Anyone who handles mass proceedings knows the problem: the sheer volume of cases running at the same time places considerable demands on any organisation. Without powerful orchestration in the background, that volume can hardly be managed any more.
At what case volume does automation actually pay off? How can you tell a genuine enterprise platform from a law firm tool with a coat of paint? And which twelve functions must a solution for mass proceedings absolutely offer? This article answers these questions soberly, without sales promises.
01 Volume
When manual handling reaches its limits
Every growing law firm and every legal expense insurer eventually reaches a point where the existing way of working no longer holds up. This is rarely down to the team. It is because case volume crosses a threshold beyond which purely manual processes structurally cannot keep pace.
Experience shows that this threshold usually lies between 300 and 500 cases a month. Below it, well-organised teams and solid law firm management software can still work reasonably efficiently. Above it, quality begins to erode gradually: deadlines get tighter, responses take longer, errors increase and the strain on staff grows.
The typical reaction is expensive and not very effective: hiring additional staff, outsourcing tasks or both – without solving the actual problem.
Because the actual problem is not a staffing issue. It is an automation deficit.
Three patterns that arise without automation:
- Linear growth: if volume doubles, staffing needs double – costs rise 1:1 with the number of cases.
- Declining quality: beyond a certain workload, the depth of case handling and the error rate suffer systematically.
- Bottleneck effects: individual process steps – such as document intake or deadline monitoring – block the entire workflow.
02 Automation, but how?
What “automation” really means in mass proceedings
The term automation is used far too freely in the legal tech sector. Many vendors really mean nothing more than digitised manual work: a document can be uploaded faster, a follow-up can be set up with one click, a form can be completed digitally instead of on paper.
That is digitisation, not automation – and it does not solve the scaling problem.
| Digitisation (not enough) | Automation (what is needed) |
|---|---|
| Documents arrive digitally but are still opened and sorted manually | Incoming documents are classified and assigned automatically by AI |
| A follow-up is set with one click | Follow-ups are created, postponed and completed automatically on the basis of rules |
| A pleading is sent digitally | Pleadings are generated in context and sent fully automatically |
| A case is assigned to the responsible handler by hand | Assignment is automatic, based on fixed criteria such as area of law, workload or priority |
| The case status is updated manually | The status updates itself automatically, triggered by events in the process |
| A deadline is entered in the system by hand | Deadlines are extracted from documents with AI support and set automatically |
True automation means: nobody has to push a case forward actively any more. The process runs by itself – people decide only where their judgement is genuinely needed.
03 Bottlenecks
The four most common bottlenecks in high-volume business
Anyone who runs mass proceedings knows these bottlenecks – even if their cause is not always immediately obvious:
- Document intake – Hundreds or thousands of documents a day, spread across email, fax, beA (the special electronic lawyers’ mailbox), post and upload. Each one has to be opened, read, assigned and filed. Without automation, this step alone consumes enormous staff capacity. With AI-supported classification and event-based workflows, it becomes the trigger for a fully automatic end-to-end process.
- Deadline management – In mass proceedings, deadline monitoring is the biggest liability risk. A missed deadline in five cases a week is an outlier. With 500 cases a day, it is statistically almost unavoidable if deadlines are recorded manually. Automated deadline detection by AI and rule-based escalation paths remove this risk structurally.
- Communication with the parties involved – Status enquiries from clients, reminder letters to debtors, acknowledgements of receipt to courts: each of these messages takes time when handled manually, and across thousands of cases that time adds up to an enormous total effort. Fully automated, event-triggered communication solves this problem without additional staff.
- Handling exceptions and process changes – Mass proceedings are rarely truly homogeneous. Special constellations, deadline extensions or unexpected objections create exceptions that can throw a standardised process off course. An enterprise platform must model such exceptions explicitly in the workflow – through predefined alternative paths, not through manual workarounds.
“Companies confronted with mass claims and proceedings often only approach us once the pain is already severe. The earlier processes are orchestrated on a platform, the sooner peak loads and costs can be contained.” – Tim Platner, Managing Director, Legal Data Technology GmbH
04 Must-haves
The 12-point checklist: what a platform for mass proceedings must be able to do
Not every legal tech platform is suited to mass proceedings. Many are excellent for sole practitioners or boutiques with complex individual cases – but fail as soon as volume rises. The following checklist helps you identify the right solution.
Mandatory criteria
Without these functions, a platform is not suitable for mass proceedings:
- Event-based workflow triggers: Processes start automatically on defined events – receipt, timer, status change.
- Batch processing: Workflows can be started for thousands of cases at once.
- AI integration in the workflow: AI steps are a fixed, defined part of the process – not an isolated add-on tool.
- Explicit exception handling: Errors, escalations and special cases are modelled in the workflow itself.
- Audit-proof process history: Every step and every decision is documented in an auditable way.
- Multi-tenancy: Different clients, products or tariffs can be managed on one platform.
- Workflow versioning: Process changes are versioned and can be rolled back without interrupting live operation.
Quality criteria
These functions distinguish a good platform from a very good one:
- BPMN standard: Processes are modelled in an open, cross-system standard – portable and extensible.
- Human-in-the-loop as part of the process: Human decisions are explicitly anchored in the workflow.
- Flexibility with AI models: Different language models (LLMs) – local, cloud-based, specialised – can be swapped and combined.
- Parallel workflow versions: Several versions of a process can be active at the same time.
- Integrated receivables management: Postings, bank transfers and receivables tables are mapped directly in the platform.
05 Standards
BPMN as the technical foundation: why the standard makes the difference
Behind many automation promises in legal tech lies a proprietary workflow engine – a tool that works only within its own platform, whose logic cannot be exported and which requires developers to change.
The problem becomes apparent at the third scaling step at the latest: the process landscape grows to hundreds of workflows. Every adjustment needs coordination, and every new requirement costs development time. The platform that was supposed to save work becomes the bottleneck itself.
BPMN (Business Process Model and Notation) solves this problem at its root: processes are mapped in an internationally established standard that is both a visualisation and executable code.
What this means in practice:
- Business users model processes themselves – Lawyers, case handlers and project leads design processes graphically, without needing development skills.
- A single reliable source – The BPMN model is both the documentation and the actual execution logic: what is modelled runs in the system exactly as drawn.
- Safe to adapt – Process changes are made in a controlled way, versioned and documented traceably – instead of through risky interventions in the code.
- Usable across systems – Workflows can be exported and imported between systems without ending in vendor lock-in.
06 Examples
Automation before and after: a look at the numbers
The following benchmarks come from typical enterprise projects involving legal mass proceedings. They serve as a guide – actual results depend on process depth, case type and the maturity of the implementation.
| Process area | Manual effort (before) | Automated effort (after) |
|---|---|---|
| Document intake & classification | 6–12 min / document | < 30 sec. (AI + workflow) |
| Deadline detection & setting | 3–8 min / pleading | Fully automatic (0 min. manual) |
| Standard correspondence (acknowledgement of receipt, reminder letters) | 5–15 min / letter | Generated and sent fully automatically |
| Case assignment & forwarding | 2–5 min / case | Rule-based, < 5 sec. |
| Status updates | 1–3 min / event | Automatic, event-based |
| Escalation handling | 15–30 min / exception | User task: 2–3 min. decision |
| Bulk dispatch (e.g. 500 beA pleadings) | 2–3 working days | One batch run (< 1 hour) |
| Overall degree of automation | 5–15% (manual tools) | 75–92% (enterprise platform) |
Anyone who wants to start workflows in hundreds of case files at once with a single click will find more in DEPLAW’s workflow batch processing.
07 ROI
When does automation pay for itself? An example calculation
Many decision-makers shy away from investing in an enterprise automation platform because the implementation costs are visible immediately, while the savings initially seem abstract. A transparent example calculation brings clarity:
Example: law firm with 800 cases a month
Assumptions: €45 staff costs per hour, an average of 90 minutes of manual handling time per case
- Monthly manual effort: 800 cases × 90 min × €45/h = €54,000/month
- With an 80% degree of automation: 800 × 18 min × €45/h = €10,800/month
- Monthly saving: €43,200 – annual saving: €518,400
- Typical payback period for the platform investment: 3–7 months
On top of this come non-monetary effects: fewer errors, greater deadline security, scalability without additional hires and more satisfied employees.
For legal expense insurers with several thousand cases a day, these amounts multiply accordingly. The free DEPLAW ROI calculator gives a first indication of your own savings potential.
08 DEPLAW for enterprises
Who DEPLAW is right for – and who it is not
DEPLAW is not a solution for everyone. An honest assessment:
| DEPLAW is a good fit if you … | DEPLAW is probably not a good fit if you … |
|---|---|
| regularly handle more than 300 cases a month | run a small law firm with fewer than 50 cases a month |
| want to fully automate recurring legal processes | mainly handle complex individual cases with no recurring patterns |
| need beA, AI and document processing in one system | are merely looking for simple case file management without process automation |
| want to make process changes yourselves – without developers | have no internal team that can design workflows from a practitioner’s perspective |
| need an audit-proof, auditable solution for a regulated industry | are looking for an inexpensive entry-level solution for first digital steps |
09 FAQ
Frequently asked questions about legal tech for mass proceedings
At what case volume does an enterprise platform pay off?
As a rule of thumb, an automation platform pays off economically from around 300 cases a month with recurring process structures. For law firms with debt collection, traffic law or many parallel court proceedings, and for legal expense insurers, the threshold can be lower – especially if beA communication, deadline management and document processing are already becoming a manual bottleneck today.
Can we start with a simple process and scale up later?
Yes – we even expressly recommend it. A good starting point is recurring reminder routines: send a letter, wait for the deadline, remind automatically, proactively inform the claimant. More complex processes can then be built on this step by step – such as AI-supported analysis of replies, automatic detection of deadline extensions or fully automated court payment orders.
How do you deal with special cases and exceptions?
In a BPMN workflow, exceptions are not a disruption but a modellable part of the process. Through event subprocesses, error events and clearly defined alternative paths, special cases are handled not through manual workarounds but as a regular part of the process – traceable, auditable and consistent.
How long does implementation take?
In our experience, the first automated beA workflows go live within a few days. DEPLAW is designed so that business users without development skills can create their own workflows in the BPMN editor and put them into operation straight away. Implementation time depends mainly on the complexity of the processes you want – not on technical hurdles.
10 Conclusion
Mass proceedings are orchestration projects, not staffing projects
Anyone who tries to cope with legal mass proceedings using purely manual processes or simple digitisation tools is fighting symptoms rather than causes. The real challenge is architectural: how do you build a process that works as reliably for the thousandth case as for the first?
The answer lies in a complete workflow platform that detects events, integrates AI, orchestrates decisions and involves people only where their expertise is really needed.
This is not a future scenario but achievable today – and for law firms and insurers with the relevant volume, it is no longer a nice-to-have.