← Back to Executive Insights
Executive Insight · Systems, standardisation & operational control

Standardisation ends where it makes operations worse.

Why a shared platform can create transparency and scalability – while still reducing operational speed.

Category Experience Context Cross-functional system rollout
Role Overall operational responsibility · Implementation & integration
Environment B2B services · Multi-site
Scope Quotation · Order · Resource control · Inventory · Documentation
Systems Core platform · Back-office system · Operational interfaces
Starting point

Introducing a shared platform was necessary. Too many operational processes were still managed through Excel, individual files and local ways of working. At the same time, the project made one point very clear: a standardised system does not automatically mean that every business area works better afterwards.

Core thesis

Standardisation is valuable as long as it reduces complexity. When employees have to perform additional work simply to satisfy the logic of a system, system architecture becomes a management question again.

01 The starting point

Operationally, there was hardly any common system.

When I assumed overall operational responsibility, an established core system already existed. In day-to-day operations, however, it was only used for a limited part of the processes.

Quotation information, operational lists, transport documents, labels and parts of inventory management were organised through Excel and other local files.

As a result, there was no consistent history. Information was stored in different places and many processes depended heavily on individual employees and their personal ways of working.

The decision to introduce a shared platform was therefore fundamentally understandable – and, in my view, necessary.

02 What improved

Standardisation delivered real benefits.

The platform created a significantly more consistent operational structure for the first time.

Sales Standardised calculation Pricing components, quotations and sales information could be structured more consistently.
Control Greater transparency Quotation volumes, status and other operational information became centrally visible.
Operations Central resource planning Vehicles and operational resources across several locations could be planned jointly.
Documentation One shared history Documents, labels, transport information and process histories became centrally available.

The platform itself was therefore not the problem. It addressed genuine weaknesses in the previous structure. The tension only emerged where process logic developed for other areas met a significantly different operational business model.

03 Implementation & change

The rollout was prepared intensively.

On the company side, I held overall operational responsibility for implementation and integration. At the same time, I pushed for our area to be included in the broader rollout as early as possible.

Together with managers and responsible functional teams, the individual process stages were reviewed early. Managers received in-depth training and specialist functions were involved through workshops.

Key users tested functions, existing operational data had to be prepared and partly connected with existing core systems. Change requests were collected and – where permitted by the central system architecture – incorporated into further development.

After individual project stages were completed, the results were assessed and the next process steps coordinated. For the actual rollout, employees in the operational areas sometimes received several days of training.

My responsibility

Overall operational responsibility on the company side · Coordination with project and system owners · Implementation steering · Involvement of managers and key users · Process feedback · Change requirements · Data migration and system integration · Rollout steering and support for employee training

04 System logic meets operations

At some points, process security began to cost operational speed.

In many situations, our business was highly time-sensitive. Customers sometimes required services within a few hours. The operational organisation therefore had to be able to react very quickly.

The new system logic, however, required defined upstream steps for many activities. Before a transport could be planned, for example, customer and master data had to be complete, quotations created and internal order structures established.

For new customers, additional creation in a connected back-office system could also be necessary. What was operationally a simple transaction could therefore become a longer digital process chain before the actual service had even started.

The same applied to quotations. Existing or very similar transactions could not always simply be reused. Information had to be rebuilt even though the operational situation was already known.

A standardised process is not automatically a good process. The decisive question is whether it creates security without destroying the speed the business requires.

05 Physical reality

A data model and a physical process are not always the same thing.

This became particularly visible with physical transport units.

One transport unit could contain several objects that had to be documented individually. For employees working operationally, however, the transport unit itself was initially the relevant physical object.

The system sometimes required each individual object contained within it to be processed and recorded. Because direct labelling of the objects was not always possible, several different labels had to be attached to the outer transport unit and processed individually during goods handling.

From a data perspective, the logic was understandable. Operationally, however, it created additional work.

A system can represent data perfectly correctly and still make the real process more complicated.

06 Architecture question

Does every function really have to sit in the same system?

As experience with the rollout increased, a different question emerged for me: do all operational functions necessarily have to be represented within the same platform?

My proposal was to separate certain highly operational functions from the large platform and use a leaner, specialised application for them. Through an integration or middleware layer, the relevant data could still have been exchanged with the core systems.

This alternative was not implemented. The selected system strategy deliberately aimed to keep as many processes as possible within one shared platform.

Precisely for that reason, the experience became important to my current view of system architecture. A standardised system can bring advantages. But a single system name does not automatically mean that every function has to take place within the same technical tool.

07 Operational vs. analytical

An operational system must work operationally first.

At the same time, the platform contained a large number of dashboards, charts, evaluations and analytical functions. This transparency had value. In day-to-day operations, however, it also became clear that a very extensive system landscape can affect performance and user experience.

For me, this leads to a clear distinction: operational systems must support transactions quickly, reliably and as simply as possible. Management and analytical information is equally important – but does not necessarily need to be created within the same technical layer.

Particularly with large volumes of data, a separate analytics and reporting layer can make more sense than continuously expanding the operational system.

Conclusion

Standardisation created transparency – and revealed its limits at the same time.

The experience was not an argument against the platform. It was an argument for differentiated system decisions.

Multiple locations Operational resources could for the first time be planned much more centrally across locations.
One history Transactions, documents and operational information became available much more consistently.
Greater transparency Quotations, operational data and management information became centrally visible.
A clear boundary Where process standardisation reduced operational speed, the architecture itself became the issue.
Operational consequence

Standardise where things are genuinely the same.

For me, standardisation is not an end in itself. It makes sense when it reduces complexity, increases transparency and makes repeatable processes easier.

Different business models, however, create different operational realities. If those differences are ignored technically, the organisation eventually begins adapting its work to the system – even where the customer or the operational process actually requires something else.

Every major system decision should therefore distinguish: Which processes are genuinely standardisable? Where does a business-critical exception exist? And which functions really belong in the core system?

The same applies to data and analytics. Management needs KPIs and transparency. Operational employees primarily need a system that allows them to perform their work quickly and reliably.

Good system architecture standardises where standardisation helps – and allows differences where they are part of the business model.
SYSTEM ARCHITECTURE & OPERATIONAL REALITY Are you facing a system implementation or standardisation initiative where central requirements and operational reality have to be brought together?

An initial confidential conversation is enough to assess the situation, responsibility and potential assignment framework.

Discuss an assignment confidentially