Image credit: Pexels
From prescriptive decision-making to experience-driven insights and engineering visibility, emerging platforms are moving analytics beyond reporting to real-world action.
Business intelligence has long helped organizations understand past performance, but the growing volume and complexity of enterprise data are exposing the limits of traditional reporting tools. While conventional BI platforms excel at showing what happened, they often stop short of helping decision-makers determine what should happen next. As a result, a new generation of analytics platforms is transforming data into actionable recommendations, measuring real customer experiences, and providing end-to-end visibility across increasingly complex operations.
Rather than relying solely on historical dashboards, these platforms are combining advanced mathematical models, experience-level analytics, and unified data architectures to enable faster, more informed decisions across marketing, customer engagement, and software engineering.
Prescriptive Analytics Takes Marketing Beyond ROI
For marketing organizations, measuring return on investment has long been a standard practice. However, Josh Lucas, VP of Product at Keen Decision Systems, believes that focusing exclusively on ROI can obscure the larger objective of maximizing profit.
“You could take that $10 million budget, and you’re going to end up with like a $3 ROI or a $2.50, but you’re going to make $2 million more in profit. Wouldn’t you rather have more profit than a higher vanity metric?” Lucas said.
Instead of simply reporting historical campaign performance, Keen Decision Systems uses mathematical modeling to determine where additional investment is likely to produce the greatest total profit while identifying the point at which returns begin to diminish. The platform also allows organizations to model future scenarios before market conditions unfold.
“If you really want to make that jump to look in the future, you must account for the past, but you must also have a way to war game the future and then make decisions based on that.”
Artificial intelligence supports the platform by helping users interpret complex outputs, but Lucas emphasized that AI complements rather than replaces the underlying analytical framework.
“We’ve been adding AI for a couple of years to our platform, and we use it — think about the last mile. It might help you configure our application better to then run the real math, the real analytics on it… But it isn’t making the decision. And I think that’s the clear distinction.”
Keen is now developing a closed-loop system that connects executive strategy directly with tactical execution in real time, extending analytics from planning into operational decision-making.
Measuring Customer Experience Instead of System Performance
While many analytics platforms focus on customer identities and historical behavior, Conviva argues that organizations often overlook the quality of the experience customers actually have.
“What all tools miss is the how and the why — meaning I could know you by name, and I could know your purchase history… That doesn’t mean I know you,” said Keith Zubchevich, CEO of Conviva.
The company initially built its expertise in streaming video analytics, where user experience became a competitive differentiator. That philosophy is now being applied to AI-powered conversational agents as businesses increasingly deploy them across customer service and support.
According to Zubchevich, traditional AI performance metrics fail to capture whether customers leave conversations satisfied.
“LLM as a judge, where it judges every individual answer and says the answer was accurate — that misses the whole point. Consumers are going to look for an elegant and seamless and frictionless experience.”
To address that gap, Conviva is developing its “Agent Experience” product, which will monitor conversational efficiency alongside real-time customer sentiment.
“If you’re going to launch an agent, I should know how many turns it took to get the consumer’s intent. I should know in real time what the sentiment of my consumers is right now…”
By shifting the focus from technical accuracy to customer perception, the company aims to help organizations optimize AI interactions based on how users actually experience them.
Bringing Engineering Data into One View
Analytics challenges also extend to software development, where fragmented tools frequently prevent organizations from understanding how engineering work progresses.
Dan Hess, co-founder of Minware, said that disconnected systems have made it difficult to measure software development consistently.
“Software engineering has very much been a black box for a very, very long time. Disconnected systems, understanding what happened when — all of these things are major, major challenges that the industry has faced.”
Minware addresses this issue through its Hypercube data model, which consolidates engineering information from systems such as GitHub and Jira into a unified source of truth. By standardizing metrics across teams, organizations can more easily identify schedule overruns and understand where development time is being spent.
The platform also tackles an increasingly important issue as AI coding tools become more widely adopted.
“One of the things that we’re seeing that’s helpful is linking the token costs to the work that happens in commits or pull requests and then linking that to tickets…” Hess said.
Connecting AI token usage directly to engineering outputs enables organizations to evaluate whether AI investments are producing measurable value and, importantly, whether AI is appropriate for a given task.
Hess also noted that AI is changing the nature of software engineering itself.
“The bottleneck used to be generating code, and that’s where people used to spend the majority of their time — that’s been completely inverted on its head. Now we see that the software engineering role is switching to more of an architecture and strategic type role.”
As AI assumes greater coding responsibilities, engineering leaders are increasingly focused on architectural oversight and strategic planning rather than solely on code generation.
Analytics Evolves From Reporting to Decision Support
Across marketing, customer experience, and software engineering, the common thread is a shift away from descriptive reporting toward analytics that enable better decision-making. As organizations continue to generate larger and more complex datasets, expectations for business intelligence platforms are evolving accordingly.
Rather than simply documenting past performance, the next generation of analytics platforms is designed to recommend actions, reveal how customers actually experienced interactions, and expose the operational costs behind critical business decisions.
