Analyze & Build

This article covers how to explore your Labguru data in Analytics: building queries, summarizing and reshaping results, adding your own calculations, and saving your work to return to later. When you are ready to present what you have found, Visualize & Present explains how to turn results into charts and dashboards. You can start a query in two ways — by asking the Analytics Assistant in plain English, or by building it manually in a workbook. Everything below applies to both, because a query the assistant drafts is a normal query that you can keep refining.

This guide does not need to be read from cover to cover. If you are new to Analytics we recommend reading Querying Data first, then turning to the topics that interest you most.

What you can do

The features described here are the ones available to everyday analysts, covering two roles. Access is controlled by a single Labguru permission:

• Access Dev Tools grants build access: users with this permission can build and edit queries, write calculations, add their own data, save analyses for reuse, create dashboards, and publish and share them with others.
• All other users have view access: they can open dashboards shared with them, filter and drill into the data, and download results, but cannot build or edit queries or dashboards.

Tasks that belong to system administrators and data modelers, such as defining the underlying data model or managing connections, are outside the scope of this guide.

Querying data

Workbooks are the core analysis environment in Analytics. Each workbook holds one or more tabs, and each tab is a query you build and refine. Because every query runs against the curated data model, results stay consistent and respect your permissions no matter how you build them.

Results are cached, so re-running the same query returns quickly. If you need the latest figures, refresh the query to bypass the cache.

Ways to build a query

You can start in either of two ways, and refine the result by hand afterwards:

• Ask the Analytics Assistant — open Analytics Assistant in the Analytics sidebar and describe what you want in plain English. The assistant drafts a query and a result, which you can then adjust with the field picker. Check the result before relying on it. See Analytics AI Assistant.
• Build it manually in a workbook — select fields from the field picker and results appear automatically. This is the most flexible everyday method.

Selecting fields

Fields are split into two groups in the field picker, each nested under the dataset it belongs to:

• A dimension is something you group by, such as a date, project, or sample type.
• A measure is an aggregation over your dimensions, such as a sum, count, or average.

For example, to see total samples logged per day, you would pick a date dimension and a count measure. As soon as you select fields, the query runs and the results table fills in. Click an active field again to remove it.

Use the search box above the field picker to find fields by name. Hovering over a field and clicking the options icon (or right-clicking it) reveals additional actions — such as changing how a field is aggregated, or the timeframe a date is grouped by.

Limiting rows

A query returns a limited number of rows by default; you can raise the limit from the row limit menu, and the results table paginates so you can scroll through them. The limit only affects how many rows are shown, not how many are processed.

Caution. Use high row limits with care. Displaying very large result sets at once can slow your browser.

Working across multiple tabs

A workbook can hold several tabs, each with its own query. You can rename and reorder tabs, and duplicate a tab to reuse a query. A linked tab shares the same underlying query, so several visualizations stay in step when you change the source query once.

The results of one tab can also act as a filter on another — useful for defining a subset on one tab and exploring it on a second.

Controls

Controls let a viewer change what a query shows without rebuilding it — for example switching a date between daily, weekly and monthly, or toggling between two measures.

Controls added in a workbook do not automatically appear on a dashboard. See Dashboard Filters and Controls.

Filtering and sorting

Filters let you focus a query on the rows you care about, and sorting controls the order in which results appear. Both are applied directly from the field picker or the results table.

Adding a filter

Any column can be added as a filter. The options you see depend on the field type, with different controls for dates, numbers, and text.

Filters support the options you would expect for each field type — for example filtering on several values at once, filtering dates by an absolute range or relative to today, and combining more than one condition on the same field. When you filter a field, Analytics suggests possible values so you can pick from a list rather than typing them by hand.

Sorting

Click a column header in the results table to sort by it. Clicking again cycles through descending, ascending, and unsorted.

Removing fields and filters

• Click an active field in the field picker to remove it from the query.
• Open the options menu (or right-click) on a column header in the results table and remove it from there.

Pivots, totals and formatting

Once a query is returning results, you can reshape and tidy the results table before you chart or share it:

• Pivot a field so its values run across the top of the table as columns.
• Add column and row totals.
• Format number and date columns.
• Fill in gaps in the data, such as a date with no activity.

These options live in the results table and its Options menu, and affect how the data is presented, not the data itself.

The results table is a working view of your query. How a chart looks on a dashboard is controlled separately in the Chart tab — see Creating Visualizations.

Calculations and custom fields

When the fields in a dataset do not give you exactly what you need, you can extend a query in two ways:

• Table calculations work on the results after a query has run, much like formulas in a spreadsheet — for example a percent of total, a running total, or a period-over-period change.
• Custom fields define new reusable dimensions and measures that become part of the query itself. Once saved, a custom field appears in the field picker like any other.

Both work on top of the curated data model, so results stay consistent and respect your permissions.

Creating a custom field from existing modeled fields is available to any analyst who can edit the workbook. Writing SQL references inside a custom field requires the Access Dev Tools permission.

SQL in field definitions

Free-form SQL querying is not enabled in Analytics — you do not need to write SQL to build a query. SQL does appear in field definitions: when you edit a field (from a column's context menu, choose Modelling > Edit), a user with Access Dev Tools can define how it is calculated or aggregated using a SQL expression.

Learn more

These features come from Omni, the platform Analytics is built on, and Omni's help center covers them in much more depth, including every supported function and worked examples:

• Using Excel-style functions to create table calculations
• Creating custom fields in Omni workbooks

Omni's pages describe the full product, so some options may differ in Labguru, depending on your permissions and what is enabled for your organization.

Adding your own data

Not everything you want to analyze lives in the main Labguru database. Analytics lets you bring in your own data, either by uploading a file or typing it in, and gives you a spreadsheet environment for working on top of live query results.

Uploading a CSV or Excel file

A data input table lets you add your own data — by uploading a CSV or Excel file, or typing it in — without touching any data pipelines. This is handy for enrichment or a one-off exploration. Once added, the table is available in the field picker like any other, and you can query it and combine it with other data.

Adding data input tables requires the Access Dev Tools permission, and the upload ability must be enabled for your organization and the document.

Spreadsheet tabs

A spreadsheet tab gives you an Excel-like surface inside a workbook that connects to your queries. You can format, reshape, and build on top of live data using familiar formulas, without exporting the data or refreshing it by hand. Because the spreadsheet is connected to the query, it stays current as the underlying data changes.

This is useful when you want to lay out results in a particular way, add supporting calculations alongside a query, or prepare figures for a report.

Saving and reusing analyses

Workbooks can be saved, versioned, copied, and explored, so you can return to your work and build on it without losing earlier states.

Saving a workbook

Save a workbook from File > Save, giving it a name and choosing the folder to save it into. You can move it later.

Where you save a document determines who can reach it. A document saved in My documents is only visible to others if you explicitly share it. See Organizing and Finding Content.

View and edit modes

• In view mode you are looking at the latest saved version.
• In edit (draft) mode your changes save automatically as you work. Edits also affect any dashboard attached to the workbook.

Version history

To see the history of changes to a workbook, click File > Version history. The panel lists changes with timestamps and authors, and you can restore an earlier version from here.

The version history covers changes to the workbook itself, not to attached dashboards.

Exploring without changing the original

To experiment without affecting the saved version, click Explore near the top right. This opens a fresh copy of the workbook that you can work in safely.

Reusing queries

• Copy a tab into another workbook to reuse its query there. The destination workbook must use the same connection and shared model.
• Save a query as a view so it becomes a reusable building block for other queries. Saving a view requires the Access Dev Tools permission; making it available outside the current workbook requires a modeler to promote it to the shared model.