We have the data. Why can’t we decide?
We asked Bogdan Jelić, Business Development Manager at Comtrade System Integration, why companies can have more data, dashboards and technology than ever, and still struggle to make decisions quickly. In his answers, he explores the role of Excel, ERP, BI and data silos, and what it really takes to turn data into business decisions.
Companies have more data and technology than ever before. So why can it still take days to answer a relatively simple business question?
Bogdan:
Because having data is not the same as having the information you need to make a decision.
In most companies, data is there, but it is spread across ERP systems, CRM platforms, HR systems, different databases, BI reports, and a large number of Excel files. Each system shows one part of the business, while the answer management needs usually requires connecting several of those parts.
Take a simple question: what would a 10% increase in sales actually mean for the company’s results?
The answer isn’t simply 10% more revenue. You need to understand whether the company has the operational capacity to support that growth, how many additional people would need to be hired, whether it would require more inventory or working capital, and what impact it would have on logistics, collections, cash flow, and ultimately profitability.
Sales sees the opportunity, HR sees the need for more people, operations sees capacity, and finance sees the financial impact. The problem starts when no one can quickly bring those four perspectives together.
That’s why the real value of digital transformation isn’t about producing more data or building better-looking reports. It’s about being able to quickly translate a business change into its operational and financial impact.
Excel is often seen as one of the main problems in planning and reporting. Is it really an obstacle to digital transformation?
Bogdan:
Excel isn’t the problem. The problem is what companies try to do with it.
It’s an extremely flexible tool, and it’s almost impossible to imagine serious work in finance and analytics without it. But there’s a big difference between using Excel for analysis and having Excel act as the entire infrastructure for running the business.
The problem starts when one file becomes the database, planning model, approval system, consolidation tool, reporting source, and the only record of who changed a particular assumption.
At that point, the business starts depending on files that are difficult to control and on people who are the only ones who understand how those files work.
A setup like this can work perfectly well for years. But as the company grows, so does the number of users, markets, products, and business units. And with them comes more versions, more manual work, and more room for error.
That’s why I don’t believe in an “Excel is bad” approach. Excel should stay where it adds the most value: quick analysis, modelling, and exploring data. But we shouldn’t expect it to handle process management, system integration, access control, versioning, and corporate data governance on its own.
Excel is usually a symptom. When key processes are being managed through dozens of disconnected spreadsheets, it often means that the formal systems and processes don’t really match the way the company operates.
Many companies already have an ERP and well-developed BI reports. Why isn’t that enough for effective business management?
Bogdan:
Because ERP, BI, and business planning solve different problems.
An ERP is primarily a system of record. Its job is to reliably process and store business transactions: invoices, purchases, payments, accounting entries, inventory, contracts, and other events that have already happened or are happening now.
BI is great for organising, analysing, and visualising data. It can clearly show what happened, where something went off track, and how certain metrics are changing.
But managing a company also means looking at a third dimension: what do we expect to happen, and what are we going to do if things change?
A dashboard can show that sales have dropped. But it can’t decide whether the company should change its pricing, slow down hiring, move the budget around, cut certain costs, or invest in a new sales channel.
To make those decisions, you need business drivers, assumptions, scenarios, clear ownership, and an understanding of how different parts of the business affect each other.
A company can have a top-class ERP, a modern data warehouse, and visually impressive dashboards, and still not have an aligned view of where the business is heading.
Technology can process the data, but it can’t decide what that data actually means for the organisation, who owns it, or how a change in one area affects the rest of the business.
In other words, the ERP can be the source of truth for transactions, while BI can provide a good view of what’s happening. But to make decisions, you need a shared model of how the business works.
Are data silos mainly a technology problem, or is there usually a deeper organisational issue behind them?
Bogdan:
Data silos rarely exist simply because two systems aren’t technically connected. More often, they exist because different parts of the company use different definitions, assumptions, and goals.
Sales might look at revenue based on signed contracts, finance might look at invoiced revenue, while management might look at cash collected. All three perspectives can be valid. But if it isn’t clear when each one should be used, the company ends up with three versions of the same metric.
You can see a similar problem between HR and finance. HR might talk about the number of employees, while finance looks at the cost of labour.
But the number of employees alone doesn’t tell you what the financial impact of hiring will be. You also need to consider the start date, salaries, bonuses, taxes, benefits, turnover, open positions, and how long it takes for a new employee to become fully productive.
Take a bank, for example. One part of the organisation might be planning growth in its loan portfolio, while the impact of that decision shows up in capital requirements, liquidity, expected credit losses, operational capacity, and regulatory metrics.
The same applies to insurance. Growth in premiums can’t be looked at separately from claims, reserves, reinsurance, distribution costs, and product profitability.
In situations like these, simply connecting databases isn’t enough. If different parts of the business understand the same business event differently, integration will only move that inconsistency from one system to another faster.
That’s why data management is both a business and a technology discipline. IT can provide the architecture, integrations, security, and access to data. But the business has to define what the data means, what the rules are, who owns it, and how it should be used to make decisions.
Having one database doesn’t guarantee one version of the truth. For that, you also need one shared understanding of the business logic behind the data.
Why do companies sometimes invest heavily in digital transformation, only to end up with almost the same problems after implementation?
Bogdan:
Because transformation is often treated as buying and implementing software.
A company takes an existing process, including all its manual steps, exceptions, and workarounds, and asks for it to be moved into a new system.
Technically, the implementation can be a success. The system works, the data has been migrated, and users have been trained. But the organisation has simply ended up with a more modern and more expensive version of the old way of working.
The most expensive mistake in digital transformation is successfully automating a process that should never have been kept in the first place.
Before choosing a technology, you need to understand which decisions the company wants to make faster and better. Only then should you define what data is needed, where it comes from, who is responsible for it, and how a change in one part of the business affects the rest of the organisation.
That’s where close cooperation between finance, business teams, and IT becomes critical.
The business understands the commercial and operational reality. Finance can translate that reality into measurable financial impact. IT makes it possible to apply that model reliably through data, systems, and automated processes.
None of these three perspectives is enough on its own.
When IT leads a project without enough business context, the company can end up with a technically strong solution that users simply work around.
When the business leads a project without enough architectural discipline, it can create yet another isolated solution.
And when finance gets involved only at the end, the system may work well operationally but have no clear connection to profitability, cash flow, or strategic goals.
Successful transformation therefore doesn’t start with the question, “Which tool should we buy?”
It starts with a different question: “Which decision can’t we make quickly enough today, and what is stopping us?”
