In this programme, we spoke to István Végh, Deputy Chief Executive of Allianz, and András Érczfalvi, a presenter at TrendFM, about how Allianz Hungária Insurance’s digital transformation took place. The programme was recorded on 10 November 2025 and broadcast on 13 November 2025.
Participants
István Végh – Allianz
András Érczfalvi – TrendFM
András Tresch - Quattrosoft
Lessons from a 15-year partnership for insurance executives
The trap of a comfortable market position
In 2009, Allianz Hungária was the undisputed leader of the Hungarian insurance market. It was a dominant player in the motor and corporate insurance sectors, enjoying stable profitability and a comfortable oligopolistic position.
And that was precisely the problem.
István Végh, Deputy Chief Executive of Allianz Hungária, put it this way during the discussion: “If I had to assess the situation at the time, I would have given the company a mark of seven or eight. But if I had projected the expected changes five years ahead, then certainly a six. Because by then, the advantage we had would have been exhausted.”
The figures spoke for themselves: nearly 5,000 employees, 300–400 different IT systems, and an organisation that was on an unsustainable path, both technologically and in terms of its human resources.
The ‘50 Lotus Notes applications’ syndrome
Does this sound familiar? Dozens of separate systems, each of which solves a specific sub-problem – sort of. In the case of Allianz, this meant more than 50 Lotus Notes-based applications, which István Végh aptly referred to as ‘quick-fixes’.
These systems were capable of handling emails and storing documents in some way. But what they lacked was:
Conscious process control. No one had a comprehensive overview of where a case was in the processing chain.
Skill-based task allocation. There was no way of ensuring that a given task was assigned to the employee with the appropriate skills.
Process-based measurement. It was impossible to say how long each step took or where the bottlenecks were.
Performance-based incentives. How could performance have been measured without any data?
Managers made decisions ‘on a hunch’. And whilst human intuition is valuable, in an organisation processing one and a half million transactions a year, it is simply not enough.
Why do most digitalisation projects fail?
Before we discuss the solution, let’s take an honest look at why most digitalisation initiatives fail.
The ‘giga-project’ trap. Many leaders believe that the company must be transformed into a new way of operating in one big, comprehensive step. One switch, one big bang, and that’s it. This almost never works. The organisation doesn’t have time to digest the change, the risks become unmanageable, and at the first serious hurdle, the whole initiative collapses.
A diversion of management’s attention. We had a project where the client’s top executive found himself a new priority every three months. IT support vanished, all management commitment to the project evaporated – and the administrative side did everything in its power to avoid having to change.
The power of the status quo. Let’s not underestimate it: people don’t like change. In a large organisation, resistance can be enormous. Without management commitment, any digitalisation project becomes a bumpy ride.
The lack of measurement. If matters are handled via email, it is almost impossible to generate measurement data from this. It may be possible to track whether something has been completed, but not how long it took, who worked on it, or where each sub-task got stuck.
Complaints handling as the perfect pilot project
Allianz’s management took a different approach to the issue. They did not launch a mega-project. Instead, they looked for an area which:
- Small enough to make the risk manageable
- Complex enough to include all the important elements
- Suitable for properly testing the system’s capabilities
The choice fell on complaints handling. A team of just a few people, dealing with a few thousand cases a year – compared with the administration’s one and a half million transactions.
But this small area was a real ‘workhorse’: something came in from the customer (input), the organisation processed it (process), and then provided feedback (output). If the system works here, it will work elsewhere too.
And it did work.
After the pilot, the organisation moved forward step by step. There was a major roll-out, then the organisation digested it and accepted it. Then came the next area. This was not a sprint, but a marathon – as István Végh put it: “an extended marathon”.
The ‘Train the Trainer’ approach
Introducing the technology is only half the job; the other, often neglected half is preparing the organisation.
In the case of Allianz, we used the ‘Train the Trainer’ method. We trained key users, who were then able to deliver integrated training to their colleagues. This was particularly important because staff do not just use the workflow system – they have to work across several systems whilst carrying out a process.
Change management is not a one-off event. It is part of everyday life, as there are constantly changes, both big and small. However, this issue becomes particularly prominent in the case of major structural reorganisations.
The results: in figures and in terms of competitive advantage
Fifteen years on, the results speak for themselves:
Staff numbers. From its original total of nearly 5,000, the organisation has been reduced to 800 in-house staff and an external sales network of around 1,200. Operations and claims processing are now carried out by 400 staff.
Administrative costs. Allianz Hungária operates with administrative costs of 6.8 per cent of premium income, whilst the average for major insurers is around 8.7 per cent. This 2-percentage-point difference translates into savings in the billions on an annual basis.
Market position. Allianz is by far the most efficient insurance company operating in Hungary today. This position is not down to chance, but to a deliberate and persistent drive towards digitalisation.
Performance-based pay. Around 2015–17, Allianz was the first – and perhaps remains the only – organisation in Hungary to introduce performance-based pay in its administrative departments. This was only possible because the system provided reliable, indisputable data on performance. Today, the salaries of two colleagues sitting next to each other can vary significantly based on their measured performance – and this is accepted by everyone because the data is transparent.
The significance of architecture
One important decision was made right at the start: the core system and the workflow system operate side by side; they did not attempt to merge one into the other.
This decision saved us a great deal of money and a lot of hassle. The workflow system handles task management and processes, measures KPIs and ensures transparency. The core system continues to handle insurance transactions.
It was a similarly deliberate decision to store the documents initially in the workflow system’s database and then, after five years – when the volume made it necessary – to transfer them to a dedicated document management system. Such gradual, well-planned migrations are key to long-term success.
Next chapter: Process Mining and Artificial Intelligence
Today, Allianz finds itself in a similar situation to that of 2009–10. It operates in a stable and efficient manner – but the technological and market environment around it is changing.
Last year, the company analysed its operations using the Allianz Process Mining tool (Celonis). They processed more than 10 million transactions and identified 170 opportunities for optimisation.
The workflow system can tell you how many tasks there are and what their turnaround times are. However, it cannot tell you where there are unnecessary steps in the process, where there are unjustified waiting times, or where tasks are being ‘passed back and forth’ between colleagues. Other tools are needed for this.
And this is where artificial intelligence comes into the picture. There are tasks that neither robots nor traditional automation can handle – but which can be tackled using AI.
Customer expectations are changing too. The younger generation expects a reply within minutes, not days. If we get back to an enquiry in two days’ time, they may not even remember us – or they may simply have changed their mind.
What can insurance executives learn from this?
Don’t think in terms of mega-projects. Start small, with a well-defined area where the risks are manageable but all the key elements are in place.
Ensure leadership commitment. It is not enough to be enthusiastic at the start. Digitalisation is a marathon, and leaders must be there throughout to steer the train back on track if it goes off the rails.
Measure. Without data, there can be no optimisation. You need to know how long each step in the process takes, and where it is worth taking action.
Think in terms of architecture. Thoughtful system architecture pays off both today and ten years from now. Don’t try to cram everything into a single system.
Get ready for the next wave. AI is not the future – it is the present. Anyone who does not start engaging with it today will find themselves at just as much of a competitive disadvantage as those who failed to start digitising 15 years ago.
Final thoughts
The 15-year partnership between Allianz and Quattrosoft proves that successful digitalisation is not a matter of a single grand gesture, but of perseverance, informed decisions and a partnership in which both parties understand one another.
The question is not whether an insurance company should digitise its operations. The question is how – and with whom.
This article is based on a conversation broadcast on TrendFM’s “Guest of the Day” programme on 12 November 2025, in which István Végh, Deputy Chief Executive of Allianz Hungária, and András Tresch, Managing Director of Quattrosoft, shared their experiences with reporter András Érczfalvi.
