Top Companies to Watch: Fall 2026
Agentic AI has changed what product teams can build. But before chasing every AI-powered feature, winning teams are taking a step back and asking themselves what they can create to make a meaningful difference for their business and their customers.
Those who focus on what creates a real advantage will propel themselves forward using the power of AI.
For SaaS product leaders, analytics is where that question gets tested first. Your customers now expect to ask questions of their data inside your product and get answers back, not export a CSV or dig through static dashboards. An AI demo makes that look easy. Shipping it is only the beginning.
Governance, security, query performance, hallucination management, and scale all become yours to own, and that permanent ownership pulls your best engineers away from the workflows and experiences that make your product different. Meanwhile, the backlog grows, costs climb, and your customers' expectations keep rising.
The eight companies here took a different path. Rather than building and maintaining their own analytics layer, they decided what only they should focus on: their data, their domain logic, and their customer experience. Then they embedded ThoughtSpot for the rest. The result is conversational, agentic analytics inside their own products, in front of customers fast, and a sharper product for the people who use it every day.
Endpoint Clinical: Closing the Embedded Analytics Revenue Gap in Three Months
Endpoint Clinical has run roughly 2,500 clinical trials across 90+ countries, and trial supply can account for 10 to 30 percent or more of a trial's budget. Analytics done well has real financial consequences. For 15 years, though, building it in-house meant a compounding maintenance burden.
Jeff Rubinson, VP of Product, is blunt about where that left the team:
"As a product leader, I always say there are a couple of things you don't want to do internally. One is analytics." - Jeff Rubinson, VP of Product, Endpoint Clinical
Endpoint chose to buy instead. It signed with ThoughtSpot in December 2025 and launched Elo AI in March 2026, with roughly 1.5 full-time resources on the implementation.
Elo AI gives clinical trial managers conversational AI for their supply chain data, a deep analysis mode for complex investigations, and recommendations that tie insights directly to decisions. Rubinson describes the customer reaction simply:
"When we demo this solution, it's almost like magic to some of our customers."
His advice to other product leaders is to focus on core competencies and let experts handle the rest, because analytics "is not a one-time implementation. It's a lifestyle."
onlyfy by XING: Recruiters Who Ask Instead of Wait
onlyfy by XING, a digital recruiting and talent acquisition brand, manages the full recruiting journey, backed by a XING talent pool that onlyfy reports at 21 million profiles across the DACH market. As the product grew, its analytics ambitions started competing with its core focus on recruiting.
Anastasia Koneva, Director of Product at onlyfy by XING, describes the fork in the road plainly:
"As our product grew, development speed slowed and we could no longer move as fast as our customers needed. At some point it became clear: we didn't want to become a BI vendor. We wanted to focus on what we do best, and that's recruiting." - Anastasia Koneva, Director of Product, onlyfy by XING
So onlyfy made a build versus buy decision and split responsibilities cleanly. Snowflake holds the data foundation. On top of it, ThoughtSpot Embedded with Spotter delivers the AI analytics and the agent experience. Onlyfy keeps the domain layer: KPI definitions, semantic context, and recruiting logic.
On top of that split, the team embedded Spotter into Mira, an AI agent that lives natively inside the onlyfy platform. Recruiters now ask questions in plain language, such as how a funnel is performing or where candidates are dropping off, and Mira answers with visuals. No separate login, no new tool, and no SQL. Guided prompts give non-technical users a place to start, so analysis happens inside the recruiting workflow rather than after it.
Within three months of launch, the adoption numbers moved quickly. Onlyfy reports more than 850 customers on the new experience and over 1,000 questions asked through Mira. In that same window, users created and shared more than 35,000 liveboards. Koneva reads that engagement as the real signal:
"When customers start asking questions instead of navigating static dashboards, something fundamental shifts. For us, the clearest sign that Mira is working is the depth of engagement. People are exploring their data directly, without relying on manual reports or Excel workarounds."
There is a commercial payoff too. Onlyfy says its sales conversations have shifted from demoing applicant-tracking software toward strategic discussions about faster and better hiring decisions.
Zencargo: Turning Freight Data Into an Analyst in Your Pocket
Zencargo, a digital freight forwarder, deals with a different version of the same problem. It moves goods by sea, air, road, and rail, yet its real edge is the data behind those moves. All of it comes from hundreds of APIs, stitched into something a customer can actually question.
Sam Greenhalgh, the company's CRO, calls early supply chain insight "the number one lever for how businesses can outperform their competitors."
To close the gap between a question and an answer, the team built Ask Luca, an embedded assistant it calls an “analyst in your pocket.” Zencargo moved from data plus visualization to data plus brain and visualization. That brain is a semantic layer encoding shipping context, performance, risk, and opportunity. ThoughtSpot Spotter handles the natural-language layer on top.
Because the model is proactive, Ask Luca surfaces opportunities, risks, and decisions instead of waiting to be asked. Zencargo describes the aim as turning “what ifs” into “what’s next.” The analytics layer became a retention engine, and Customer Success moved from tactical KPI readouts toward root-cause strategy.
Adoption backs that up. Ask Luca launched in October, and Zencargo says more than 100 customers adopted it quickly, with a waiting list for further models. Internally, roughly 150 people use ThoughtSpot every day as part of normal work. Greenhalgh puts the stickiness bluntly:
"We're the victim of our own success; customers are addicted to Luca." - Sam Greenhalgh, Chief Revenue Officer, Zencargo
His build-versus-partner strategy is the lesson data leaders will recognize. The real moat was the data, not the visualization. So Zencargo builds and owns its live data platform, while ThoughtSpot owns the analytics layer above it. Looking ahead, the roadmap for Ask Luca 2.0 moves past answering, weighing past decisions, supply chain signals, and outside events to recommend a next action.
ASK BOSCO®: Scaling Embedded Analytics From Weeks to Minutes
ASK BOSCO®, a marketing and e-commerce analytics platform, had a scaling problem rather than a starting one. It needed embedded analytics that could serve hundreds of clients without a BI team behind every account. According to founder John Readman, the gap was never a shortage of dashboards:
"Just having access to more dashboards isn't the answer. The answer is getting insights to the answers to the questions you want in an accessible format." - John Readman, Founder, ASK BOSCO®
Before landing on ThoughtSpot, the team looked around. Readman reports that Looker Studio and PanIntelligence lacked the scalability and robustness they wanted, and that Tableau's commercial model did not fit their growth plans. ThoughtSpot's technology, flexible licensing, and roadmap let them onboard customers quickly.
The result is ASK BOSCO® AI Analyst, powered by Spotter. Non-technical users like CEOs and CMOs get answers during a meeting through natural language search. ASK BOSCO® cut the time to answer important business questions from weeks to minutes and scaled embedded analytics to hundreds of clients.
Gallus Insights: Two Hundred Dashboards, One Question
Gallus Insights builds analytics for enterprise mortgage lenders and servicers, an industry of huge loan portfolios and razor-thin margins. Its clients had invested millions in data infrastructure, yet the people running the business day-to-day still couldn't get answers. Augie Del Rio, Gallus's founder and CEO, describes the real problem:
"Having 200 dashboards doesn't mean you have too much information; it means you have too much friction between the question and the answer." - Augie Del Rio, Founder + CEO, Gallus Insights
Gallus built on Snowflake for performance, security, and scale, and embedded ThoughtSpot directly in the Gallus application. Loan officers and operations leaders now type questions in plain English, such as how this quarter's delinquency rate compares to last year's.
As Del Rio puts it, "If you can Google, you can Gallus." Gallus reports 10x faster performance and reduced data costs, and queries that took minutes now return in seconds.
ThoughtSpot’s semantic layer made that trust possible. Gallus used it to turn complex mortgage business logic into governed answers about aging loans, bottlenecks, and P&L impact, and Del Rio warns that most companies underestimate the modeling work.
The payoff is behavioral. Clients went from waiting two weeks on a data request to managers asking dozens of questions a day on their own.
ELMO Software Group: Ending Spreadsheet Hell for HR Teams
When CTO Josh McKenzie evaluated ELMO's internal analytics, he saw a wall. The homegrown tool demanded too much specialized expertise, leaving HR teams in "spreadsheet hell" and waiting on analysts for answers that should have been instant.
His fix was not to teach non-technical users data modeling. ELMO partnered with ThoughtSpot to bridge that gap with natural language, so users could query data inside their own workflows. McKenzie points to Spotter as the deciding factor:
"The big thing for us was the Spotter product. Allowing users to bridge that data analyst gap was really important. So, that product has yielded really, really great results for us." - Josh McKenzie, Chief Technology Officer, ELMO Software Group
By embedding Spotter, ThoughtSpot's AI analyst, ELMO removed the friction of traditional reporting. HR teams can now cross-reference payroll, benchmarking, and performance data on demand, without an analyst translating for them.
McKenzie also shares his thinking on build versus buy, on mapping every role to rebuild ELMO's engineering lifecycle around AI agents, and on why domain expertise remains the moat for SaaS companies, in an episode of The Data and AI Chief podcast.
The Agentic Analytics Pattern for Product Leaders
Across eight very different businesses, the same advice keeps showing up:
Decide what only you should build. Partner with a leading analytics provider for the analytics layer, and spend your engineering time on the domain context only your team can build.
Inspire curiosity to drive adoption. Move your customers from reading dashboards to asking questions in their own words.
Keep governance in the semantic layer. Self-service can then scale without giving up trust or control.
Measure engagement, not output. Track questions asked and repeat use rather than the count of dashboards shipped.
Measure business impact, not just usage. Trace analytics to a decision, a behavior change, or new revenue.
Sequence the roadmap. Go from governed answers today toward human-approved, agentic actions next.
How to Build What Differentiates You
Across Endpoint Clinical, onlyfy, Zencargo, ASK BOSCO, Gallus Insights, and ELMO, the through-line is one decision: what to build, and what to leave to someone else. When product teams stop owning the analytics layer, they get their roadmap back and put a better experience in front of their customers sooner.
If your team is weighing what to build versus embed, the next step is to see it in practice. Join a weekly live group demo to experience ThoughtSpot Embedded in action.
Common Questions From Product Leaders
What does moving from dashboards to conversational analytics mean for my product?
Your customers stop hunting through prebuilt views and start asking questions in plain language, inside your product. A dashboard shows what someone decided to chart last quarter, while conversational and agentic analytics let your customers ask the next question themselves. They can keep going, one question after another, without filing a support ticket or waiting on your team to build another report.
Should you build the analytics layer or partner?
The companies here landed on partnering, and their reasoning is consistent. Your differentiation lives in your data and domain context, not in charting libraries or a query engine. Endpoint Clinical spent 15 years attempting to build analytics in-house before partnering with ThoughtSpot. ELMO's CTO weighed the same dilemma and chose to partner. Onlyfy and Zencargo kept ownership of their data and semantics, then partnered for the analytics and agent layer so engineers could stay focused on the product.
What should product teams still build themselves?
The layer that makes your product irreplaceable: industry expertise, domain logic, and proprietary workflows. Onlyfy kept its recruiting logic, Zencargo its shipping context, and Gallus its mortgage business rules. The AI-native experience your customers interact with, and the governance, security, and performance foundation beneath it, are better embedded than owned.
How do you drive self-service adoption without losing governance?
Put the governance in the semantic layer rather than in access restrictions. When KPI definitions, metrics, and business context live in a governed model, every answer traces back to the same trusted logic. Guided prompts help too, giving non-technical users a safe starting point instead of a blank search box.
How do agentic analytics tools keep answers trustworthy?
They ground responses in your live, governed data model rather than guessing from documentation. The payoff is an answer that is correct, not just plausible.
