Ask a question. Trust the answer.
Connect your data, ask in plain English, read the chart, then pin it to a dashboard and have it sent to your team.
Which regions grew revenue fastest last quarter?
Revenue growth by region
Quarter over quarter
End to end
Five steps to a dashboard.
Nothing here is a black box. You can see every step, and the one that checks the number stops when it cannot be sure.
01
Connect read-only
Give iDash a read-only login. It can look at your data and never change it, and your credentials are encrypted before they are stored.
02
Agree what the numbers mean
iDash reads what tables you have and how they relate. Your team then names things in plain business terms and settles what revenue actually means.
03
Ask the question
Ask in plain English. iDash picks from the measures your team already agreed on, so it is not inventing a definition on the spot.
04
Prove it or stop
iDash works out where combining tables would double a total, and counts each customer once even when they have twenty orders. If it cannot be sure, it says so.
05
Pin, schedule, share
Pin the chart to your project dashboard, put it on a schedule, set an alert on the number, or drop the view into a product your customers already use.
One shared definition of every number.
Revenue means one thing across your whole company. iDash starts from your own data, then your team decides what each number really means.
- Starts from your own data
- iDash reads what tables you have and how they relate, so you start with something real instead of a blank page.
- Curated by the people who know
- Rename fields into the words your team actually uses, hide the ones nobody should touch, and mark what is safe to add up.
- One definition of revenue
- Define revenue once, with the rules that go with it, and every question that mentions revenue gets the same answer. It stops living in someone's private spreadsheet.
- Connections it can trust
- iDash knows how your tables link up, which is what lets it count each record once. When two ways of connecting the data disagree, it refuses rather than picking one.
- order_idOrder IDUnique ID
- customer_idCustomerForeign key
- placed_atOrder dateTime
- net_totalRevenueMeasure
Metrics
- Revenue
- Sum of net total
- Orders
- Count of distinct orders
- Average order value
- Revenue divided by orders
Ask like you would ask a colleague.
Type the question the way you would say it out loud. When it needs real number crunching, iDash does that in a sealed workspace.
Questions iDash answers straight away
- What was revenue by region last quarter?
- How many customers ordered more than once in the last 90 days?
- Show monthly recurring revenue by plan for the last twelve months.
- Which accounts have not ordered since March?
- What is the refund rate by product category this year?
It picks, it does not guess
iDash chooses from the measures and groupings your team already agreed on, so an answer never depends on it remembering the shape of your database.
A fallback that admits what it is
Some questions sit outside your agreed definitions. iDash can still answer those, and every answer shows where the number came from, so nobody mixes up the two.
A sandbox for the harder maths
Cohorts, forecasts and reshaping run in a walled-off workspace that sees the results and nothing else. No internet unless you allow it, and never your database login.
Charts checked before they render
Every chart carries the figures behind it. That check is why a chart you pinned last month still shows a real number today.
Pin the question, not the picture.
A pinned chart remembers the question behind it. Refresh asks your data again, so the tile is as current as your database.
Revenue analytics
Saved results. Every tile shows what it got when it last ran.
Net revenue
Monthly, trailing 12 months
Orders
12,847
+11.2%vs the prior 90 days
Average order value
$333.30
+4.0%vs the prior 90 days
Revenue by channel
Quarter to date
- Organic search37.0%
- Paid social24.0%
- Email19.0%
- Referral12.1%
- Direct7.9%
Revenue by region
Quarter to date
Pin from the conversation
Any chart in a chat pins to the project dashboard in one click, along with the question, the filters, and the note about where the number came from.
Refresh is a re-run
There is no stale picture pretending to be current. Refresh asks the question again, under the same read-only login and the same counting rules.
Arranged for the people reading it
One dashboard per project, ordered the way your team runs a review: the headline number first, the breakdown under it, the detail on request.
A dashboard you did not have to build.
Answers that arrive on schedule.
Run a question every morning, watch a number for a threshold, and send the result to Slack, to email, or into your own product.
Automation
1 trigger · 1 condition · 2 outcomes
Every weekday at 07:00 UTC
Compared with the trailing 4 week mean
#revenue-alerts
No notification sent
Scheduled runs
Put a dashboard or a single question on an hourly or daily cadence. Each run asks your data again, so the number in the message is live.
Alerts on the number, not the chart
Set a threshold on refund rate, pipeline coverage or anything else you have defined, and iDash tells you when it crosses. Silence means nothing crossed.
Delivered where work happens
Results land in a Slack channel or an inbox with the chart, the number, and where it came from, so the thread does not turn into an argument about the source.
Embedded in your product
Put a dashboard or a single chart inside the application your customers already log into, under the same access rules that apply inside iDash.
#revenue on Slack
Revenue last week was $412K, up 6% on the week before. Refund rate held at 3.1%.
- Monday revenue digest
- Every Monday at 08:00
- Refund rate above 4%
- Watching, last checked 6 minutes ago
Who sees what
People see only what they should.
Roles, workspaces, and rules about rows and columns come from who signed in. Nothing typed into a chat can widen what a person sees.
Four roles, ranked
Owner, admin, member and viewer. Every workspace checks who you are before it shows you anything, and a workspace can never be left without an owner.
One workspace cannot see another
A workspace you do not belong to does not tell you it exists. You get nothing back, not a polite refusal that confirms the name.
Down to the row and column
Show a regional manager only their own region, or hide salary from everyone outside finance. The rules follow the person who signed in, not the question they asked.
Credentials that stay put
Connection details are encrypted where they sit. They are used only to open a read-only session, and are never handed to the AI or the sandbox.
See it against your own data.
Connect a read-only copy of your data, ask a question you already know the answer to, and check what comes back.