Getting started
In about 15 minutes you'll build a flow in Strata, chart its result in Vista, save a version, read a diff and go back to that version. You need a desktop browser. You don't need to sign up.
What you'll build
A flow that joins retail orders to retail stores and totals revenue for each country, and a dashboard with one column chart of that total.
As a guest, your project runs in your browser and syncs to a temporary My Cloud. Guest copies are deleted about 30 days after the guest session started.
Step 1: Add the two sample tables
- Open Strata. You see an empty canvas with a block palette above it.
- In the palette, select IO. Drag Input onto the canvas, or double-click it. A single click only shows a description.
- Select the block, then look at Properties on the left. Open the dropdown that says Local File and choose Example.
- Open Dataset and choose Retail Orders. The preview under the canvas shows columns starting with
order_idandstore_id. - Drag a second Input onto the canvas, below the first. Select the new block, then choose Example again, then the Retail Stores dataset. It has 40 rows and 5 columns, including
store_idandcountry.
A red message such as file_path is required is normal until you pick a dataset.
Step 2: Join orders to stores
Both tables have a store_id column, so that's the join key.
- In the palette, select Combine and drag Join to the right of the two inputs.
- Drag from the dot on the right of the Retail Orders block to the upper dot on the left of Join. That's the left input.
- Drag from the dot on the right of the Retail Stores block to the lower dot on the left of Join. That's the right input.
- Select Join. Under Join Keys, select Add.
- Set Left Column to
store_idand Right Column tostore_id. - In the preview, select OUTPUT INNER. You should see 14 columns: the 10 order columns followed by
store_name,city,countryandstore_format.
Join has three output dots, in this order from the top: OUTPUT LEFT (orders with no store), OUTPUT INNER (matches) and OUTPUT RIGHT (stores with no orders). See Join for details.
Step 3: Total revenue by country
- In the palette, select Transform and drag Summarize to the right of Join.
- Drag from the middle output dot of Join to the input dot of Summarize.
- Select Summarize. Under Actions, select Add. Leave Action as Group by and set Column to
country. Output fills in by itself. - Select Add again. Set Action to Sum, set Column to
net_sales_eur, and typerevenuein Output. - The preview now shows 10 rows and 2 columns,
countryandrevenue. - Open the dropdown that says Preview: 100 rows and choose Preview: All rows. The default is a sample, so its totals aren't the real totals. Finland is now highest at about 11,758, followed by Sweden and Belgium.
Step 4: Send the result to Vista
- In the palette, select IO and drag Vista Source to the right of Summarize.
- Drag from the output dot of Summarize to the input dot of Vista Source.
- Select Vista Source. Display name holds a random name. Replace it with
Revenue by country.
A dashboard can only use tables that reach a Vista Source block.
Step 5: Add a chart
- At the top of the page, select Vista. You see an empty page called Page 1.
- With Bars & columns selected in the palette, drag Columns onto the page.
- The chart appears straight away. Select it and open the Data tab in Properties. Data source shows
Revenue by country, Group by iscountry, Bar value isrevenueand Operation is Sum.
Vista chose the source because the project has only one. With more sources, select Change next to Data source and use the Table dropdown to choose one.
Step 6: Save a version
Your work autosaves, but only a saved version gets a name you can find later.
- On the right, select History.
- In What changed?, type
Revenue by country, then select Save. - A new row with your message appears at the top of the list. It has a small colored tag with a branch name that Oplema picked, such as
papa.
Step 7: Change something and read the diff
- Select Strata at the top, then select Summarize.
- In the second action, change Action from Sum to Average. The preview now shows averages.
- In History, a row named <current> marked unsaved appears at the top. Leave it unsaved.
- Select the row below it,
Revenue by country. On the canvas, Summarize is highlighted because it changed. Everything else is dotted. - Above the canvas, select Script, then select the
Revenue by countryrow again if needed. You see the changedactionsline, with the old line in red and the new line in green. - Select Graph to return to the canvas.
Step 8: Open the earlier version
- On the
Revenue by countryrow, select Open. - Select Summarize. The action is Sum again.
- Look at the list. Your average change wasn't lost. It's now a row named <current> with an Open button, which brings it back.
Saves explains what happens when you edit from an older version, and Diffs explains the Logic, Canvas and Full buttons.
Compare with the finished project
This is the version you saved in step 6, with Preview: All rows: Finland is highest at about 11,758.
Finished Getting started project: flow and dashboard
Loading interactive example…
The live example needs a desktop screen. The result it produces is shown here, and the explanation and settings below work on any screen.
| country | revenue |
|---|---|
| Belgium | 11554.9 |
| France | 9688.5 |
| Norway | 8575.3 |
| Finland | 11758.3 |
| Ireland | 8443.5 |
| Germany | 8486 |
| Sweden | 11650.3 |
| Austria | 10770.2 |
Where to go next
- Learn Strata step by step. The first course, Build a report from a CSV export, goes deeper, with hands-on lessons that turn one shop's export into a weekly report you can check. It starts with Build your first data flow from a CSV.
- Sign up to keep it. Select Sign up at the top right, in this browser. You also need an account for the next three.
- Ask Elara to change it. Start at Set up Elara.
- Run it on a schedule. Scheduling needs an Output block with a saved destination. See Scheduling.
- Publish the dashboard. Give people a link to it. See Publish.
- Find any other task. See How-to guides and the Glossary.
Prepare your data. Build it in Strata.
Strata opens in your browser, with no account.