Customer Experience in ISO: How 10001, 10002, 10004 and 23592 Integrate into One System
Four ISO guidance standards govern customer experience: 23592 for service excellence, 10001 for codes of conduct, 10002 for complaints handling and 10004 for satisfaction measurement. This guide explains how they integrate into one system built on a shared core — and why they should not be built as four separate systems.
When an organisation decides to improve its customer experience through an established methodology, it quickly runs into four numbers: ISO 23592, ISO 10001, ISO 10002 and ISO 10004. Then the confusion begins. Do we build four systems? Which one comes first? And what is the difference between measuring satisfaction and handling complaints when both deal with a dissatisfied customer?
The common answer is to build each standard as an independent system with its own policies, procedures and records, then collect them in one folder and call the result integrated. That answer is both expensive and wrong: expensive because it produces four context policies, four management reviews and four databases that never speak to one another; wrong because the drafters of the standards themselves stated the links between them in the body of those standards, not in commentary written afterwards.
This guide sets out the architecture behind the Integrated Service Excellence and Beneficiary Satisfaction Framework: one core and five layers, grounded in the text of the four standards, and translated into 117 working documents and 16 weighted performance indicators. Six of the system files are attached at the end of this article for direct download.
Four standards, not four systems
One idea summarises the whole article: integrating processes is not the same as merging standards. Merging is only permissible where an activity can be executed as a single operational process without any of the reference standards losing its purpose. The architecture was built on that criterion alone: whatever could be unified was unified in the core, and whatever belonged to a specific standard remained a distinct module retaining its own terminology, forms and records.
The clearest expression of this principle is the beneficiary journey itself: every stage is grounded in a standard, and the loop only closes when the effect of improvement returns to the design of the service.
What does each standard actually do?
ISO 23592:2021 — Service excellence
A capability model of four dimensions and nine elements defining the vision, culture, employee engagement and experience design of a service organisation. It is not an operational backbone, which is why it sits above the process layer rather than beside it.
ISO 10001:2018 — Codes of conduct
Guidance for drafting the promise made publicly to the customer: what the organisation commits to, within what timeframe and to what standard, and how fulfilment is evidenced. It is primarily preventive — a clear commitment stops the complaint before it occurs.
ISO 10002:2018 — Complaints handling
The third edition, confirmed in the 2023 systematic review. It covers the full complaint cycle: receipt, acknowledgement, tracking, classification, escalation and closure, then the root cause analysis that turns a complaint from an incident into knowledge.
ISO 10004:2018 — Satisfaction measurement
It defines satisfaction as the gap between what the beneficiary expected and what they perceived — not a single aggregated score. The system must therefore hold two separate fields, stated expectation and measured perception, and a third calculated for the gap.
A fact many vendors overlook: these are guidance standards
All four standards are advisory. Three carry the word Guidelines in their titles, and the fourth is a set of principles and a model. They contain no auditable requirements against which a certificate of conformity can be issued, and no international accreditation exists for them as it does for ISO 9001. Anyone offering an accredited ISO 10002 certificate is offering a certificate issued by a private body against its own criteria, not an international accreditation.
This works in your favour. The absence of certification means the value of the system is measured by its operational effect alone — a shorter complaint closure time, a higher rate of meeting published service times, a narrower gap between expectation and perception — and not by a document on the wall. For that reason this framework is not marketed as an accredited integrated management system, but as an operational framework grounded in four reference standards.
When there is no certificate to conceal weak implementation, operational effect becomes the only evidence that the system works.
The textual basis that makes integration legitimate
No integration decision in this framework rests on inference. Each rests on a link stated in the body of the standards themselves. These are the principal documented links:
- ISO 23592 places feedback management, the substance of ISO 10002, inside the second level of its own service excellence pyramid. Complaints handling is therefore not a system running parallel to 23592, but a foundation layer within its own model.
- ISO 10001 states in its introduction that a code of conduct forms part of an effective approach to complaints handling: preventing the complaint through the code first, then internal resolution, then external dispute resolution.
- The relationship between 10001 and 10004 runs in both directions: measurement processes assess satisfaction with the codes, and the codes govern the measurement processes themselves, including confidentiality of beneficiary data.
- ISO 10002 and ISO 10004 intersect explicitly: measurement tools assess satisfaction with the complaints process itself, and the frequency and nature of complaints are an indirect indicator of satisfaction. This is the bridge that makes a single database a methodological requirement.
- Each of the three standards 10001, 10002 and 10004 contains a dedicated annex on the interrelationship between them, meaning integration is intended by the drafters, not assumed by the reader.
The architecture: one core and five layers
The first basis for the core-and-modules decision is the structural backbone shared by 10001, 10002 and 10004: each begins with context, then framework, then planning and design, then operation, then maintenance and improvement. That correspondence makes building them as separate systems an administrative duplication without justification.
Context, framework, resources, review and improvement are therefore built once in the core, and only operation remains specific to each module. ISO 23592 sits above that layer as a framework for capability and strategic direction, because it is a capability model rather than a process model.
From architecture to numbers: a weighted indicator system
An architecture that does not end in numbers a decision maker can read is of little value. The five layers are therefore translated into sixteen indicators across five strategic objectives, with weights summing to exactly one hundred per cent, so the entire system reduces to a single figure: weighted institutional performance.
The figures come from the live indicator system, running on data for the Riyada Governorate Municipality, an entirely fictitious organisation used for demonstration and training whose figures are attributed to no real body.
- Every indicator is grounded in a clause: rate of meeting published service times cites ISO 10001 clauses 6.5 and 8.2, while average expectation-perception gap cites ISO 10004 clause 4.2.
- Owner separated from measurer: the indicator owner is accountable for the result, while measurement is the responsibility of a different function, so no department measures and judges its own performance.
- Achievement is capped at 120 per cent with a floor of zero, preventing a single outlying indicator from distorting the weighted average of the whole system.
- Four status bands: underperforming below 70 per cent, needs attention up to 85, on target up to 99, and outstanding at 99 and above, each with a predetermined response.
- Negative-polarity indicators use an inverted formula, so a fall in average days to close a complaint reads as improvement rather than decline.
What this means in practice: 117 interlinked documents
The package is distributed across eleven modules: the core (12 documents), service excellence under 23592 (30), codes of conduct under 10001 (12), complaints under 10002 (17), satisfaction measurement under 10004 (15), shared processes (12), the data model (9), implementation and maturity (6), together with the project plan, the architecture and the indicator system. By format: 38 Word documents, 68 Excel workbooks and 11 PowerPoint decks.
All of it is governed by a single index: no file is created outside it, and no file is renamed without updating its row. That is the difference between a folder of documents and a managed system.
How to start in your own organisation
Start with the crosswalk, not the documents
Map your operating domains, then set against each one what every standard says about it. The crosswalk reveals which activities can be unified and which must stay separate, before a single line is written.
Unify terminology before anything else
Decide early that customer and beneficiary are one term across all your documents, and build a single glossary with a source for every term. Inconsistent terminology corrupts later reporting more than any methodological error.
Build the core once
Context, leadership, roles, principles, resources, confidentiality and document control: one document per subject serving all modules, not one document per standard.
Begin operation with the complaints module
It produces immediate data and shows early effect, and it simultaneously feeds both satisfaction measurement and the code of conduct.
Separate expectation from perception in the data model
Two mandatory separate fields and a third calculated for the gap. Merging them into one score destroys the concept on which ISO 10004 rests, and it is hard to undo once data collection has begun.
Never launch an indicator without a clause and an owner
Every indicator needs a clause that grounds it, an owner accountable for the result, a separate function responsible for measuring it, and a relative weight within one hundred per cent. An indicator lacking these four is a number for display, not for management.
Six files from the live system — free download
The files attached below are not illustrative templates. They are six documents from the package itself, under their approved codes:
- IMT-ISE-IM-PLN-000 — the master project plan: the crosswalk across twenty-one domains, the documented integration links, the shared structural backbone, the unified charter of principles, the coding system, and twelve architecture decisions each recorded with its rejected alternative and its basis.
- IMT-ISE-IM-RPT-000 — the analysis and architecture decisions report, in Word.
- IMT-ISE-CR-REG-001 — the master document index: the path, approved filename and status of every file, with a calculated production-status sheet.
- IMT-ISE-SH-REG-001 — the performance indicator system: a dynamic card, a properties register, a readings log and a weighted dashboard.
- IMT-ISE-SH-REG-002 — the improvement opportunities register: the instrument that closes the loop and converts measurement and complaint findings into tracked actions.
- IMT-ISE-CS-PRO-001 — the beneficiary satisfaction measurement procedure: ten executable steps from defining the scope of a measurement wave to assessing the effectiveness of the resulting actions, with a clause traceability matrix.
All names and figures belong to the Riyada Governorate Municipality, an entirely fictitious organisation created for demonstration and training.
These six files form part of the Customer Experience Pro complete package, the fully documented system of 117 Arabic documents covering all four standards. You can find it here: the complete package on the Imtithal store.
Customer Experience Pro — the complete package
The fully documented system: 117 Arabic documents covering all four standards — policies, procedures, codes of conduct, forms, records, matrices and indicator dashboards running on real formulas, with consistent demonstration data that shows the system working before you enter any of your own.
View the complete package