This post describes the CORE check — a framework for evaluating which business processes are worth re-evaluating or redesigning, and what the redesign should focus on in each case. It concludes with a comparison of CORE against established business process management frameworks and a complete assessment concept.
The CORE check
CORE stands for four letters, four assessment dimensions:
- C — Correctness: Does the process produce correct results?
- O — Operating effort: What does it cost to produce the output?
- R — Resilience: Does the process survive disruptions?
- E — Essentiality: How important is the process to the organization?
C, O and R are health dimensions — they describe how the process is doing. E is a prioritization dimension — it does not describe health but decides whether investing in the process is worthwhile at all. The health dimensions are independent: a process can be perfectly fine on one dimension and critical on another — the dimensions do not compensate for each other.
C — Correctness: Does the process work?
The first question is whether the process delivers what it is supposed to deliver. Errors here are particularly expensive because they multiply downstream: incorrect data produced by one process causes rework, wrong decisions or complaints in every subsequent process.
Indicators:
- Defect rate of the output that becomes visible downstream
- Rework and correction effort in subsequent processes
- Complaints from internal or external “consumers” of the process output
- Frequent follow-up questions because results are incomplete or implausible
O — Operating effort: Is the process itself in need of optimization?
Independently of whether the output is correct: what does it cost to produce the output? A process can deliver error-free results and still be dysfunctional — for example because it ties up staff capacity unnecessarily, consists of media breaks, or is error-prone (i.e. errors are only caught by laborious manual checks before they leave the process).
Indicators:
- Staff effort per run and cycle time
- Share of manual steps, copy-paste work, media breaks
- Internal correction loops: how often does work have to be fixed inside the process before the output is right?
- Cost per case relative to the value of the output
The distinction from C matters: error-proneness (O) is not the same as defect rate (C). A highly error-prone process whose errors are caught internally produces correct outputs — but ties up capacity and is latently risky.
R — Resilience: Does the process survive disruptions?
The third question is what happens when something goes wrong: an employee falls ill, a tool fails, a supplier drops out, case volume suddenly spikes.
Indicators:
- Bus factor: how many people truly master the process? At 1, there is no backup at all.
- Documentation level: is the process documented well enough for a non-expert to execute it?
- Single points of failure: people, specialized tools, individual interfaces or suppliers
- Recovery time: how long until the process runs normally again after a disruption?
A typical pattern: a process runs correctly and efficiently but depends on a single person. As long as nothing happens, it looks perfectly healthy — the R dimension makes the hidden risk visible.
E — Essentiality: How important is the process?
Essentiality is deliberately not a health dimension but the prioritization dimension: it does not assess the state of the process but its significance — revenue, compliance and customer impact, consequences of failure, frequency. A dysfunctional process with low E is not a redesign candidate; a mediocre process with high E is. Without E you can assess processes but not prioritize them.
Comparison with standard frameworks
The following table maps the usual business process management frameworks onto the four CORE dimensions — details per framework below:
| Standard framework | C — Correctness | O — Operating effort | R — Resilience | E — Essentiality |
|---|---|---|---|---|
| Devil’s Quadrangle (Brand & van der Kolk) | Quality | Time, Cost | Flexibility (partial) | – |
| BPR selection criteria (Hammer & Champy) | Dysfunction (partial) | Dysfunction (partial) | – | Importance, Feasibility |
| FMEA | Severity × Detection | Occurrence | – | Severity |
| Lean / DOWNTIME | Defects | remaining 7 waste types | – | – |
| SCOR | Reliability | Responsiveness, Cost, Asset Management | Agility | – |
| PEMM (Hammer) | Metrics | Design, Infrastructure | Performers, Owner | – |
| BPM Goal Hexagon | Quality | Time, Economics | Flexibility, People (partial) | – |
Devil’s Quadrangle
The classic for business process redesign (Brand & van der Kolk, 1995; popularized by Reijers & Liman Mansar) assesses processes on four dimensions: Time, Cost, Quality and Flexibility. The key insight of the model is that the dimensions stand in trade-off relationships: more quality assurance costs time, more standardization costs flexibility. The quadrangle was designed to evaluate the impact of redesign measures — not to select candidates for redesign.
Difference to CORE: Quality is a single dimension in the quadrangle; CORE splits it into external correctness (C) and internal error-proneness (O) — the FMEA shows why this matters: the two cases require completely different redesign levers. Flexibility only means coping with variation (case types, volume); Resilience in CORE additionally covers resource failure. And the quadrangle has no prioritization dimension like Essentiality at all — it assesses process states, not process importance. What you should take from the quadrangle is the trade-off lens (see step 6). For the actual selection question, CORE directly provides the answer the quadrangle was never designed to give.
BPR selection criteria after Hammer & Champy
Hammer & Champy assess processes for reengineering selection using three criteria: Dysfunction (how broken is the process), Importance (how strongly it affects customers/results) and Feasibility (how realistic a redesign is).
Difference to CORE: Hammer & Champy say that you should judge dysfunction — but not how. CORE is essentially the operationalization of dysfunction via the three health dimensions C, O and R: instead of a gut-feeling score, you get concrete indicators per dimension. Importance corresponds to the E in CORE. Feasibility is deliberately absent: in a first iteration concerned only with taking stock, it would merely complicate the assessment — it becomes relevant once prioritized candidates turn into concrete redesign projects.
FMEA
Failure Mode and Effects Analysis assesses individual failure modes of a process via Severity × Occurrence × Detection and prioritizes them using the risk priority number. The analysis works bottom-up: you list every conceivable failure individually and rate it.
Difference to CORE: FMEA provides exactly the logic needed to cleanly separate C and O — high occurrence with good detection (controls catch everything) is an operating-effort problem, poor detection is a correctness problem. Severity roughly corresponds to the E in CORE. FMEA does not cover resilience: it assesses failure modes, not the ability to recover after disruptions. Once CORE has flagged a C-suspect process, you can dig deeper there with a real FMEA — but for a rough survey across dozens of processes, FMEA’s bottom-up effort is oversized.
Lean / DOWNTIME
Lean assesses processes via eight types of waste: Defects, Overproduction, Waiting, Non-utilized Talent, Transportation, Inventory, Motion, Extra-processing (mnemonic DOWNTIME). The method is not a selection technique but a diagnostic technique: you walk a concrete process step by step and hunt for waste.
Difference to CORE: Defects corresponds to C; the remaining seven waste types operationalize O — R and E simply do not appear in Lean. So Lean tells you where the process is bleeding, but not whether it is the right process. The combination is still powerful: CORE locates the O-suspect process, DOWNTIME then provides the checklist to dissect the deficit in detail.
SCOR
The Supply Chain Operations Reference model (the name similarity to CORE is a coincidence — the two are independent) assesses processes on five attributes with concrete level-1 metrics:
| SCOR attribute | Level-1 metric (example) | CORE dimension |
|---|---|---|
| Reliability | Perfect Order Fulfillment (right output, right time, right documentation) | C — Correctness |
| Responsiveness | Order Fulfillment Cycle Time | O — Operating effort |
| Cost | Cost of Goods Sold, SCM costs | O — Operating effort |
| Asset Management Efficiency | Cash-to-Cash Cycle Time, Inventory Days | O — Operating effort |
| Agility | Upside/Downside Flexibility & Adaptability, Value at Risk | R — Resilience |
SCOR is a KPI framework with benchmark data (ASCM) and a fixed process taxonomy (Plan/Source/Make/Deliver/Return/Enable) — a measurement framework for benchmarking processes against industry figures.
Difference to CORE: SCOR splits O into three (speed, process cost, capital tie-up) and provides with Agility the most concrete resilience operationalization of any standard framework — e.g. days until a process can handle 20% more unplanned volume, or Value at Risk. If you want to instrument the R in CORE, you will find ready-made metric definitions here. Reliability includes punctuality alongside correctness, which CORE deliberately separates. There is no E equivalent: SCOR measures processes but does not prioritize them. The price for metric depth is weight — SCOR is tailored to supply-chain processes and presumes a measurement infrastructure. For a fast, process-agnostic first assessment without a KPI data basis, CORE is the more accessible instrument.
PEMM (Hammer’s Process and Enterprise Maturity Model)
PEMM (Hammer’s Process and Enterprise Maturity Model) assesses not process outcomes but process enablers: Design, Performers, Owner, Infrastructure and Metrics — supplemented by enterprise capabilities such as Leadership and Culture. It is a maturity model: you rate how institutionalized a process is managed.
Difference to CORE: CORE asks “how healthy is the process”, PEMM asks “how well is it managed” — complementary perspectives with overlaps at R (Performers: knowledge/backup; Owner: accountability in case of failure) and indirectly at C (Metrics: is output quality even measured?). For the selection question, CORE’s outcome view is the more direct indicator: a process can be PEMM-mature and well-documented yet still tie up capacity or deliver wrong outputs — and conversely run healthily without anyone ever having formalized it.
BPM Goal Hexagon
The BPM Goal Hexagon (2026) is the most recent extension of the Devil’s Quadrangle: it replaces Cost with the broader dimension Economics and adds People (does the process burden employees, is the work meaningfully designed?) and Environment (sustainability of the process). Like the quadrangle, it assesses redesign goals and impacts, not candidate selection.
Difference to CORE: People is only partially covered in CORE via O (effort) and R (personnel dependency) — overloaded employees are often the very cause of outages. Environment plays a role in the question “is redesign worthwhile” only in regulated or ESG-relevant contexts. CORE deliberately stays at four dimensions — anyone working in a People- or ESG-sensitive environment can add the hexagon dimensions as an optional extension without restructuring the framework.
In summary: what CORE takes from the standard frameworks — and what it lacks
- Essentiality (from BPR selection criteria and prioritization matrices): decides whether a redesign is worthwhile — built into CORE as E.
- Feasibility (Hammer & Champy): deliberately not in CORE — relevant only once prioritized processes become concrete redesign projects.
- Trade-off relationships (Devil’s Quadrangle): improvements in one dimension can worsen others — see step 6.
- Detection logic (FMEA): makes the separation of C and O operational.
- Waste types (Lean/DOWNTIME): diagnostic checklist once O is flagged.
- People dimension (Hexagon): optional when employee burden or ESG is relevant in your context.
The complete concept: CORE in practice
Step 1: Inventory your processes
List the relevant processes with output, downstream consumers, frequency and rough effort. The 10–20 most important processes usually suffice for a first pass.
Step 2: Score health (C, O, R)
Assess each process on the three health dimensions. A rough 3-tier scale (good / medium / critical) is sufficient for prioritization; if you measure, you can use KPIs:
| Dimension | Good | Medium | Critical |
|---|---|---|---|
| C — Correctness | Hardly any downstream errors | Occasional errors, catchable | Regular defective outputs |
| O — Operating effort | Little manual effort | Noticeable manual share | Ties up disproportionate capacity |
| R — Resilience | Backup available, documented | Backup theoretically possible | Bus factor 1, undocumented |
Step 3: Assess Essentiality (E)
Assess separately from the health score: revenue/compliance/customer impact of the process, consequences of failure, frequency. Here too, low / medium / high suffices. The third BPR factor Feasibility (is the process even changeable?) is deliberately not part of CORE — it only comes into play once prioritized candidates become concrete redesign projects.
Step 4: Prioritize
Combine health (C, O, R) and Essentiality (E) in a 2×2 matrix:
- High E + critical health → redesign candidate, immediately
- High E + medium health → schedule a re-evaluation
- Low E + critical health → take quick wins only, or question the process (is it needed at all?)
- Low E + good health → do nothing
Step 5: Derive the redesign focus from the dominant deficit dimension
The health dimension with the worst score determines what the redesign concentrates on:
| Dominant deficit | Typical redesign levers |
|---|---|
| C — Correctness | Validation at the source, automated plausibility checks, poka-yoke, post-processing before handover |
| O — Operating effort | Automating manual steps, removing media breaks, eliminating waste (Lean/DOWNTIME checklist), standardization, parallelization |
| R — Resilience | Documenting knowledge, cross-training, backup arrangements, automation (reduces personnel dependency), defining fallback procedures, redundant suppliers/tools |
Step 6: Make trade-offs explicit
The lesson from the Devil’s Quadrangle: measures in one dimension can worsen others. More quality controls (C) cost effort (O), full automation can reduce resilience (tool outage), documentation (R) costs effort (O). So define target values on C, O and R before redesigning — not only on the problematic dimension.
Summary
- CORE = Correctness (does the output work?), Operating effort (what does producing it cost?), Resilience (does the process survive disruptions?), Essentiality (how important is the process?).
- C, O and R are health dimensions and determine what a redesign focuses on; E is the prioritization dimension and determines whether a redesign is worthwhile. Feasibility only becomes relevant during concrete redesign planning.
- Compared to standard frameworks: the quality split into defect rate vs. error-proneness (C vs. O) follows FMEA logic (occurrence × detection), R goes beyond the quadrangle’s flexibility (closest to SCOR agility), and E comes from the BPR selection criteria of Hammer & Champy.
