Administrator guide: analytics and insights

Zion’s analytics suite gives administrators four connected views of school performance: time-series trends, subject and centre breakdowns, termly-goal predictions, and pattern insights. Administrators see data across the whole organization. Supervisors see the same tools scoped to their assigned centre or centres. This guide covers how to read each view and turn a finding into an action.

​

Where to find analytics

Navigate to Analytics in the sidebar to reach the analytics index, which links out to Trends, Centres, Subjects, Predictions, and Insights, plus a student picker for drilling into an individual record. The dashboard also embeds a small insights widget showing your top current insights, with a link through to the full Insights page.

​

Trends

The Trends view charts goal-achievement over time for a student, subject, or centre. Pick a window of 7, 30, 60, or 90 days, or the current term, and the chart updates accordingly. From a student’s trend detail page you can also see:

• Consecutive-misses highlighted directly on the chart
• A subject-level breakdown for that student
• A comparison against their centre’s average
• An intervention notes panel
• A printable Student Progress Report

Use trends to confirm whether a pattern flagged elsewhere (a low centre score, a flagged insight) is showing up in the raw achievement data over time, not just in a single snapshot.

​

Subject and centre breakdowns

​

Subjects

The subject view shows goal-achievement by subject, a grade-band by subject matrix, a PACE-level completion-rate table, and a radar overview. Filter by date range (7 days, 30 days, or current term), grade band (1 to 4, 5 to 8, 9 to 12, or all), and subject. Export the filtered view as CSV or PDF.

​

Centres

The centre view compares learning centres on goal achievement: mean and standard deviation across centres, outliers highlighted, a comparison chart, a top-5 and bottom-5 trend chart, and a full metrics table. Export as CSV.

​

Predictions

Predictions project whether each student is on track to meet their termly PACE-count goal, based on their completion velocity so far this term. Every student and subject gets one of three statuses: The Predictions page shows a donut summary of on-track, at-risk, and off-track students, plus searchable and filterable student cards (by centre, status, and subject). Click a card to open a detail panel for that student and subject. Export the current at-risk and off-track list as CSV from Export At-Risk Students. Early in the term, a student with zero completions is not automatically marked off track. Zion checks how far the term has progressed: less than a quarter of the term elapsed still counts as on track, less than half counts as at risk, and later than that counts as off track.

​

Pattern insights

Pattern Insights runs six statistical detectors across the school’s goal data and surfaces findings you might not notice by scanning trends manually. Each insight carries a relevance score from 0 to 100 and a confidence level of high, medium, or low. Confidence reflects sample size: a finding based on a handful of goals is flagged low confidence even if the pattern looks strong, while a finding backed by dozens or hundreds of goals is flagged high confidence. Low confidence does not mean the finding is wrong. Treat it as worth watching rather than acting on immediately.

​

Generating and refreshing

Insights regenerate automatically on a weekly cycle. You can also click Generate Insights to run a fresh analysis on demand, useful after a batch of new data or before a leadership meeting. Generating replaces the current non-dismissed set with a new one. Insight generation, dismissal, and export require an active subscription on your organization.

​

Filtering and export

Filter insights by confidence (all, high, medium, low) or by type (day-of-week, subject correlation, centre performance, seasonal trend, student velocity, subject difficulty). Insights sort by relevance score by default. Export a PDF of the full insight list, or export a single insight, from the Insights page.

​

Building insight history

Pattern detection needs enough goal data to find a real signal. A school that has just started using Zion will see few or no insights at first. As daily goal data accumulates over several weeks, more patterns become detectable. Consistent daily goal marking across centres is what feeds this.

​

Related pages