When planning an S/4HANA transformation, most organisations focus on system capabilities, and rightly so. The platform offers real-time analytics, a simplified data model, and embedded best practices that can fundamentally improve performance. But understanding what the technology can do is not the same as understanding what your organisation actually needs it to do, and the distance between those two things is where most S/4HANA programmes run into trouble.
The issue is straightforward. Without a clear, evidence-based understanding of how your processes work today, even the most advanced ERP will simply accelerate the problems you already have. Manual workarounds that slow delivery will be carried forward. Compliance gaps that currently rely on informal controls will be embedded into the new system. Recurring issues that should be automated or redesigned will persist, only now they will be running on a platform that costs significantly more to implement.
A recent Horváth study of 200 companies found that budgets were exceeded in more than six out of ten S/4HANA migrations, with nearly two-thirds of organisations reporting significant quality deficiencies after completing the transition. The study’s director attributed many of these problems to an incorrect programme setup, in which the project’s complexity and required resources were underestimated, while organisational competence was overestimated. Process understanding sits at the heart of this challenge. You cannot correctly scope, resource, or govern a transformation programme if you do not have an honest view of how work actually flows through the organisation today.
The AS-IS model as a diagnostic, not documentation
This is where the AS-IS process model comes in, and where it is most commonly misunderstood.
In many S/4HANA programmes, AS-IS modelling is treated as a documentation exercise. The programme team runs workshops, captures how processes work in the current system, and produces a set of models that describe the status quo. These models are then filed away as a reference point while the real focus shifts to the TO-BE design and the configuration of the new platform.
This approach misses the point entirely. The value of an AS-IS model is not in recording what exists. It is in revealing what needs to change. When done properly, the AS-IS model functions as a diagnostic tool that surfaces the issues which would otherwise be carried silently into the new system.
Consider what a well-constructed AS-IS analysis exposes. It identifies where manual steps are creating bottlenecks in processes that should be running much faster. It highlights where informal workarounds have been built around system limitations, which will not translate cleanly to a new platform. It shows where compliance controls depend on individual knowledge rather than system-enforced rules. And it quantifies the scale of process variation, revealing how differently the same process is executed across regions, business units, or teams.
In SAP Signavio, this diagnostic capability is supported directly. Process models in SAP Signavio Process Manager provide the structural view of how the process is designed. Process Intelligence adds the data layer, using process mining to show how the process actually executes in the system. The combination of both is what gives transformation leaders the evidence they need to make informed decisions about what to standardise, simplify, and redesign as part of the S/4HANA programme.
Why skipping this step is so costly
The temptation to move quickly past AS-IS analysis is understandable. S/4HANA programmes are expensive, time-pressured, and politically complex. There is a natural desire to focus energy on the future state rather than spending time documenting the present. But the consequences of skipping this step are well documented and consistently expensive.
Without a clear AS-IS baseline, fit-to-standard workshops become speculative rather than evidence-based. The programme team is comparing SAP’s best practice processes against an assumed understanding of current operations, which is almost always incomplete and often inaccurate. Decisions about which processes to standardise and which to customise are made without data, increasing the risk of unnecessary customisation (which drives up cost and complexity) or premature standardisation (which creates resistance and adoption problems downstream).
Scope creep is another common consequence. When the programme lacks a shared, visible understanding of the current process landscape, new requirements surface late in the Realise phase as teams discover gaps and dependencies that were not captured during Explore. These late discoveries are among the most expensive problems in any ERP programme because they force rework when the system is already being configured, and they introduce delay at precisely the point where the programme can least afford it.
There is also a governance dimension. In large, multi-entity organisations, the AS-IS model is often the only artefact that creates a shared language across the programme. Without it, different workstreams operate with different assumptions about how the business runs, leading to design decisions that conflict with each other and integration testing that reveals fundamental misalignments far too late in the programme lifecycle.
Changing the nature of the programme
When AS-IS analysis is treated as a genuine diagnostic rather than a tick-box exercise, it changes the character of the entire transformation. The programme is no longer simply migrating processes from one system to another. It is actively improving them during the migration, which is where the real return on investment lies.
This is a critical distinction. Organisations that treat S/4HANA as a system migration tend to measure success in terms of go-live dates, data migration accuracy, and functional parity with the old platform. Organisations that treat it as a process transformation measure success in terms of cycle time reduction, automation rates, compliance improvement, and cost per transaction. The AS-IS model enables the second approach by providing the baseline against which improvement can be defined, targeted, and measured.
In practical terms, this means using the AS-IS analysis to identify a focused set of process improvements that can be designed into the TO-BE state from the outset. Rather than configuring S/4HANA to replicate today’s processes and then attempting to optimise later (which rarely happens once the programme team has moved on), the transformation itself becomes the vehicle for improvement. Manual approvals are replaced with automated routing. Exception handling that currently requires offline communication is built into system workflows. Process variants that exist for no defensible reason are harmonised into a single standard.
SAP Signavio supports this transition directly. Process Insights can provide a rapid initial view of where the most significant gaps and opportunities lie across the SAP landscape. Process Intelligence enables deeper analysis of specific end-to-end processes, quantifying the scale of the problem and prioritising where improvement will have the greatest impact. In addition, the Collaboration Hub ensures that the agreed-upon TO-BE process is visible and accessible to everyone involved in the programme, creating a single source of truth that connects design decisions to how work will actually be done in the new system.
The foundation question
Every S/4HANA transformation is ultimately built on a foundation of process understanding. The question is not whether to model the AS-IS state, because virtually every programme does so in some form. The question is whether you are using that model as a genuine diagnostic that informs the design of the future state, or simply producing documentation that gets filed alongside the programme governance pack and never revisited.
Organisations that treat AS-IS modelling as a strategic input to the transformation consistently make better scoping decisions, experience fewer late-stage surprises, and deliver programmes that change how the business operates rather than simply changing the system it operates on. The technology is only as strong as the foundation it is built upon, and that foundation starts with an honest, evidence-based understanding of where you are today.