Why classification decides everything downstream
Categorisation feels bureaucratic in the moment a complaint arrives — the customer is upset, the engineer wants to act, and stopping to pick a category and a priority feels like friction. But those two fields are the hinge on which the rest of the process turns. They decide where the complaint goes, how fast it must be handled, and whether, a year from now, you can see any pattern in your complaints at all. Skip them and you have a fast-feeling process that produces no management information; do them consistently and you get routing, SLA and analysis almost for free.
This guide is a practical companion to the broader complaint handling process: it zooms in on the second step — categorise and prioritise — and gives you a template you can adapt. The core idea is simple: a category answers what kind of problem is this? and a priority answers how urgently must we act? Everything else follows.
Categories — your vocabulary, not a vendor's
A complaint category classifies the complaint by type, and the single most important rule is that the list should be your organisation's own vocabulary. A machine builder thinks in field failures, installation issues and documentation errors; a process manufacturer cuts the list by defect mode and batch; a distributor thinks in delivery, damage and short-supply. A fixed vendor taxonomy forces your complaints into someone else's boxes, and mis-fit categories are worse than none because they look like data while being noise.
Categories do two concrete jobs. First, routing: warranty complaints go to the service team, billing disputes to accounts, quality defects to the engineer who owns that line — the category is often what drives the assignment. Second, analysis: the monthly Pareto of complaints by category, customer and product is where continuous improvement starts, and that chart only exists if complaints are consistently categorised. Get the list right and both happen automatically; get it wrong and you spend every review meeting arguing about what the numbers mean.
A starter category taxonomy for manufacturers
You should adapt this to your own products and failure modes, but a practical starting point for a manufacturer or equipment maker looks like this. Keep it short — ten well-chosen categories beat forty overlapping ones.
| Category | Typical complaints | Usually routed to |
|---|---|---|
| Product / quality defect | Off-spec part, dimensional issue, functional failure | Quality / production engineer |
| Warranty / field failure | Product failed in service within warranty | Service team → 8D |
| Installation / commissioning | Problem during install or start-up | Field-service engineer |
| Delivery / logistics | Wrong, short, late or damaged shipment | Dispatch / logistics |
| Packaging | Transit damage, inadequate packing | Packaging / dispatch |
| Documentation / labelling | Wrong manual, certificate or label | Quality / documentation |
| Service quality | After-sales visit unsatisfactory | Service manager |
| Commercial / billing | Invoice, pricing or payment dispute | Accounts |
Notice that several categories imply a path into root cause: a warranty or field failure almost always deserves an 8D investigation, while a documentation error may just need correction and a quick check. The category is your first signal of how deep the investigation should go.
Priority, severity and the SLA
Where category answers what, priority answers how urgently. It is worth separating two ideas that people often blur.
Severity is how bad the problem is in itself — a safety-related failure is high severity no matter who reported it. Priority is how urgently you will act, and it combines severity with business context: the customer's importance, contractual SLAs, and how many units are affected. A cosmetic issue at a key account may be prioritised above a moderate issue at a small one — not because it is more severe, but because the business consequence is larger. In practice severity informs priority, and priority sets the due date the team actually works to.
| Priority | When to use it | Typical SLA intent |
|---|---|---|
| High | Production-stopping, safety, or a key account affected | Acknowledge same day, resolve fastest |
| Medium | Real problem, limited units, workaround exists | Resolve within the standard window |
| Low | Cosmetic or minor, no operational impact | Scheduled, longest window |
The software turns priority into a concrete SLA due date on the ticket, drives a pending and ageing view sorted by how close each ticket is to breaching, and escalates anything overdue to the owner and then the supervisor. Complaints rarely fail loudly — they expire quietly in queues — and priority-driven ageing is what prevents that. The mechanics are covered under SLA & follow-up, and the whole lifecycle in how the software works.
Want your categories and priorities set up for you?
In a 30-minute demo we will map your complaint categories, priorities and SLA rules into Fast Complaint Software on screen.
Getting classification to stick
A category scheme only works if handlers actually use it consistently, and there are a few reliable ways to make that happen.
- Keep the list short. Every extra category is another chance for two handlers to file the same complaint differently. Start with eight to ten and add only when a real gap appears.
- Make category and priority mandatory. If a ticket can be saved without them, some will be — and those are exactly the ones that vanish from the analysis.
- Review the Pareto monthly. If one catch-all category swallows half your complaints, it needs splitting; if a category never gets used, retire it. The taxonomy is a living thing.
- Let AI help spot patterns. Clustering complaint remarks into named themes can reveal categories you did not think to create — see Dhruv AI.
Categories and priority in Fast Complaint Software
Fast Complaint Software, built by Improsys in Pune, is built to run on your own vocabulary. Feedback and complaint categories are maintained as master data you define, so the list is yours rather than a fixed vendor set. Priority is selected on each ticket from a configurable priority master, and together category and priority set the SLA due date that drives the pending-ticket schedule, ageing views and overdue escalation. Because every ticket is tagged by category and linked to a customer and product, the dashboards and MIS slice complaints by category, by customer and by product for the Pareto trend analysis that turns classification into improvement. For the full picture, start from the complaint management pillar.
Frequently asked questions
What are complaint categories?
Complaint categories are the classification you apply to each complaint by type — for example quality defect, warranty, delivery, documentation or billing. They should be your own organisation's vocabulary rather than a vendor's fixed list. Categories do two jobs: they route the complaint to the right team, and they turn a pile of individual grievances into analysable data, so you can see the monthly Pareto of complaints by category, customer and product and aim improvement effort where it counts.
How do complaint categories and priorities affect the SLA?
Category and priority together set the SLA — the due date by which a complaint must be acknowledged and resolved. Priority ranks urgency (for example High, Medium, Low), and category often implies where the complaint goes and how serious it typically is. The software converts that into a due date on the ticket, drives a pending/ageing view sorted by how close each ticket is to breaching, and escalates anything overdue. Without category and priority, every complaint looks equally urgent, which means none of them really are.
What complaint categories should a manufacturer use?
A practical manufacturing starter set is: product/quality defect, warranty or field failure, installation or commissioning issue, delivery or logistics (wrong, short or damaged), documentation or labelling error, and commercial or billing dispute. Add a packaging category if you ship fragile goods, and a service-quality category for after-sales. Keep the list short enough that handlers classify consistently — ten well-chosen categories beat forty overlapping ones, because analysis depends on complaints landing in the same bucket every time.
What is the difference between priority and severity?
Severity describes how bad the problem is in itself — a safety-related field failure is high severity regardless of who reported it. Priority describes how urgently you will act, which combines severity with business context such as the customer's importance, contractual SLAs and how many units are affected. A cosmetic issue at a key account may be prioritised above a moderate issue at a small one. In practice severity informs priority, and priority sets the due date the team works to.
How does categorisation help complaint analysis?
Because tickets are tagged by category and linked to a customer and product, the software can produce Pareto analysis — complaints by category, by customer and by product — which is where continuous improvement starts. Unclassified complaints in an inbox tell you nothing, and the same defect can surface ten times without anyone noticing it is the same defect. Consistent categories are what let you see that one product line or one supplier generates a disproportionate share of trouble and fix it at source.