×

Lorikeet Overview for Finway

The universal AI concierge for pre-accounting, built for the procure-to-pay stack.

Prepared for Adriana Palaioroutas · 20 August 2026
Alex Holder, Lorikeet

The Opportunity

Turn pre-accounting from reactive checking into proactive prevention. Cut admin approval time in half, catch errors before they reach accounting, and free your admins to focus on the exceptions worth their judgement.

Better admin experience

No more hunting for §14 UStG issues, vendor price sanity, or budget context. The agent surfaces every issue on the invoice at a glance, with the source data cited.

Proactive, not just reactive

Skonto windows, payment deadlines, budget overruns, and stalled approvals flagged before they cost you money. SMS, WhatsApp, email — the admin picks the channel.

Real efficiency

Admin approval time cut by 50%+. PDFs pre-fill into expense requests. Cost centre coding suggested from vendor rules and historical patterns. The admin reviews, doesn't retype.

Redeployed for growth

The admin and CX capacity you free up flows into PLG onboarding walkthroughs, expansion conversations with your paying customers, and the trial-to-paid conversion motion you've been under-resourcing.

State of Play at Finway

~450 customers, ~5 approvers each, ~100k invoices/month, with heavy spikes at month-end close; the admin and CX teams carry the weight of pre-accounting review
Fin (Intercom) covers Pillar 1 — support tickets — at ~40% resolution. Reasonable ceiling for FAQ, but it can't see the invoice, the budget, or the vendor history
Two personas, one platform: admins want error detection and autofill; basic users want a one-click PDF-to-expense. Same UI, different jobs

Three motions to automate end-to-end:

Admin invoice / PO review, run the §14 UStG compliance check, compare unit prices against six months of vendor history, propose cost centre 1/2 + expense account from historical patterns, flag budget overruns before the admin clicks approve
Basic-user expense creation, admin drops a PDF and the agent parses, proposes a full coding with reasoning, asks only for what it can't confidently derive, and submits
Proactive outbound, Skonto and payment-deadline warnings via SMS/WhatsApp with reply-handling, plus in-app onboarding walkthroughs (cost centres, invite employees, bank, DATEV) grounded in your help centre

How we can solve your problems

Four use cases built as a working sandbox, synthetic data, real workflows, real agent.

Inbound (admin)

Admin invoice review

Admin opens a pending PO or invoice and clicks "Ask AI". Agent runs a §14 UStG check on the attached document, compares line-item unit prices against six months of vendor history, proposes cost centre 1/2 + expense account from historical patterns, and flags the €3,272 PO against the €300 IT budget.

The highest-value admin request from your survey. Multi-source reasoning Fin cannot do.

Basic-user PDF-to-expense

Basic user drops an invoice PDF into the New Invoice modal. Agent parses vendor, dates, amounts, VAT; proposes cost centre 1/2 + expense account with reasoning (e.g. "11 of 12 Hotel Sonne invoices coded to Reisekosten / 6650"); submits on confirmation.

Turns a twelve-field form into a drop-and-glance flow for non-power-users.

Outbound & proactive

Skonto & deadline warnings

Agent SMSes the admin: "Skonto on HS-2026-04187 closes in 2 days, pay by 21.08 to save €5.71." Recipient replies "pay", "defer", or asks any question — the agent handles the conversation, schedules the payment or notes the missed discount.

Live demo on the next slide. Real phone, real SMS, real reply.

Onboarding walkthroughs

Trial admin lands on the dashboard. Agent greets them and walks them through the four setup steps they typically skip: creating cost centres, inviting the team, connecting the bank, and setting up DATEV — all grounded in your existing help-centre articles.

Fixes the "most trials stay solo and miss approval workflows" pain point Adriana called out.

Try the live agent yourself.

A mock Finway admin UI with the AI concierge embedded — same widget architecture Adobe, Paddle and Airwallex use. Two scenes and one proactive SMS, all live against a real Lorikeet agent.

Scene 1: Admin invoice review

Click "Ask AI to review" on the Muster Metallbau PO. Agent runs §14 UStG + price history + cost centre + budget in one turn.

Scene 2: PDF-to-expense

Switch to "Basic user — New expense" and drop the Hotel Sonne PDF. Agent extracts every field, proposes coding at 92% confidence, submits on confirm.

Proactive: Skonto SMS

Enter a phone number in the Skonto card, hit Send. Recipient gets a real SMS in ~5s and can reply naturally — "pay", "defer", or any question.

Password on the demo: Finway2026! · Both scenes run against a real Lorikeet workflow with 8 mock tools mimicking Finway's endpoints.

Open the live demo

finway.demo.lorikeetcx.ai

Password: Finway2026!

Appendix

Lorikeet overview, the team, investors, customer voice, architecture, POC plan, security, and pricing.

Lorikeet is building the leading AI concierge for complex businesses

One concierge, every channel. The agent takes actions in your systems, not just answers from a KB — the reason regulated fintechs, healthtechs, and platforms like Finway pick us over Fin, Sierra, Decagon.

Re: PO-20251001-0002 — Muster Metallbau, ready to approve?

Finway assistant

To: Steve Frost (Approver)

Hi Steve, quick review before you approve this PO. Two mandatory §14 UStG fields are missing on the invoice (Steuernummer + USt-IdNr.) — the vendor record has neither on file.

Also worth knowing: approving this puts the IT cost centre at 1091% of budget (€2,972.50 over). Suggest edit-before-approval.

Email

Lorikeet

0:08

"Finway here — Skonto on HS-2026-04187 (Hotel Sonne, €305.57) closes in 2 days. Pay by 21.08 to save €5.71. Reply 'pay' to authorise, 'defer' if cash is tight, or ask any question."

SMS & Voice

What cost centre should this hotel invoice go to?

S

Reisekosten / Hotel & Accommodation, expense account 6650. 92% confidence.

Why?

S

11 of 12 Hotel Sonne invoices in the last 12 months were coded exactly that way.

Chat

Our people have decades of experience with AI and building great products

CEO and Co-founder

Steve led product teams at Stripe and Watershed, building tooling to enable complex processes like carbon accounting and financial reporting at scale.

CTO and Co-founder

Jamie was a research tech lead at Google Brain, leading research on factual grounding in large language models. Third named author on Google's breakthrough 2022 LaMDA paper, and fourth named author on the 2020 predecessor Meena paper.

Our team comes from leading companies applying AI and driving large-scale corporate transformation

Lorikeet is backed by top global investors

We've raised over $50M from investors who backed Nubank, Klarna, Canva, Airwallex, Supabase, Midas, Clearscore and more.

Announcing our

$35M

Series A

Led by QED Investors

Blackbird

Square Peg

Capital49

Skip Capital

Airtree

Operator Partners

www.lorikeetcx.ai

Lorikeet consistently outperforms other vendors in the market

"We tested AI solutions head-to-head and Lorikeet was a winner in every metric."

Lindsay Boland

Product Manager, Flex

"We ran POCs with Fin AI, Decagon... Lorikeet was a clear winner."

Jiaona Zhang

Former CPO, Linktree

"Considered Sierra and other well-known players... [Lorikeet] could handle nuance, de-escalate emotional moments, and follow SOPs without breaking a sweat."

Jessica Mishlove

Head of Customer Relations, Arbor

Read case studies

Journey of a Support Ticket

How Lorikeet's AI agent resolves customer issues end-to-end

1

Customer sends message

Via chat widget, email, WhatsApp, voice, or ticketing system

2

Channel Intake

Mode (chat / email / voice / SMS) · Escalation rules · Auto-close config

3

Brand + Config Loaded

Finway voice · German & English · Admin vs basic-user persona

4

Inbound Guardrail

Evaluate message → pass to triage, or escalate immediately

5

Intent Classification

AI matches customer intent against all available workflows

Matched Workflow

Run response SOP

FAQ Fallback

Search knowledge base

No Match

Escalate to human

6

Workflow Executes

Structured decision graph or Natural Language agent follows step-by-step SOP

Tools

Finway API · Kontenrahmen · Vendor rules · Budgets

Knowledge

Help-centre articles · SOPs · §14 UStG rules

Actions

Invoice ops · Notes · Slack · DATEV export

7

Outbound Guardrail

Check AI response against policies → pass or escalate

8

Reply Sent to Customer

Brand voice · Formatting · Citations · Side effects (tags, CSAT, Slack)

If customer replies → loops back to step 5 with full context

If no reply → auto-resolved after idle period

Proof of concept

2–4 weeks, offline, free, to prove Lorikeet executes against real Finway invoices, admin queries, and expense flows
We'll mock data to simulate your APIs and Kontenrahmen, rather than integrate directly during the trial
Zero risk to you, if we don't meet your needs, no obligation to proceed
Sized for your end-of-October target, first launch of admin invoice review + Skonto alerts inside 6–8 weeks; wider rollout for onboarding + PLG in the following release

Train and test AI

Integration scoping

Legal / privacy / security review

Week 0–1

We will:

Train Lorikeet on Finway voice, tone, and §14 UStG rules
Run response testing over 20–50 real admin queries and invoices
Lock the anchor use cases (admin review + PDF-to-expense)

We need:

SOPs, tone of voice, sample invoices & tickets
Fast feedback from Adriana + a nominated admin

Week 1–2

We will:

Build the admin invoice review + basic-user PDF flow against Finway sandbox APIs
Wire the Skonto SMS outbound with a Finway-registered sender
Iterate on response quality and coverage

We need:

Finway PM + tech lead on the endpoints we need (vendor history, budget, master data)

Week 2–3

We will:

Layer in onboarding walkthroughs (cost centres, invite, bank, DATEV) grounded in help centre
Lorikeet 101 with your champion
Start security and legal review

We need:

Access to the Finway help centre for KB ingestion

Week 3–4

We will:

Complete security + GDPR review (EU data residency option confirmed)
Agree commercial terms & sign
Ship first release ahead of the end-of-October target

We need:

Sign-off from the Finway board on scope + commercials

Lorikeet is built securely for enterprise

Compliance-grade architecture, day one. Designed for fintechs, healthtechs, and regulated platforms handling invoice PII, banking data, and GDPR-scoped records.

SOC 2 Type II
& ISO 27001

Both certifications independently audited. Your data is always protected and is never used for training.

GCP + VPC

Built on Google Cloud, running in a virtual private cloud. Audit-grade logging end-to-end.

Ex-Stripe security

Founding engineering team includes an ex-Stripe Staff Security Engineer. Security reviewed by industry leaders.

Handles regulated financial data

Already deployed with FCA-regulated subscribers and payment platforms handling invoice PII, IBANs, tax IDs, and payroll data. EU data residency available.

For Finway's data posture: invoice PII, vendor tax IDs, IBAN/BIC, DATEV records, and customer accounting data all need strict handling under GDPR. Happy to walk your security and legal team through our controls during the pilot — including EU data residency, access controls, and incident response.

View trust center →

You only pay for resolutions

Aligned incentives by design, we both win when tickets are resolved.

Pay only for resolved tickets

Lorikeet charges per ticket successfully resolved, as determined by you. If we don't deliver, you don't pay.

No platform or per-seat fees

No setup costs, no per-seat licenses, no monthly minimums. The only line item is resolutions.

Volume discounts

Per-resolution price decreases as volume grows. Available across email, chat, voice, and SMS.

Aligned incentive structure

You buy credits from Lorikeet up front
Credits are only consumed when the AI reaches the desired outcome
You don't pay if the AI fails and the case is passed to a human
Credits sized to your forecast volume, buy top-up packs at a discount mid-contract, no surprise overages

Sizing for Finway: exact per-resolution price depends on the workflows we scope together during the pilot, sized to your admin-facing query volume plus outbound Skonto/onboarding usage. We can baseline a target range once we've agreed the anchor use cases and estimated monthly volume together.

Alex Holder

alex.holder@lorikeetcx.ai