Technical support & field dispatch
Move from reported symptom to the right field visit, first time
A wasted field visit is the most expensive outcome in after-sales service, and it usually starts with a vague symptom description and a manual nobody had time to open. This blueprint grounds diagnosis in model- and revision-matched documentation, confirms entitlement before work is promised, and sends the engineer with the right parts and a complete brief — while safety-critical symptoms stop guided troubleshooting on the spot.
- The team
- 4 AI worker roles escalating into 1 human decision owner.
- The process
- 6 orchestrated stages, each with a named decision owner.
- The channels
- Voice, Web chat, Email, WhatsApp on one shared context.
The shift
What breaks today, and what changes
The challenge
- Customers and distributors describe faults imprecisely, so the wrong engineer or the wrong part is dispatched.
- Technical knowledge sits across manuals, revisions, and past cases that nobody can search under time pressure.
- Entitlement and warranty status are checked late, after expectations have already been set.
What changes
- Diagnosis is grounded in documentation matched to the exact model and revision in front of the customer.
- Entitlement and warranty position are established before any work or timeline is promised.
- Engineers arrive with the right parts and a structured brief drawn from the diagnostic conversation.
The team
Every seat has one job and a defined level of autonomy
This is what gets configured in Team Canvas: role-scoped AI workers, the humans they escalate into, and a trust level on each connection that sets what a worker may do on its own. Trust starts tight and widens on evidence.
AI workers
Technical intake worker
Identifies the exact asset by serial and revision, and captures the symptom precisely.
- Owns
- Asset identification and symptom capture
- Trust level
- Acts within policy — read-only against asset records
Diagnostic worker
Suggests approved troubleshooting steps from documentation matched to that asset.
- Owns
- Guided diagnosis
- Trust level
- Acts within policy — approved procedures only, stops on safety-critical symptoms
Parts & entitlement worker
Confirms warranty position and checks compatible part availability.
- Owns
- Entitlement and parts readiness
- Trust level
- Acts with approval — reserves parts, never approves warranty claims
Dispatch coordinator worker
Matches skill, region, and urgency, and assembles the engineer brief.
- Owns
- Dispatch recommendation
- Trust level
- Suggest only — dispatch is confirmed by a person
Human decision owners
Service manager
Owns warranty decisions, dispatch confirmation, and safety judgement.
Owns: Warranty and dispatch authority
The process
The conversation starts the work. The workflow finishes it.
Each stage records the decision that was made and the role accountable for it — which is what makes the run reviewable afterwards rather than a black box. Where the platform executes a stage rather than a person or agent performing it, that is named too, and accountability still sits with the role.
- 1
Identify the asset
Which serial, model, and revision — and is entitlement valid?
Technical intake worker
- 2
Assess safety
Is this symptom safe to troubleshoot conversationally at all?
Diagnostic worker
- 3
Guide diagnosis
Which approved procedure for this revision narrows the fault?
Diagnostic worker
- 4
Plan resolution
Resolved remotely, parts shipment, or field visit?
Parts & entitlement worker
- 5
Dispatch
Which engineer, which parts, and is the brief complete?
Service manager
- 6
Learn
Does this outcome change the approved guidance for this fault class?
Service manager
Run by workflow
Decision rules
The non-negotiables encoded in the workflow, not left to a prompt.
- Safety-critical symptoms stop guided troubleshooting immediately and route to a qualified person.
- Only documentation matched to the asset’s model and revision may be used — never a near match.
- Warranty and goodwill decisions are made by authorised staff, never by a worker.
- Dispatch is always a recommendation confirmed by a person before an engineer is committed.
Controls & guardrails
What makes this safe to run in production, and provable afterwards.
- Asset-matched knowledge with no cross-revision guidance.
- Hard safety cutoffs that end conversational troubleshooting.
- Entitlement verification before any commitment is made.
- Human confirmation on every dispatch and warranty decision.
Platform capabilities
What this blueprint runs on
Nothing here is bespoke. Each blueprint is a configuration of the same platform, which is why the second one you launch reuses the governance you already reviewed.
Channels
- Voice
- Web chat
Systems it reads and writes
- Asset & serial registry
- Technical documentation library
- Parts inventory
- Field service management
Measurement
What we instrument from day one
These are the measurements the pilot puts in place, not benchmark results. You set the targets against your own baseline — and the same instrumentation is what decides whether a worker's trust level widens.
- First-visit fix rate
- The metric that governs after-sales cost: how often one visit resolves the fault, with no return trip.
- Diagnostic grounding
- Confirms each recommended step traces to documentation for the correct model and revision.
- Parts accuracy
- Tracks how often the part identified during the conversation matches the part actually needed on site.
Dispatch effectiveness
Source-matched answers
Right part, first time
Rollout
How this one goes live
A deliberately narrow start, supervised, with autonomy widened per role once the record supports it.
Weeks 1–3
One product family, triage only
Identify assets and capture symptoms for a single product family. No guidance yet — this stage proves the asset match is reliable.
Weeks 4–6
Approved guided diagnosis
Enable the approved procedure set for that family, with safety cutoffs and entitlement checks live from the start.
Week 7+
Parts and dispatch briefs
Add parts checks and dispatch recommendations, and measure first-visit fix rate against the pre-blueprint baseline.
Questions we get asked
Technical support & field dispatch, in practice
- How do you stop it giving guidance from the wrong manual?
- The asset is identified by serial, model, and revision first, and retrieval is scoped to documentation for that exact revision. A near match is treated as no match — the worker escalates rather than offering guidance that might apply to a different build.
- What happens with a safety-critical fault?
- Guided troubleshooting stops on the turn the symptom is recognised, and the conversation routes to a qualified person. Safety cutoffs are enabled from the first day of the pilot, before any diagnostic capability is switched on.
- Can it approve a warranty claim?
- No. The worker establishes entitlement position from your registry and can reserve parts, but warranty and goodwill decisions remain with authorised staff, and dispatch is confirmed by a person before an engineer is committed.
What changes in Manufacturing
Diagnosis is grounded in model- and revision-matched documentation, and safety-critical symptoms stop guided troubleshooting.
Read the full Manufacturing analysisRelated blueprints
Teams that run next to this one
Next step
Build this team around one measurable win.
We map your version of this process, compose the worker roles, connect your systems, and agree the escalation points with the people who own them — then run it supervised.