From first customer interest to start of supply: how process intelligence exposes rework, seasonal construction times and compliance risks in district heating connections – and turns them into prioritised measures.
Lead-to-Activation is the end-to-end process from a newly created lead to an active service at the customer. For a district heating connection it covers sales and offer, the parallel coordination of civil works, materials and house connection, construction and commissioning, and finally the handover report, billing and start of supply. From the customer’s side it sounds simple – interest, offer, contract, construction, connection, heat. In reality it is one of the most complex processes in the energy sector.
Note: the figures shown come from a synthetic demonstration dataset. They illustrate analysis paths and methodological relationships and are not production figures of any customer.
13%
Offers reworked
154 of 1,208 cases between offer preparation and sending
+20 days
Construction in winter
A recurring seasonal pattern from November to March
121
Supply before handover
connections start supply before the handover report is finalised; 45 skip it entirely
119
Missing permits
of 1,208 cases lack a timely official permit before construction starts
Even before the order, information sits in different systems and objects. After the order the flow branches into parallel strands that have to be joined again before construction, commissioning, billing and start of supply.
Phase 1
A lead is created and qualified. Qualified leads get a technical calculation, on which the offer is prepared, sent and finally signed as a contract.
Business objects
Guiding question
How often does an offer go back for rework – and why?
Phase 2
After the order several strands run at once. They have different durations, objects and cardinalities – a single connection can create several purchase orders, deliveries and invoices.
Business objects
House connection & planning
Subcontracting & civil works
Materials: pipes, cables, equipment
Guiding question
Which prerequisite is holding back the shared next step – the start of construction?
Phase 3
Construction only starts once all strands are joined. The end of construction is followed by acceptance and technical commissioning. When several connections are bundled into one civil-works job, the cases also depend on each other.
Business objects
Guiding question
Which part of construction slows down, from when, by how much – and does it repeat?
Phase 4
The handover report has to reach certain states before supply starts. Commercial finalisation with billing runs in parallel.
Business objects
Guiding question
Were all prerequisites met at the moment supply started?
A business case is not the same as a single business object. What matters is that the analysis keeps objects and their relationships instead of aggregating them early into a flat event log – only then can every finding be traced back to the affected orders, purchase orders or reports.
The following patterns were uncovered with Noreja. Each answers a concrete business question – and moves it from “What happened?” to “Why did it happen and what follows from it?”.
Finding 1 · Sales
Instead of counting every repetition as a new process variant, it is interpreted as rework: a recurring error pattern that delays contract signature and ties up sales capacity.
The same pattern shows up again later: in billing, about 106 of 1,089 cases show a later correction. Rework is therefore not an issue of one department but a cross-process theme.
154 / 1,208
reworks between “Offer prepared” and “Offer sent” (≈ 13%)
Instead of “Which variant is this?”
the analysis asks: “Which error pattern is this, why does it occur and what is its impact?”
Finding 2 · Construction
A long lead time in a single case says little. So the same activity is looked at across all process instances and turned into a time series: from November the duration of construction rises noticeably, stays high through the winter and drops again from April.
The time series stays connected to the process structure and can be traced back to concrete periods, cases and objects.
≈ 20 days
longer construction in the winter months
Instead of “Is the process slower in winter?”
the analysis asks: “Which specific part slows down, from when, by how much – and does it repeat?”
Finding 3 · Compliance
The handover report has to reach certain states before supply starts. The business question therefore matters more than the plain sequence of events.
For field processes one distinction is crucial: a late timestamp in the ERP – caused by missing connectivity or later synchronisation – does not mean the real process was late. Process errors and data errors look the same in an event log but need completely different measures.
121 · 45
connections start supply before finalisation (≈ 11%) · finalisation skipped entirely (4.1%)
Instead of “In which order did the events happen?”
the analysis asks: “Were all prerequisites met when supply started?”
Finding 4 · Trade coordination
No single activity decides when construction can start – several prerequisites have to be met together. 393 units had to be sourced from another supplier because the first could not deliver; 119 of 1,208 cases lack a timely official permit; and in some cases construction does not start despite completed technical work because a commercial prerequisite such as the customer’s down payment is missing.
That is the core of causal process logic: not only knowing what happened, but which prerequisite is missing for the next step to happen.
393 · 119
units sourced from a second supplier · cases without a timely permit
Instead of “Which activity takes longest?”
the analysis asks: “Which prerequisite is holding back the shared next step?”
The AI assistant Minerva summarises the process areas and names the biggest time levers. The process manager starts with a business question instead of first choosing a view, filter and visualisation.
| Process area | Key statement | Biggest time lever | Duration |
|---|---|---|---|
| Sales | High conversion of qualified leads into orders | Technical calculation to offer | 27 days |
| Trade coordination | Several parallel prerequisites decide construction start | Civil-works permit to construction start | 77.8 days |
| Connection & commissioning | Stable flow, but long construction and acceptance phase | Construction plus wait until acceptance | 69.2 days |
| Handover reports | Documentation marker before order receipt | Time impact not measurable | – |
A finding does not improve a process yet. In the Impact Board, findings are rated by impact, implementation effort and uncertainty – process excellence and the business decide together which measure pays off economically and organisationally.
| Priority | Measure | Expected benefit |
|---|---|---|
| 1 | Bundle permit and construction-start management | Shorten the biggest bottleneck of 77.8 days |
| 2 | Link construction planning to a binding start-approval process | Reduce the 55.7 days after planning ends |
| 3 | Standardise technical calculation and offer preparation | Shorten the 27 days in sales |
| 4 | Record handover reports completely, in time and content | Transparency on documentation and handover delays |
| 5 | Prepare construction acceptance early | Reduce the 27.8 days between construction end and start of acceptance |
| 6 | Analyse invoice and supplier approvals | Less waiting after deliveries and fewer invoice corrections |
| 7 | Make multiple orders transparent | Fewer avoidable procurement and coordination loops |
Improvement does not end with implementation. The question is not “Have we used AI now?” but “Has the process become faster, more stable or better in quality?”
Insight
Spot the error pattern, time series or missing prerequisite
Action
Derive, rate and prioritise a measure
Impact
Quantify the economic effect and implement
Feedback
Measure the effect against the process data again
Business objects and their relationships are not aggregated early into a flat event log.
Analyses answer concrete process questions without losing the underlying information.
Process behaviour is assessed against business and causal dependencies.
Findings become measures, are valued economically and then measured against the process data again.
Lead-to-Activation is the end-to-end process from a newly created lead to an active service at the customer. For a district heating connection it runs from lead qualification, technical calculation, offer and contract through the coordination of civil works, materials and house connection, construction and commissioning, to the handover report, billing and start of supply.
Because after the order the process branches into several parallel strands – civil works or subcontracting, material procurement, house connection and commercial prerequisites – with different durations and cardinalities. One connection creates several purchase orders, deliveries and invoices, and bundled civil-works jobs create dependencies between cases. A single process map or a flat event log cannot represent that.
About 13% of offers go back for rework (154 of 1,208), construction takes about 20 days longer in winter, 121 connections start supply before the handover report is finalised, and construction start depends on several prerequisites – such as missing permits in 119 cases. The biggest time lever lies between civil-works permit and construction start, at 77.8 days.
In the event log both look the same. A late ERP timestamp can come from missing connectivity or later synchronisation in the field without the real process being late. Only a business assessment against objects, states and dependencies shows whether a process measure or a data measure is needed.
No. The figures come from a synthetic demonstration dataset. They illustrate analysis paths and methodological relationships and are not production figures of any customer.
Event knowledge graph, perspectives, error patterns, causal process logic, bundling and the path from finding to improvement initiative – step by step in the Noreja Help Center (in German).
Read the case in the Help CenterJoin companies that are redefining growth and efficiency with intelligent process analysis.
