Ask questions where the answers live
With most form tools, analysis begins by leaving: export, spreadsheet, pivot table, repeat next week. Faldaro runs the grouping, the aggregates and the search on the server, over the live submissions — and keeps the reference data behind your forms on the platform too.
-
A report builder over live submissions
Pick columns, filter, group and aggregate — sums, averages, counts, distinct counts, minimums and maximums, with date grouping by day, week, month, quarter or year. The queries run on the server against current data; nothing is copied out to be counted.
-
Pinned to the dashboard
Any saved report pins to the organisation’s dashboard and renders there as a live card, beside submission trends, workflow-state counts and top forms. The morning glance and the deep dive are the same definition.
-
Plain English to a filter
“Unpaid claims from July over $5,000” compiles into the exact filter — field keys, operators and workflow states checked against the form before anything runs. The interpretation is shown, so you can see what the words became.
-
A form’s list is its pipeline
Above every form’s submissions, its workflow states are counted filters in the order they flow — one glance for where work is piling up, one click to see only those records. The name column stays pinned while a wide table scrolls, columns you hide are remembered per form, and ticking rows brings a bar that assigns or archives them together.
-
Describe a report, refine it in the builder
Say what you want measured and a full definition — columns, filters, groupings, aggregates — lands in the report builder for you to adjust and save. AI proposes; you dispose.
-
Insight reports, on demand or on a schedule
Every form’s Insights page analyses its submissions — trends, outliers, correlations, workflow bottlenecks — with the statistics computed by the server and AI narrating them into ranked findings. Every count on a finding is an executed query, and one click opens exactly those submissions. See the AI page for the full story, including outcome prediction from your own history.
-
CSV export, without the booby traps
Reports and raw submissions export to CSV. Any cell that could execute as a spreadsheet formula is neutralised on the way out, so a hostile submission cannot attack whoever opens the file.
-
A funnel on every share link, with nothing to consent to
Every public link counts views, starts and completions — per link, per day, shown beside the link and as a dashboard funnel. No cookies, no IP addresses, no fingerprinting: the server increments a counter and stores nothing about the person. There is nothing to put in a consent banner, because nothing is collected.
-
Datasets: your reference data, versioned
Rate tables, product catalogues, region lists — edited in the browser or imported from CSV, versioned like forms, and bound to dropdowns and combo boxes. One dataset, every form that uses it, no options drifting out of step.
Mechanism
A report definition is data, never SQL
Field names are validated against the form's live schema and bound as parameters; operators, aggregates and date buckets are looked up in closed maps. Nothing a report author types is ever concatenated into a query — and every report runs under a server-enforced time budget, so an expensive one is cut off rather than taking the service with it.
Isolation
The engine has no special access
Reports execute under the same row-level security as every other query in the platform. However a definition is written — by hand, by AI, by mistake — it can only aggregate submissions its own organisation could already read.
Questions
Reports and datasets, plainly
Do I have to export to analyse anything?
No — that is rather the point. Group, aggregate and filter run against the live submissions, on the server, whenever the report is opened. Export exists for when you want the file, not because the answers live anywhere else.
Is a report live, or a snapshot?
Live. A report is a saved definition, not saved results: every run — on the report page or as a pinned dashboard card — executes against the submissions as they are now.
Where do my dropdown options come from?
A dataset: a versioned reference table you edit in the browser or import from CSV. Bind a field to a column and the options stay in one place — update the dataset, publish, and every form using it offers the new list. A second field can cascade from the first, region narrowing to district.
Can the AI invent a report over fields that do not exist?
No. A described report is validated field by field against the form’s real schema before it reaches the builder — unknown fields are dropped, operators are checked against each field’s type, and nothing is saved until you save it.
Can a report see another organisation’s submissions?
No. Reports run under the same row-level security as every other query — the report engine holds no special access, so a definition, however written, can only ever aggregate what its organisation could read anyway.
Retire the Tuesday export
Build the report once, pin it to the dashboard, and let the spreadsheet be something you choose to make — not the place the answers live.
No card required.