If you picked 100 data and analytics users in any organization, you’d likely spot the following three distinct groups:
1. The guessers
This group includes users who have a general idea of what they need but get stuck navigating their data. They’re constantly second-guessing themselves, whether it’s about which Business View (we call it a BV—a Business View [BV] is a business-ready, logical layer that organizes relevant columns, metrics, and relationships from underlying datasets) to choose, which fields are relevant, or even where to start.
- “I never know which BV to pick.” They spend too much time trying to identify the right dataset for their analysis, often selecting the wrong one and causing delays.
- “Which columns do I even have?” They lack visibility into the available fields and metrics, leading to frustration and inefficiency as they search for answers.
- “Where do I even start with a new BV?” New hires or those exploring unfamiliar datasets often feel overwhelmed and need guidance to avoid analysis paralysis.
This group needs capabilities that provide clarity and guide their decisions to reduce wasted time and errors.
2. The power users
These are the experts tackling the hardest questions—complex, multi-step analyses that demand precision and speed. They’re looking for direct answers without wasting time stitching together datasets or jumping between tools.
“Getting answers for complex questions is burdensome.” Power users often face bottlenecks when combining datasets, calculating benchmarks, or applying advanced filters. The process involves repetitive, manual steps that slow down their ability to deliver impactful insights.
3. The community builders
These users are focused on scaling knowledge across their teams. They understand the importance of standardizing terminology, sharing best practices, and ensuring everyone works from the same playbook.
“I’m tired of re-explaining the same terminology.” They repeatedly define and clarify key metrics or terms, which is time-consuming and might lead to inconsistencies across the organization.
Such community builders need centralized tools for defining and sharing organizational knowledge, ensuring alignment and reducing redundancy across teams.
Tellius 5.3 addresses these three challenges. In this release, Kaiya—our conversational AI for analytics—pushes boundaries to help you:
- Take out the guesswork by eliminating the friction of searching for the right Business Views or columns.
- Crack complex queries without typical back-and-forths, so you can tackle sophisticated, real-world problems head-on.
- Centralize your company’s Learnings as Kaiya helps turn every definition into an asset that the entire organization can reuse and build on.
Below, we’ll demonstrate how these new features work in real-world scenarios. Each example is a representative use case—whether you’re a current Tellius user or you’re just discovering us, these enhancements simplify your analytics journey so you can focus on answers, not the mechanics of finding them.
Written by: Ramya Priya,
Product Content and Strategy Lead at Tellius
In This Post
Kaiya takes out the analytics guesswork
Scenario 1: “I never know which BV to pick.”
Most often, business users rely on analysts to select the correct dataset for their queries. Even when they tried to select datasets independently, they’d often end up choosing irrelevant ones, leading to delays and confusion.
That’s where Kaiya’s SmartSelect comes in. By analyzing the context of each query, Kaiya will automatically select the most relevant BV. For example, when someone asks, “What’s the project cost trend for the last quarter?,” Kaiya instantly identifies “Project_Costs_BV” as the best match and delivers results on the spot. Similarly, for a query like, “Which regions have the most delayed projects?,” Kaiya would auto-select “Regional_Timelines_BV” and provide accurate insights.
If Kaiya ever picked the wrong BV, users could correct it, and Kaiya would learn from their feedback—getting smarter and more accurate with every interaction.
Scenario 2: “Which columns do I even have?”
Identifying which columns in your datasets are relevant to a specific query or analysis often required hours of effort spent digging through technical documentation and repeatedly seeking clarification from the IT team.
Enter Kaiya’s metadata queries. Instead of wasting hours manually hunting for the right fields, you can simply ask Kaiya, “What’s in Pricing_BV?” and get your answer in seconds. Kaiya will instantly display a detailed breakdown of all the relevant columns—dimensions and measures. Kaiya can even answer specific questions like “What date fields are available in this BV?” and “What are the formulae of the calculated columns?”.
No more back-and-forth with IT. No more sifting through endless files and dashboards. Having instant clarity on available columns will help you avoid errors while creating questions and ensure no critical fields are missed.
Scenario 3: “Where do I even start with a new BV?”
New hires would often struggle to navigate large datasets. Without familiarity, they might often feel overwhelmed and need constant guidance from managers or senior analysts. This would lead to delays in onboarding and slow down critical reporting processes.
Kaiya’s starter questions provide an intuitive entry point for analyzing datasets. For example:
- When accessing “Budget_BV,” Kaiya will suggest questions like “Which departments exceeded their budgets last quarter?” or “What are the top expense categories for this fiscal year?”
- For “Forecast_BV,” new hires can see prompts such as “What are the projected revenue trends for the next six months?” or “Which regions are expected to meet their targets?”.
- New hires can ask Kaiya to generate starter questions, which they can review and store, creating a reusable library of queries.
- Admins can also customize the library of starter questions to match organizational priorities, ensuring alignment with business goals.
With Kaiya’s guidance, new hires can quickly gain confidence to identify patterns and insights without constant hand-holding, freeing up senior analysts to focus more on strategic tasks.
Crushing complexity with compound queries
Scenario 4: “Getting answers for complex questions is burdensome.”
To identify the regions underperforming in sales, analysts needed to benchmark regional sales against the national average and then isolate the regions that fell below that benchmark. Sounds simple, right? It wasn’t. The traditional process was time-consuming and error-prone—calculating the national average in one tool, exporting the results, and then using another tool to compare the numbers.
With Kaiya’s compound queries, the process gets streamlined. Instead of jumping between tools and juggling manual steps, analysts can instantly ask, “Identify the sales channel with the highest revenue growth and find the top-performing products in that channel.” From there, Kaiya does all the heavy lifting behind the scenes:
- First, Kaiya calculates revenue growth across all sales channels.
- Next, it identifies the channel with the highest revenue growth.
- Finally, it filters for and delivers the top-performing products within that channel—all in one cohesive result.
For another query like “What’s the trend of new prescriptions for prescribers who have not been called in the last 6 months?”:
- First, Kaiya identifies prescribers who have not been contacted in the last six months.
- Next, it retrieves prescription data for these uncontacted prescribers.
- Finally, it generates a trend chart comparing their new prescriptions over time, helping you spot meaningful patterns—all in a single, streamlined output.
No advanced syntax. No technical expertise required. Just plain-language questions and seamless answers. What used to take analysts days of manual effort can now be done in minutes.
Curated Learnings: Building and scaling knowledge
Scenario 5: “I’m tired of re-explaining the same terminology.”
If a pharmaceutical company frequently uses specialized terminology, such as “high-risk patients” or “priority regions,” in their queries team members had to define these terms repeatedly across the organization. This might turn into a challenge, and moreover, certain terms frequently used by individual users could benefit the entire organization if standardized.
Using Kaiya’s Learnings Management, an organization can standardize these terms across their datasets. For example:
- Defining “high-risk patients” as [Patient IDs 101, 203, 305] within the “Clinical_Trials_BV.”
- Admins can create organization-wide Learnings to define “priority regions” as areas with “fewer numbers of healthcare facilities,” ensuring consistency across all Business Views.
With the ability to add or edit Learnings and control whether they apply to a single user or the entire organization, miscommunication and redundancy can be drastically reduced. Kaiya will automatically apply these Learnings for all relevant queries, leading to better alignment across teams.
This highlights how Kaiya not only resolves specific challenges but also transforms the way teams interact with their data, saving time and driving better outcomes. Take Kaiya for a spin—our AI-driven assistant helpful at every step: from discovering your data’s structure, to handling multi-step questions, to preserving insights for the entire organization.
Curious to see how metadata queries, SmartSelect, compound questions, and curated learnings come together in real time? Join our webinar on January 29th for a hands-on walkthrough. We’ll showcase the latest enhancements, walk you through real scenarios, and answer your questions.
