The data ecosystem is rapidly evolving, and that means better interfaces are being built to consume data and analytics for everyone in an organization.
Traditionally, the dashboard has been used as the cure-all for all data consumption in an organization. Tools like @tableau, @LookerData and @Domotalk are the de-facto ways for business users and technical users to consume analytics.
There are two new interfaces that are being introduced in the market. 1/ the notebook/doc and 2/ the app.
1/ The notebook/doc format: tools like @hyperquery, @_hex_tech, @DeepnoteHQ and @observablehq are pioneering this space. It's a notebook/doc format that turns into reports and apps. It can be used for exploratory analytics as well as productionized analytics.
2/ The app format: tools like @streamlit, @plotlygraphs, @datapaneapp, and @TopcoatData allow technical users to use python or sql to build interactive web applications. It's not used for exploratory analytics; it's used for productionized applications.
These three data interfaces (notebook/doc, dashboard, app) are as fundamental as the doc, the slide, and the spreadsheet in traditional productivity.
As analogy: 1/ The analytics doc is to the traditional doc.
2/ The dashboard is the traditional slide.
3/ The app is the traditional spreadsheet (it's most often used as a business application and a SSOT).
Yes, there are overlapping feature-sets between these three interfaces. But as the data ecosystem matures, all three interfaces will thrive and will be used for different use cases in a data-driven organization.
โข โข โข
Missing some Tweet in this thread? You can try to
force a refresh
There is an industry-defining unbundling of BI happening right now. The BI app is unbundling into two composable parts - the metrics layer and the consumption layer. A ๐งต๐:
Firstly, WHY is this happening? A massive tailwind propelled by @SnowflakeDB and @getdbt. In particular, a well engineered, infinitely scaleable analytics warehouse and a transformation and modeling layer that extends into the metrics layer.
"Accessories" to the data warehouse help tremendously as well: @fivetran, @AirbyteHQ, @getcensus, and @HightouchData for forward and reverse ELT (piping data to and from operational tools - e.g. @salesforce - the data warehouse) are the must-haves.