I’m sure some version of this is already happening and teams are certainly embedding live dashboards inside everyday tools, but this idea has only recently started being productized.ĪirOps is super early but has a compelling vision - take the outcome of analyses to everyday tools like Google Sheets, Airtable, and Notion. Stop receiving messages like “is this up-to-date?”ĭo what they do best - analyze - instead of design/present Stick to their workflows instead of jumping between tools If the outcomes of analyses can be presented to stakeholders wherever they wish to consume them, while not worrying about data becoming stale, then analysts can: “The best interface is no interface” is an idea that Golden Krishna first shared in 2013 (he’s since turned the idea into a book ). Or the optimism towards modern BI tools such as Preset as well as spreadsheet-based analysis tools like Equals and Canvas. Not if you look at the growth in the adoption of BI tools like Mode. So, as the earliest analysis interfaces, spreadsheets and dashboards should have been long dead by 2022 - but are they? The term, OLAP, however, was coined in 1993 (only 23 years after the first OLAP product) by Edgar F. The first OLAP database was released in 1970 and it took 28 years for OLAP to become mainstream after Microsoft released its OLAP server in 1998. Microsoft released the first version of Excel in 1985 (for Mac) followed by the Windows version in 1987 ĭashboarding (or BI) tools were first developed in the 1980s but gained adoption only in the 1990s when data warehousing and OLAP (online analytical processing) databases made dashboards function properly Spreadsheet and Dashboardĭue to the ubiquity of spreadsheets and dashboards as interfaces for data consumption, I thought it might be fun to share a brief history lesson. The tools mentioned below are references, not endorsements. I’ve been thinking a lot about analysis interfaces and this post is the outcome of my thinking (and analyses). Thankfully, the fast-paced innovation over the last 2.5 years in the data tooling landscape has given us some new interfaces, and some modern takes on legacy ones. While ad-hoc analyses lead to outcomes being presented as messages, once again, I hope we can all agree that messaging as an analysis interface is a bane. Third, the analyst presents the outcome to a stakeholder for consumption - this is where the analysis interface comes into play Second, the analyst arrives at an outcome as a result of the analysis This post is all about deriving insights so hear me out:įor insights to be derived, there needs to be - an interface - to present - the outcome - of analyses - for consumption.įirst, an analyst performs an analysis by wrangling some data Infrastructure and quality aside, I hope we can all agree that data’s true potential is unlocked when people are able to, without dependencies, derive insights and take action on available data. For the last 3 years, I’ve been a keen observer of the data landscape and fortunately, I’ve had the opportunity to learn from many innovative teams who spend their days reimagining how humans work with data, and how data can do more for them.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |