241 more to go" or "$1,200 ahead"). For metrics where lower is better, the direction flips automatically, same as Status.
Tips and Best Practices
Start simple. Begin with 3–5 key metrics on one dashboard. Too many numbers at once defeats the purpose. A good starting set for most stores:
- Revenue (Metric, with a monthly growth goal)
- Orders (Metric, with an order count goal)
- Avg Order Value (Metric, watch it closely during promotions)
- Sessions/traffic (Metric, to track visits alongside conversions)
Match your review cadence. Use calendar period windows if you review performance monthly or quarterly. Use rolling windows if you review daily or weekly.
Use multiple dashboards. One dashboard per use case works well — daily ops (rolling windows, quick morning check), monthly goals (calendar windows with OKR targets), product focus (top products + inventory for operational oversight).
Use Alert to catch issues instantly. A red pulse on a card is immediately visible even from across the room — turns your dashboard into an at-a-glance status board.
Use Trend smoothing for noisy data. If your daily numbers are naturally choppy (flash sales, promotions), apply Rolling 7-day smoothing to see the underlying trend more clearly.
Relative goals vs. absolute goals. Relative goals (grow X%) work best when you don't know the exact number ahead of time. Absolute goals (hit X) work best for fixed targets — quotas, volume commitments, etc.
Still need help?