Independent IFS Cloud practice · Tailoring strategy
IFS Cloud customization strategy: build each requirement at the lowest tailoring level that meets it, and write down why
An IFS Cloud customization strategy is a rule for choosing, per requirement, between configuration, extending on the outside through APIs and extending on the inside with code. IFS documents these three tailoring types, and each one costs something different at a release update. The strategy picks the cheapest type that still meets the requirement and records the choice before anything is built.
Key takeaways
- IFS Cloud has three tailoring types: configuration in designers, extending on the outside through REST APIs, and extending on the inside in IFS Developer Studio.
- The tailoring level of the solution (core, configured, customized) decides the release update experience: automated, impact analysis, or developer work with assisted merging.
- Configuration is not free at updates. IFS warns that configurations can be affected by future releases and that it is tempting to overuse them.
- Two IFS rules prevent costly mistakes: a customization never depends on a configuration, and
@overtakeis a last resort after@override. - A CRIM describes the business outcome; the tailoring items that deliver it, with their level and owner, belong in a register that is checked before every release update.
1.What are the three ways to tailor IFS Cloud?
IFS Cloud can be tailored by configuration, by extending on the outside and by extending on the inside, and the IFS Tailoring Overview describes each with its own tools and its own effect on updates (Tailoring Overview, docs.ifs.com).
- Configuration: IFS defines it as “the use consultant and power-user friendly designers and tooling to extend the data model”, together with changes to pages and process automation. Page Designer and custom attributes belong here, as do custom events, projection configurations with custom actions, and the Lobby and Workflow Designers (Configuration, docs.ifs.com).
- Extend on the outside: IFS describes it as “the use of IFS Cloud open RESTFul APIs to integrate with” other applications. The code lives outside the customer solution repository. IFS advises: “Use Premium APIs where available as these are better documented and more stable over time.”
- Extend on the inside (customization): new or changed business logic built with IFS Developer Studio. IFS calls it “a powerful capability” and the only way to change some standard behaviour, and it is the most invasive of the three.
The distinction is described at the level of single objects, with a worked example, in IFS Cloud tailoring levels. This article covers the decision rule and the governance around it.
2.What does each tailoring level cost at a release update?
The tailoring level of a solution decides how much work each IFS Cloud release update brings, and IFS describes three levels with a different update experience for each.
| Level | What it allows | Release update experience, as IFS describes it |
|---|---|---|
| Core | Personalization and navigation-related configuration only | “Automated using IFS Lifecycle Experience” |
| Configured | All configuration capabilities, plus extending on the outside with any API | “Impact analysis supported by IFS Lifecycle Experience”, with assisted updates of impacted configurations |
| Customized | Adds extending on the inside with the full developer toolset | “Developer Guidance on how to address occasional changes to Platform”, with assisted merging of overridden and added functionality |
The configured level is where expectations go wrong. IFS writes that it comes with a “higher risk that some configurations could be impacted when applying future releases”, and warns that it is “tempting to 'overuse' configurations” because they look easier or cheaper. A heavily configured page is cheaper to carry than custom code, but it is not zero work at the next release. The same page also notes that smaller customizations can be relatively easy to carry through updates, while large overrides and overtakes take more effort. The trade-off is argued from the cost side in the Clean Core illusion in IFS Cloud.
3.How do CRIMs relate to tailoring types?
A CRIM describes the business outcome a project needs, while a tailoring type describes how that outcome is built, and IFS notes that one CRIM often combines several tailoring items. A supplier approval CRIM, for example, can consist of a custom attribute for the approval status, a Page Designer change that shows it, a validation workflow that blocks purchase orders to unapproved suppliers and a permission set for the approvers.
For governance this means the CRIM list alone is not enough. Each CRIM needs its tailoring items listed with their type and level, because the release update impact is assessed per item, not per CRIM. Testing works the same way: the CRIM is validated as a business outcome, the items are regression-tested one by one, as described in IFS testing: CRIM validation, ATT and cutover rehearsals.
4.In what order should a requirement be tested against the options?
A requirement should be tested against the options from the cheapest to the most invasive, stopping at the first one that meets it. IFS puts the principle in one line: “you also want to keep the tailoring level as low as possible”, without forcing a requirement into a type it does not fit.
- Standard process. Can the business change its way of working to fit standard IFS Cloud? This question belongs in the fit-to-standard workshop, with a named decision owner.
- Personalization and page configuration. Hiding fields, making them mandatory, moving them, relabelling them: Page Designer, published per configuration context.
- Data and automation configuration. Custom attributes and entities for missing data; custom events and workflows for automation; projection configurations with custom actions for new operations.
- Extend on the outside. Logic that belongs to another system, or that runs better outside the ERP, calls IFS Cloud through its APIs, preferably Premium APIs.
- Extend on the inside. Business logic that cannot be met otherwise is built in Developer Studio. IFS adds two rules here: “A customization should never have dependencies on configuration”, and “Use @overtake as a last resort only”, with
@overridepreferred.
Breaking the first rule looks like this: custom code that reads a custom attribute created in a designer. The attribute can be changed or removed by a key user without a developer build, and the custom code can then fail at runtime. Finding such failures afterwards is covered in silent custom code failure in IFS Cloud. What keeps extensions safe at updates, from the developer's side, is described in update-safe IFS Cloud extensions.
5.How do you keep tailoring under control after it is built?
Tailoring stays under control when every item has an entry in a register and every configuration travels in a package. The register is a project artefact outside IFS Cloud; the package is the IFS mechanism for moving configurations.
| Register field | Why it is there |
|---|---|
| CRIM and item ID | Links the business outcome to each object that delivers it |
| Tailoring type and level | Tells the release update check how much analysis the item needs |
| Owner | The person who decides whether the item is still needed |
| Reason and options rejected | Stops the item from being rebuilt differently by the next team |
| Package or repository location | Where the item lives: the Application Configuration Package, the customer solution repository or the external system |
| Review event | After each release update, check whether new standard functionality replaces the item |
On packages, IFS states that “configurations are imported and exported using Application Configuration Packages” (Configuration Lifecycle Management, docs.ifs.com). One detail is worth knowing before the first clean-up: making an item obsolete in a delivery does not delete it from environments where it is already installed, so it has to be removed there manually. The reason column follows the format of the IFS Cloud decision register, and the check before each release is described in IFS Cloud upgrade readiness.
6.Which risks does a customization strategy have to contain?
A customization strategy has to contain four risks, and each has a concrete counter-measure that can be agreed before build.
| Risk | Effect | Counter-measure |
|---|---|---|
| Scope creep | Every workshop adds tailoring items until budget and plan no longer hold | Change control: no item is built without a register entry and an accepted decision |
| Slow pages and processes | Workflows with loops and many API calls run inside the user's action | Performance review of workflows and custom code before deployment, with test data of production size |
| Orphaned configuration | Nobody knows why an attribute or event exists, so nobody dares to remove it | Owner and reason per item; review event after each release update |
| Users working around the system | Tailoring built for one group is ignored by another | Key users from each affected group in the decision and in UAT |
7.Frequently asked questions
What is the difference between configuration and customization in IFS Cloud?
Configuration in IFS Cloud uses designers and tooling for consultants and power users, such as Page Designer or the Workflow Designer, and is handled as part of the customer solution. Customization, which IFS calls extending on the inside, builds or changes business logic in IFS Developer Studio. Customization is the only way to change some standard behaviour and is the most invasive tailoring type.
Are configurations update-safe in IFS Cloud?
Configurations in IFS Cloud are cheaper to carry through release updates than custom code, but they are not exempt. IFS describes the configured tailoring level as carrying a higher risk that some configurations are impacted by future releases, with impact analysis supported by IFS Lifecycle Experience. Each configuration therefore needs an owner and a place on the checklist reviewed before every release update.
What is a CRIM in an IFS Cloud project?
A CRIM in an IFS Cloud project is an item on the list of configurations, reports, integrations and modifications that the project adds to standard IFS Cloud. A CRIM describes the business outcome, and IFS notes that one CRIM often combines several tailoring items, for example a custom attribute, a page configuration and a workflow. Release update impact is assessed per tailoring item.
When should we extend IFS Cloud on the outside instead of customizing it?
Extending IFS Cloud on the outside fits when the logic belongs to another system or can run outside the ERP and needs only to read or write IFS Cloud data through its REST APIs. The code then lives outside the customer solution repository. IFS recommends Premium APIs where available, because they are better documented and more stable over time.
How do we stop an IFS Cloud solution from accumulating tailoring nobody understands?
An IFS Cloud solution stays understandable when every tailoring item has a register entry that names its CRIM and its tailoring level. The entry also records an owner and a reason, plus a review event, and every configuration is kept in an Application Configuration Package. After each release update the owners check whether new standard functionality can replace their items.
8.About the author
Dariusz Myśliwiec: 18 years hands-on with IFS (Apps 7.5 to 10 and IFS Cloud). IFS Certified Associate Consultant. PRINCE2® 7. Independent practice based in Kraków, delivered remotely across Europe and globally. You talk to the consultant who does the work, not a sales layer.
Reviewing a CRIM list and deciding the tailoring level per item is part of my services for IFS Cloud. The outcome can be to keep an item, rebuild it at a lower level or drop it in favour of standard.
Selected clients from my IFS work: Fugro · LGC · BVI Medical · Betafence (PRÆSIDIAD) · Barlinek · NGK Ceramics · Newag · Oleofarm.
IFS is a registered trademark of IFS AB; this practice is not affiliated with IFS AB.
Do you know the tailoring level of every CRIM you run?
Send the CRIM list or the gap list from your project. On a technical call we take a sample of items and check the tailoring level and owner of each, and the cost each one adds to the next release update.
