Annual Tableau Conference 2023 was held in the second week of May in Las Vegas and Tableau announced some new product innovations that the company boldly sees transforming analytics, stating a new day for data is dawning. We summarize here the main product announcements and other conference key takeaways.
Generative AI has been the hype trend of the year since Chat GPT launch and Tableau is surfing the wave by announcing Tableau GPT, a generative AI engine that is aiming to create next-generation analytics experiences in various areas. Tableau is owned by Salesforce and Einstein GPT, a generative AI engine specific to Salesforce CRM data analysis was released already in March 2023.
The opening keynote of the conference highlighted a few areas where Tableau GPT could rise and shine. The feature called Tableau Pulse is powered by Tableau GPT and the key capabilities emphasize guided, personalized experience and data discovery. Pulse promises to deliver personalized metrics and insights in understandable language in the tools that people commonly use, e.g., web, email, and messaging tools like Slack, which is also part of the Salesforce ecosystem. Tableau already has some AI driven features that have natural language capabilities like Ask Data, and Pulse would probably replace some of them by taking the AI driven features to the next level. Tableau Pulse was announced to be available for piloting during H2 2023 - we already signed up to be the first ones to get info about the availability!
Demo screenshot of Tableau Pulse insights, presented at Tableau Conference.
Another feature utilizing Tableau GPT was called Analytic Assistant where the aim is to accelerate analyst productivity e.g., by aiding the creation of regular expression statement which was demoed at the classic Devs on Stage session. When using current modern sophisticated tools like Tableau, the productivity is relatively high: sophisticated analysis and insightful dashboards can be created very fast compared to some other data-related tasks, given the data layer is well modeled. In business intelligence projects typically most of the time is spent in data engineering, gathering data, polishing, and modeling it for analysis use. Our experience is the data layer build and testing often takes at minimum 75% of the project budget. Seeing demos where the creation of some code lines is automated is cool, but in analytics, the possibility to write code quickly is maybe not the biggest concern. A much greater productivity jump potential is probably available in the data layer. Would the Generative AI be able to automate data integration or modeling tasks? Hopefully soon! Generative AI definitely has a lot of potential, but it is not always necessarily a question of productivity but more of insight, at least when we talk about analytics. Getting some guidance to answer the difficult “Why?” -questions is more than welcome!
The Tableau Community, #datafam, is really something unique and is a big part of what makes Tableau so special. In this year’s conference, there was more focus than ever before on the community. During the Opening keynote, this year’s Visionaries (The highest level of appreciation from Tableau for a community member) were up on stage to receive the celebration they truly deserve. There was also a keynote session highlighting some of the visionaries and their story (available on Salesforce+).
There were also more community sessions available in the conference than ever before. Community events like Makeover Monday (https://www.makeovermonday.co.uk/), Back to Viz basics (https://www.thetableaustudentguide.com/vizbasics) and Workout Wednesday (https://workout-wednesday.com/latest/) were all hosted live during the conference. These are regular events where newbies and more experienced users alike can join in weekly challenges to learn more about using Tableau.
Knowit Tableau experts Elina Vihriälä and Matz Rasmus at Tableau Conference #DataFam sign.
Diversity was a big theme this year, and there were many different scheduled meetups for different groups such as Communi[datos] (for Spanish speaking datafam), Black in Data and Data Pride. These meetups provided an opportunity to meet like-minded individuals.
One new item on the agenda this year was “True to the Core”, a concept familiar from Other Salesfore events. The session was an open, unscripted, chat with the top management from Tableau including CEO Ryan Aytay (President and CEO Tableau), Pedro Arellano (Head of Product – Tableau) and Francois Ajenstat (Chief Product Officer). The audience was able to ask questions directly from the leadership team, and questions asked ranged from “When do we get spellcheck” (Apparently important in the US market) to “What is the future for Tableau Server compared to Cloud”. The management team members were very open and humble in their responses and pointed out that they are there to help the customers to succeed and they are open to feedback on how and what to improve and focus on.
Tableau has improved its data governance, management and modeling capabilities during the last couple of years and that trend is continuing with some new keynote announcements. VizQL Data Services is a layer that sits on top of the existing reports and data sources, and this feature allows programmatic access to the underlying data directly, and not only through a visualization. This direct access to data opens up a lot of new possibilities to utilize the work that has already been spent on acquiring, processing, loading and curating the data. The data can now be used in various use cases like integrating existing data into an automated business workflow or building new data applications. It also seems likely the new Metrics capabilities will build on this feature, taking a step towards headless BI and raising excitement in social media and Tableau community blogs. Also another data layer feature, Shared Dimensions, enables more flexibility to data modelling and is eagerly waited. Vizql Data services is targeted to be available for a developer preview during 2024.
VizQL Data Service was announced in Conference Keynote.
Tableau has long been working on capabilities to embed the data into other applications to make data available to all in relevant context. Examples of these features are Embedding API and Connected Apps, and these are available today. But so far it has only been possible to embed and control a viz (visualisation or dashboard). Another new feature to make embedding data easier is the new Embedding Playground. This is already available as developers preview right now, and makes it easier to create new visualisations for embedding by enabling creating the code needed for embedding by drag and drop functionality. Someone who is not familiar with web coding can still create the entire embedded visualisation functionality directly in the browser.
So how is it, is the new day for data dawning? Tableau has been a Tesla of BI analysis tools for a decade, showing the way to go for many competitors. After 2019 Salesforce acquisition Tableau has invested a lot of product development bandwidth to ecosystem related features like integrations, but now some new winds seem to be blowing. Maybe the most futuristic demo in the Tableau keynote session was Tableau Gestures which showed how you could augment your teams call with live and interactive data stats. With the newly announced generative AI features and emphasized community focus Tableau is taking the first steps to show where the future of analytics is heading. Maybe it is more a new day for analytics - after a long winter it is so refreshing to get a dream of spring!
Mathew Miller, Vice President of Product Management, demonstrated Tableau Gestures at the Keynote.
We offer a wide range of Tableau and other data & analytics-related services. For more information please visit our site or contact Juuso Viljanen juuso.viljanen@knowit.fi
Text:
Elina Vihriälä, Tableau Competence Lead
Matz Rasmus, Tableau Ambassador