Back to AI Web Development

Service firms & B2B · 04 / 13

Customer portal

A place customers log into to check job status, invoices and documents themselves, pulled straight from your back-office systems.

A client at a café table logs into her portal: a welcome card, an invoice with a PDF download button and project progress. Service firms & B2B
“Customers call every day for job status, invoices and the same documents. The team answers repeat questions instead of doing the work.”

The problem

Everything customers ask for already sits in your systems. Job status is in the ERP, invoices in the accounting system, contracts in a shared folder. Customers have no way in, so they phone or LINE their account contact, who opens three systems to answer one question, dozens of times a day.

When that contact is on leave or leaves for good, customers have nobody to ask. Documents get re-sent by email until nobody can find the latest version, and chasing payment turns into a phone job nobody wants.

How we solve it

We build a portal customers log into with email or LINE and see only their own data: job status, due dates, invoices and documents, read live from the ERP and accounting system you already run, with nothing re-keyed. When a status changes in your system the customer sees it at once and gets a LINE message.

AI chat inside the portal answers from that customer’s records only: where the job is, what is outstanding, where the March receipt went, and finds old documents in seconds. A question the data cannot answer, such as moving a due date or changing terms, opens a request to your team with the full context.

Decisions stay with your people. A change of terms, a discount or a refund is approved by someone in the system every time. The system makes sure the request arrives complete and keeps a record of who answered what and when.

How it runs

Work comes in from
  • Job status in the ERP
  • Invoices in accounting
  • Documents in the shared folder
  • Customer questions
What the AI does
  1. Gather that customer’s own data
  2. Answer from the real records
  3. Find documents
  4. Open a request when a person is needed
Where it lands
  • A portal customers log into
  • Alerts on LINE
  • Payments into accounting
  • Team queue with context

Before and after

Before
After
Customer calls for status, the team opens three systems to answer
Customers check the portal and get an alert when status changes
Invoice sent by email, lost, requested again
Every document in the portal, paid on the same page
The account contact is away, the customer has nobody to ask
AI chat answers from real data round the clock; the team takes only the human cases

What you get

  1. 01

    A login for each customer showing job status, due dates and full history, pulled live from your ERP or job system

  2. 02

    Invoices, receipts and contracts to download, paid by PromptPay or card on the same page, with the payment posted to accounting automatically

  3. 03

    AI chat that answers from that customer’s own records: where the job is, what is outstanding, and finds old documents

  4. 04

    A LINE or email alert when a status changes, an invoice is issued or a due date is close

  5. 05

    A request form for issues and service that queues to your team with the history attached, and a dashboard of what customers ask most

Who gets what

Business owner

Cash comes in sooner because customers see the balance and pay themselves, and you see what customers ask about most, which tells you where the work needs fixing.

IT director

Read-only API access to the ERP and accounting system, permissions per customer, an audit log of every access, deployable in your cloud or on-premise, and PDPA compliant.

The team using it every day

The account team stops answering the same questions dozens of times a day and keeps only the cases that need thought and a decision.

Who this fits

Accounting & law firmsFreight forwarders & shipping agentsContractors & interior fit-outAgencies & studiosSuppliers selling to businessesInsurers & brokers

Connects with what you already run

SAPOdooXeroFlowAccountPromptPayLINE OAMicrosoft 365

Development process

  1. 1

    Discover

    Requirements, users and success metrics, with scope and price fixed before we start.

  2. 2

    Design

    UX and system architecture; the prototype is approved before anything is built.

  3. 3

    Build

    AI-accelerated sprints with a demo every week, reviewed by senior engineers.

  4. 4

    Test

    QA, security and performance verified against the agreed scope.

  5. 5

    Launch & care

    Production deploy, team training, and a monthly care plan.

Turn your business problem into a system that works for you

Tell us today — get an executive-ready proposal with the plan and budget.

Talk to an engineer about this