Daily Movers is a daily briefing in your domain dashboard that shows which key metrics moved materially — and, for each one, surfaces a ranked shortlist of suspected contributors. It is a correlation tool, not a causation engine.
What Daily Movers is (and is not)
Daily Movers is a signal-surfacing tool, not a root-cause guarantee.
It identifies metrics that moved beyond normal day-to-day variation and ranks the candidate changes most associated with that movement. It does not:
- Prove that any listed contributor caused the movement
- Run experiments or controlled tests
- Access your internal systems, code repositories, or deployment pipelines directly
- Replace a post-mortem or engineering investigation
Every contributor in a Daily Movers card is described as a suspected contributor or likely candidate. The confidence bands reflect the strength of statistical association — not a confirmed cause. Use Daily Movers to narrow down where to look, then investigate through the linked surfaces.
Plan requirement: Daily Movers is available on Pro, Scale, and Enterprise plans. Not available on Free.
Metrics watched
Daily Movers tracks five key business metrics:
| Metric | What it measures |
|---|---|
| Revenue | Transactions, order values, and revenue totals for your tracked goals |
| Signup conversion | The rate at which visitors complete your signup goal |
| Checkout completion | The rate at which visitors reach a successful checkout event |
| Bounce rate | The share of sessions that end without a second interaction |
| Exit rate | The share of page visits that are the last in a session |
How it works
Each day, a deterministic engine compares the previous 24-hour window against a trailing 7-day baseline for each metric. A move must clear both a magnitude threshold and a minimum-volume floor to count as material. Low-volume sites or low-traffic pages may not meet the floor — those potential movers are flagged separately (see "Possible noise" below).
For each material mover, the engine collects candidate changes from the same time window:
- Code deploys and Git commits — new deployments or commit clusters that coincide with the movement window
- New or spiking error clusters — error groups that appeared or grew sharply during the window
- Web Vitals regressions — Core Web Vitals that crossed a threshold (LCP, INP, CLS) during the same window
- Traffic-source shifts — meaningful changes in channel mix (organic, paid, direct, referral)
- AI-source-mix shifts — changes in the proportion of AI-driven referral traffic
- Reconciliation context — where reconciliation data is available, a note on whether the move likely reflects a real change or a measurement/delay artefact
- Segment-mix changes — shifts in the composition of device type, geography, or browser that could arithmetically explain the metric movement
Each candidate is ranked by a weighted association score that accounts for temporal proximity (did the change happen just before the metric moved?), overlap with affected routes or segments, the magnitude and novelty of the change, and whether supporting metric deltas point in the same direction.
The engine then writes a short plain-language summary rephrasing the ranked evidence. The deterministic ranking is the authoritative output; the summary only restates what the ranking already shows and never adds information not present in it.
Confidence bands
Each suspected contributor displays a confidence band:
| Band | What it means |
|---|---|
| High | Strong statistical association — multiple supporting signals align in time and scope |
| Moderate | Reasonable association — some signals align but the picture is not complete |
| Low | Weak association — the change is present but the timing or scope overlap is limited |
| Tentative | The association is plausible but the supporting evidence is thin |
A "High" band does not mean the contributor caused the movement. It means the association is strong. Always verify through the linked surface before drawing conclusions.
Possible noise
Movers that cleared the magnitude threshold but fell below the minimum-volume floor are not shown in the main list. They appear in a collapsed "Possible noise" section at the bottom of the card. This keeps the main view focused on signals that have enough data behind them to be meaningful, while still giving you visibility into smaller fluctuations that may warrant a closer look once you have more data.
Reconciliation context
Where reconciliation data exists for a mover, each card includes a short note on whether the measured movement likely reflects a genuine change in user behavior or whether it may be a measurement artefact — for example, delayed event ingestion, a tracking gap during a deploy window, or a data-pipeline catch-up. This context helps you distinguish a real business signal from a counting anomaly before escalating.
Drill-in links
Where a contributor has supporting data, it includes a one-click jump to the relevant dashboard tab so you can verify directly. The jump applies the relevant filters and switches you to the target tab:
- Error clusters → Errors tab
- Web Vitals regressions → Performance tab (LCP, INP, CLS)
- Traffic-source shifts → Analytics tab, filtered to the channel that moved
- Deploy or commit markers → Analytics tab, where deploy markers appear as annotations on the visitor chart
You land in a filtered, in-context view rather than at the top level. Contributors without supporting drill-in data are still listed, but without a jump link.
Date range and history
Use the range selector at the top of the Daily Movers tab to change the lookback window:
| Plan | Maximum range |
|---|---|
| Free | Not available |
| Pro | Up to 90 days |
| Scale | Up to 1 year |
| Enterprise | Up to 1 year |
History accrues daily. If your account is new, or you recently added a website, Daily Movers may show fewer days than the maximum until enough baseline data has accumulated.
Reading a Daily Movers card
Cards are collapsed by default. The collapsed view shows:
- The metric name and the direction of movement (up or down)
- The measured movement: baseline value → current value, plus the magnitude (e.g. "−18 % vs 7-day average")
- The top suspected contributor with its confidence band
Expand a card to see:
- The full ranked list of suspected contributors, each with its confidence band
- The plain-language summary
- The reconciliation note (if available)
- One-click drill-in links for each contributor
Empty state: When all tracked metrics stayed within normal day-to-day variation, the tab shows "No material movers detected" — this is the expected and healthy state on most days.
Shared dashboards: Daily Movers can appear on a public or shared dashboard link when the workspace is on a Pro plan or higher. The cards stay aggregate-only — no individual visitor, session, or personal data crosses the share boundary.
What V1 does not do
To keep expectations honest:
- Daily Movers does not prove causation — every contributor is a correlation and a likely candidate
- It does not integrate directly with your CI/CD pipelines or code repositories; deploy and commit signals come from data you have configured in Zenovay
- It does not surface movers for metrics that fall below the minimum-volume floor in the main list (those appear in "Possible noise" if present)
- It is not available on Free plans