Enriching Structured Data and AI-Based Analyses

In our previous two blog posts, we provided examples of what structured data is and how it can be used to automate various tasks.

This time, let’s take things up a notch.

What can you do once you’ve collected enough data for automation?

The obvious next step is to collect more data—enrich the existing data. The internet is full of information (not in structured form), and it would be beneficial to use this in various analyses if it could be produced automatically and easily utilized. By doing this, we can easily enrich the information we need and generate even more accurate analyses, often revealing entirely new insights. With the right technology, this is now possible. But what does this mean in practice?

Let’s go through this with two practical examples.
Case 1: Enriching and Analyzing Business Target Groups

Imagine Company A, which wants to compare one hundred Finnish companies based on their financial statements. Using structured and automatically generated data, the company has compared and organized its target group. The next step, usually performed manually, is to enrich the selected companies’ data with information about their current operations and contact details. This manual step is replaced by automated and AI-assisted data collection and enrichment algorithms. Once this enrichment data, previously in document form, is now structured data, it can be fed to an AI-assisted analyzer for conclusions and further actions. This final step is known as prompt engineering, which we will explain further in upcoming blog posts.

Case 2: Valuation and Analysis of Business Target Groups

Imagine Company B, which wants to obtain valuations for, say, 50 companies and compare them. The current method involves manually extracting the desired information, entering it into an Excel file, creating formulas, and hopefully getting the desired answers. It’s easy to see that these steps take a long time and are highly error-prone. With new technology, this data can also be extracted from various document-based sources, converted into data, and fed to AI assistants for analyses.


In summary, enriched data and data extracted from document-based sources can automate many workflows previously performed manually and also produce easily AI-based analyses. When we combine this new data with prompt engineering—AI assistants trained for specific use cases—we can achieve significant benefits.