LorikeetFuse, Plugs, and a concierge across every channel.
What Lorikeet on Fuse looks like across the app, the web, voice, and outbound.
LorikeetWhat Fuse told us
Fuse today
Plugs handles most switching and billing. The gap is what comes next.
- 70–80% containment on switching and billing. Meters run lower.
- 70k inbound / month, with a bill-day spike of 5–6k in a single day.
- Support team of 30–50, scaling to 70–80 on bill days.
- Built on Google ADK / Gemini. Quality is hands-on: someone flags a bad reply, an engineer fixes it in code.
What Fuse wants to see next
Four capabilities on the roadmap that need engineering to build.
- Voice. On the roadmap. Not yet in the app.
- Proactive outbound. Signal-based reach-out. None today.
- Simulations, guardrails, quality scoring. None today.
- Long-running goal-based tickets. Change of address, complaints, backend malfunctions.
LorikeetWhat Lorikeet can deliver Fuse
Same seat inside Intercom, richer concierge underneath
Plugs stays as the brand. The team keeps its inbox. What changes is the agent underneath: authored in natural language by ops, guardrailed by default, covered by simulation.
Voice, simulations, guardrails, outbound
The four capabilities Fuse plans to build next. Available in the Lorikeet sandbox from day one, without adding engineering headcount.
Beyond containment
Plugs is at 70–80% containment today. The next layer is proactive reach-out on bill day, meter-read voice calls, and onboarding recovery.
LorikeetOne concierge, three surfaces
In the Fuse app · Plugs tab
Same entry point customers use today, contextual to the account (tariff, meter, EV, solar). Runs the switching and refund tools directly. Hands off to a human cleanly, with a full internal note.
On fuseenergy.com · widget & proactive nudge
Handles switching questions on the marketing site. Qualifies leads before signup. Answers "what's my next bill" for signed-in customers. Same concierge, different context.
Outbound · voice, email, in-app push
Bill-day preview. Meter-read reminders by voice for repeat non-responders. Onboarding recovery. Complaint follow-up. Triggered from Fuse's system of record, driven by Lorikeet.
LorikeetBill lands £34 higher. Plugs handles it in the app.
LorikeetPlugs qualifies and quotes
LorikeetProactive outbound
Bill-day preview
24 hours before bill day, Plugs reaches customers flagged by the context engine (unusual usage, first bill on a new tariff, big EV month) and explains the number before it lands.
Meter-read reminders
The meters queue has higher escalation because a missed input becomes a wrong bill. Plugs reaches out on schedule, walks the customer through submission, retries if it fails. Voice for repeat non-responders.
Onboarding recovery
The three-minute signup flow leaks people. Plugs reaches abandoned signups at the specific step they got stuck on and helps them finish.
LorikeetWhat sits behind the concierge
Simulations
Every workflow change runs against a bank of representative tickets before it ships. Regression coverage that Fuse would otherwise build from scratch.
Guardrails and message checks
Pre-processing catches vulnerable customers, sensitive intents, and PII. Post-processing catches unauthorised promises (refunds, discounts, policy statements) before the customer sees them.
Ops-owned NLW authoring
Workflows are natural-language documents in the web app, not code. Support and ops edit them directly. Engineers only get involved when new tools need building.
Long-running goals
Tickets that take five days (complaints, backend malfunctions, change of address) get a goal-oriented concierge. It holds the goal, re-engages after silence, and closes only when the goal is met.
LorikeetOps teams running Lorikeet in production
