Chat Tools to Track Decision-making

This is the continuation of the transcript of a Webinar hosted by InetSoft on the topic of "The Movement Towards Using Unstructured Data in Business Intelligence Solutions." The speaker is Mark Flaherty, CMO at InetSoft.

Collaboration is Still Primitive

There’s plenty of other places to start, you don’t have to start with collaboration but it is one of the things that information applications should provide. People are more likely or more comfortable doing this within an organization right now as opposed to either with consumers of their business partners. There is not a huge difference between those, but still directionally, people are more comfortable to deliver these types of things within an organization. We see repeatedly in all the different areas, that usability is most important.

Structured data, the point I would take away from this is that there are amounts of unstructured data that are also approaching not quite 50 percent, but if you look at the content management systems, that’s 45 percent and almost a third of the organization’s look at unstructured data from documents.

So why wouldn’t transaction data be a hundred percent?

I’m fairly certain that the way this question was asked and I don’t have the data right in front of me, but the way we typically ask this question is ‘very important,’ ‘important,’ ‘somewhat important’ and ‘not important.’ And I’m pretty sure this bar represents those who ranked it very important. So I suspect that many others also ranked it important but just not very important. And if you think about what I said earlier that some people in this space are focusing mostly on unstructured data, so there could be people responding to the survey that are more interested in unstructured data than structured data for particular applications.

So from their perspective, that information is more important. Those customers were dealing probably with unstructured information. And so to the extent that they responded to the survey, quite possible that they could have ranked unstructured information highly important. And it’s not exclusive, they can rank as many things as they want as ‘very important.’ It’s quite possible that some of our larger customers may have ranked unstructured data as very important and structured information as important or not important. Similar types of things you’ve seen in the business intelligence space come up here. This first one again, assembling information into a view is important.

Focusing on Collaboration

These are recommendations we make to organizations. They’re primarily targeted at the users of these applications. So we’re suggesting that organizations look at how they’re doing in assembling these types of applications across structured and unstructured boundaries. We encourage both business and IT to get involved. As with BI applications, if it’s driven solely by one or the other, there’s going to be gaps in the ultimate solution. Focusing on productivity and inefficiencies, this will back that notion of applying technology, whether it’s automating the entire process or hoping to at least automate collection and assembly of information or finding information that’s going to make the decision-making more efficient.

Using Chat Tools for Ad Hoc Reporting

Chat tools, when integrated with business intelligence (BI) platforms, enable ad hoc reporting by allowing users to query data and generate reports using natural language. These conversational interfaces leverage natural language processing (NLP) to provide quick, on-demand insights without requiring technical expertise. While powerful, their effectiveness depends on data integration, platform capabilities, and the complexity of reporting needs.

• How Chat Tools Enable Ad Hoc Reporting

  Modern BI platforms like ThoughtSpot, Power BI, and InetSoft’s Style Intelligence integrate chat interfaces that interpret natural language queries (e.g., “Show sales by region for Q1”). These tools connect to data sources, process queries, and return results as visualizations or reports in real time. This conversational approach simplifies ad hoc reporting for non-technical users, such as marketers or executives.

• Benefits of Chat Tools for Reporting

  Chat tools offer speed, enabling instant report generation without manual dashboard creation. They enhance accessibility, allowing non-experts to explore data via intuitive queries. Additionally, they integrate with platforms like Slack or Microsoft Teams, delivering insights directly within workflows.

• Limitations and Challenges

  Chat tools may struggle with complex or ambiguous queries, requiring clear phrasing for accurate results. They rely heavily on well-integrated, high-quality data sources to avoid incomplete or inaccurate reports. Advanced reporting needs, like custom calculations, may still require traditional BI tools or manual intervention.

• Examples of Chat Tools in Action

  ThoughtSpot’s SpotIQ uses NLP to answer queries like “What are our top-performing products?” with instant charts. Power BI’s Q&A feature generates visualizations from questions typed in Teams or the BI interface. InetSoft’s platform supports conversational analytics, embedding reporting into business applications for seamless ad hoc insights.

• Best Practices for Effective Use

  Ensure data sources are clean and integrated to maximize query accuracy. Train users to frame clear, specific questions to avoid misinterpretation by NLP algorithms. Combine chat tools with traditional BI for complex reporting, using platforms like Tableau or InetSoft for deeper customization.

Chat tools streamline ad hoc reporting by making data exploration conversational and accessible, aligning with the consumerization of BI. They are particularly effective for quick insights and simple queries but may need to be paired with robust BI platforms for complex needs. Tools like InetSoft, ThoughtSpot, and Power BI exemplify how chat interfaces are transforming enterprise reporting.

Previous: What Are Some Examples of Unstructured Data BI Applications?

Next: Tools to Simplify the Building of BI Applications