In an era where data drives decision-making, many organizations still struggle to leverage modern visualization techniques. Daria Voronova (Aria Voron), a data visualization expert, discusses why cultural resistance often outweighs technical barriers and how AI-enhanced dashboards are transforming raw data into actionable insights. She also explores the role of NLP and sentiment analysis in measuring human behavior and the future of explainable and multi-modal AI. Read on to uncover how businesses can move beyond outdated methods and embrace a data-driven future.
Explore more articles here: What is The Importance of Data Visualization and Reporting?
You mention that many organizations still rely on outdated data analysis methods from the 1980s. What do you think is the biggest psychological or structural barrier preventing businesses from adopting modern visualization techniques?
I’m not saying that sticking to raw numbers over data visualization is wrong. Tables and raw numbers are critical for decision-makers. Executives and managers depend on them. Always. Period. Choosing not to prioritize data visualization doesn’t mean a company is outdated or failing. Some startups or consultancies might not have the data volume or complexity to justify it and even big players run smoothly without leaning on it. I’m talking about something very specific. Organizations that could benefit from modern visualization techniques, but refuse to change, evolve, and integrate new technologies, and best practices because "that’s just how we’ve always done it". Also. Everything written below is not a universal rule, and generalization. Rater is a subjective outlook based on my personal experience.
The biggest barrier isn’t psychological. It’s cultural. Complacency and fear of change. "This is how it’s always been, this is how it always worked" becomes an excuse to avoid real problems, disjointed data, outdated thinking, and zero frameworks for qualitative insights. People cling to last-century methods not because they necessarily work for them. But because they’re familiar with it. Choosing comfort over growth drags people down. Often businesses go along with them.
Also, I've seen a lot of cases when companies hire a Jack-of-all-trades specialist expecting one person to handle coding, data pipelines, analytics, BI, and visualization. Management thinks this saves head counts, but ultimately they lose big. Instance, overworked analysts drowning in ad hoc requests rushed to deliver the last report on Friday at 10 PM would feel exhausted and unmotivated over a short period, less likely to produce high-quality results.
I bet you've seen this one. When stakeholders ask for "one more table". Not because it’s all they need. But because nobody has shown them a better way to visualize or represent data. Why? Too many employees won’t step outside their comfort zones to learn new skills. Creative thinking needs curiosity. Counting minutes to 5 PM kills it. The result? Missed opportunities, guess-based strategies, and wasted money on tools, including data, and visualization platforms without leveraging them fully.
Disjointed data sources create chaos. Which only promotes data dumping culture, when raw numbers are dumped with no context and no explanation. The bottom line is: that data visualization isn’t a luxury, not just "nice to have". It's a necessity.
What factors should companies consider when deciding whether to implement data visualization, and how do these factors differ based on the company’s size and stage of growth?
For small startups - Those with 10 clients and simple metrics - raw numbers often do the job. At this stage, spreadsheets can reveal enough trends to make decisions. The priority is growth, not complex data systems. Data visualization wouldn’t add much value here.
Larger companies generating billions of data records face a different challenge, on 0.0005% of their data, just a table is like peeping through a keyhole. Spreadsheets won’t explain why sales dropped in one region but spiked in another. Data Visualization turns noise into clarity. Heatmaps, dashboards, and trend charts transform overwhelming data into actionable insights.
But... Timing matters often the most. Throwing tools at a company that isn’t ready like installing high-tech security in a house with no locks. How do you know when they’re ready?
Cultural mindset. Are leaders proactive, seeing data as the biggest asset or reactive, treating it as a burden? A company in survival mode won't leverage visualization. They’ll just drown in numbers.
Operational foundation. Is the data clean, and unified visualization tools only as good as the data feeding them? If it's riddled with errors, visualize chaos, not solutions.
Strategic priorities. Does the company need deeper insights to stay competitive? For a firm losing market share, uncovering hidden trends is urgent, for one coasting comfortably, it might wait
A lot of companies bring external consultants who can deliver a quick win, but if they don’t transfer knowledge, the benefits disappear once they leave. If they managed to, let's say, increase GP by X percent - that means nothing if your team can’t adapt to and sustain those changes. The Solution is to train your people along the way and raise your internal consultants.
The solution is to build a new infrastructure that remains the same regardless of context, or industry. Know your stakeholders, how they run the business, the metrics they track, and the insights they need to build a validated data source, and then visualize if it’s done right. This makes the business transparent—an MRI scan showing its real health.
You emphasize the importance of AI-enhanced dashboards that provide context rather than just raw numbers. What exactly does it imply and what's the technology behind?
Traditional dashboards tell you what happened, sales increased by 15%. Modern AI tools can show trends or breakdown metrics, but the next level is AI which explains why it happened and suggests what to do next in plain English, considering the business context of your own team/function/ company in general. That’s what context means here. Instead of just numbers, an AI-enhanced dashboard analyzes factors like ad performance, competitor moves, or market shifts, then delivers actionable insights: “Sales rose due to a viral ad and a competitor’s stock shortage. If ad spending stays steady, growth might hold for two months.” This transforms dashboards from static reports into real-time decision-making systems.
The technology behind this is a universal five-step framework any business can adapt:
Data Integration: Combine structured data (e.g., sales figures) with unstructured data (e.g., social media buzz).
Feature engineering creates meaningful metrics like an ad impacts the school
Model selection, choose machine learning ML tools like XG Boost for predictions based on the goal
Explainability used tools like SHAP to reveal what’s driving the results
Narrative generation, natural language processing NLP to turn insights into clear advice
Training a model on your company‘s changes the game. If you feed it the full picture, how your business operates, the key metrics, and the decision driver, it gains real context. Once it understands your business. It won’t just spot in correlation. It will explain why they happen and what they mean.
Your startup integrates NLP and sentiment analysis with data visualization. What was the pivotal moment that made you realize data could be leveraged for more than just financial or operational efficiencies?
AI has expanded structured analysis beyond the data I usually worked with. Finance and operations background showed how technical skills apply across industries, even challenges once impossible to measure
One standout is natural language processing NLP and sentiment analysis. They turn human speech into measurable patterns. Quantifying something that is hard to measure. This is what sparked my interest in analyzing speech, over time, and visualizing trends that matter.
I think when it comes to using human speech as an indicator of health problems.
Many assessments today depend on subjective observations, missing the full view. What I mean by that is it's easy to make a blood test, or an x-ray and see specific quantitative results that show the before and after, if there is any progress or not. When it comes to using speech as an indicator of progress in recovery - it's not that obvious. AI can process speech data continuously, and spot patterns in word choice, sentence structure, and tone. When you integrate data visualization, what happens is that you can visually represent metrics that reflect different changes in human speech.
Hypothetically, a dashboard can then display these trends clearly, for instance, in cognitive rehabilitation speech analysis, tracking shifts in vocabulary, complexity, or coherence changes in rhythm or repetition could signal insight over time. Similarly, those regaining speech could use AI to monitor progress and tweak approaches as needed.
Our startups at this point operated in stealth mode. So no specific details yet. But the concept and idea are similar. Using AI to process speech and detect patterns to help solve a specific problem. And using data visualization to communicate sharper insights.
Disclaimer: I’m not a healthcare professional. These are my own hypothetical ideas, not clinical findings or proven outcomes. Any real application requires expert validation and strict legal compliance.
So how exactly does data visualization come into play in NLP & sentiment analysis?
Another application could be for individuals' public speaking skills. An AI model could analyze speech, quantifying filler words, consistency, logic, and confidence to measure progress, a bar chart showing the filler percentage is time, showing if practice reduces reliance on filler words, a heat map could display vocabulary shifts by topic tech versus personal vs professional, which shows diversity or repetition. A line chart might plot confidence through pacing volume and hesitation markers like fewer or tone to gauge delivery gains. Flow charts could map the logic, coherence in debates, pinpointing arguments that stray or weaken.
In education, AI dashboards could process speech for language, learners, pronunciation, fluency, vocabulary range, accent shift, offering tailored pointers, words, clouds, and sentiment. Analysis might highlight frequent student questions or confusion spots, letting teachers adjust lessons on the fly. Speech progression charts could refine debate skills, tracking fluency, and clarity over weeks. Drop-off chance might flag when online lectures lose students as a dense 10-minute mark guiding educators to tweak pacing or content.
Disclaimer: I’m not an expert in education or tech deployment. These are hypothetical scenarios, not proven solutions, any real use requires explicit consent, expert validation, and strict compliance with privacy laws like GDP or further no exceptions.
Looking ahead, what emerging trends in AI and data visualization excite you the most, and how do you see them reshaping industries beyond just business intelligence?
A major shift is ‘explainable AI (XAI).’ AI does not only make assumptions anymore, it provides a reasoning for its actions. Businesses require trust in AI reasoning, not just numbers. XAI is already providing loaning transparency to finance and assists doctors in interpreting AI diagnosis in healthcare.
The integration of text, audio, and video for richer perspectives marks a notable evolution in AI referred to as Multi-Modal AI. For instance, consider an AI assistant that actively listens to a student’s speech during their presentation and matches it against their written assignments. The goal is to issue a personalized assessment of their fluency and logical coherence. Or consider an AI model that interprets MRI scans alongside a patient's clinical history to issue a more precise diagnosis.
The education industry benefits beyond any words in the adaptation of AI-driven automation. Multi-modal AI analyzes video recordings of lessons and provides instant feedback on student responses.
Legal firms have also adopted AI to review contracts and inattentively pinpoint any prevailing risks.
Conclusion
People fear AI will replace them, but I see differently, AI unlocks human creativity. Look back at the first industrial revolution. Many resisted, saying the shift from manual labour to manufacturing was a threat. But those who adapted didn’t just survive - they drove progress. Without it you wouldn’t be reading this, and I wouldn’t be writing it right now.
AI is the same. It’s not replacing our jobs. It’s pushing us, humans, to the next level. AI takes over tedious, repetitive tasks, giving us more time to be creators rather than executors. It democratizes technology and lowers barriers for young entrepreneurs, making it easier to start a business and bring their ideas to life.
By automating routine work, AI frees us to focus on what truly matters—innovation, creativity, and self-expression. The risk isn’t AI itself; it’s failing to adapt to its impact.
