VIENJA
Get a quoteFree 30-min call
Back to case studies
AI ProjectsAn AI product built by iPlus

Orderly — AI Agent for Purchase Order Entry

Orderly reads purchase orders from fax, email and PDF, finds the right part numbers and creates the order in kintone. Unusual orders wait for a person.

Book a 30-minute callSee inside the product

On sample data · measured again before and after go-live

order entry per day
1.2 h−78%
Before: 5.5 h
to confirm an order
25 min−90%
Before: 4 h
data error rate
0.3%−1.5 pts
Before: 1.8%
How it works

How Orderly works

The problem

  1. Customers send 40–80 purchase orders a day by fax, email and PDF, each written their own way.
  2. A customer writes “Brg 6204-2RS” but the system needs part BRG-6204, so two staff look up every line and key in the order by hand: about 5.5 hours a day.
  3. So customers wait hours for an order confirmation: price, stock and delivery date.
  4. One wrong part number ships a wrong lot, with returns and lost trust.

What it does

  1. 01Orderly reads each fax, PDF or email and pulls out every item, quantity and the price the customer wrote.
  2. 02It finds the right part in your kintone catalogue: “Brg 6204-2RS” → BRG-6204.
  3. 03It checks the customer’s price against the quote you sent. With no price on the order, it takes the customer’s price list.
  4. 04An unknown part, or a price more than 10% off (an old quote, a typo), holds the order with the reason.
  5. 05Normal orders go straight into kintone; for a held order, staff read the reason and click Approve.
Real product, sample data

Inside Orderly

01/10

A faxed order is read and matched 3 of 3; the price is 12% over the old quote, so Orderly holds it for review.

Rollout

Putting it to work

Rollout with iPlus

  • Weeks 1–2 · AssessmentReview 200–300 real orders, measure today’s numbers and set up access to the order channels and kintone.
  • Weeks 3–8 · BuildOrder reading, part matching and hold rules; run alongside manual entry, then hand over.
  • Monthly · AI AdvisoryOptional, at least 3 months, to tune thresholds and add new customer wordings.
  • You provideThe catalogue, the price list, a few sample orders and one reviewer.

Running cost

  • SetupOne-time, fixed price, quoted after a 30-minute call.
  • AI APIUnder USD 100 a month for about 1,700 orders, billed to your own account with the provider.
  • HostingYour own cloud or a small server.

Where the AI gets its information

  • SourcesYour own catalogue, price list, past quotes and orders, synced from kintone.
  • EvidenceEvery result shows the quote number, its date and who issued it.
  • No guessingWith no source, Orderly stops and hands over to a named person.
  • No trainingOrder data is never used to train models.
Built withClaudeOCRMISA APIkintone APIZalo OANode.jsPostgreSQL

Frequently asked questions

Short answers to what companies usually ask before they start.

Ask anything else in 30 minutes
Can Orderly misread an order, and what then?
It can, so every field carries a confidence level. Below the threshold the order stops for a reviewer and never reaches kintone.
Does our staff need training?
About half a day. The main steps are to read the reason, then click Approve, Return or Edit.
Will it work with our software?
Usually, if it has an API or bulk import. iPlus checks this during the 2-week assessment.
What if our system has no API?
Orderly writes a file in your system’s bulk-import format, and staff import the whole batch in one step.
Is our data safe?
Your data stays in your own systems and accounts; iPlus keeps none. Order data is never used to train models.
What does it cost each month?
API usage paid directly to the provider, hosting and optional AI Advisory. At about 1,700 orders a month, the API is estimated under USD 100.
How long until it runs?
The AI Workflow Assessment takes 2 weeks, then Workflow Automation takes 4–6 weeks.
What happens if we stop using it?
The catalogue and orders stay in your own system, so you can return to manual entry at any time.
Can we start small?
Yes. Start with one channel and one group of customers, then widen it.

Related projects

The office floor: seven AI agents work beside staff, and every decision that matters goes to the approval desk
AI Projects

Atrium — An AI Agent Office That Works Beside Your Staff

Atrium is an AI agent office: seven agents sort email and WhatsApp, draft quotes, enter orders and chase receivables, while people approve every step that matters.

Luma answers a braces price from the 2026 price list and books Saturday in the chat, with no front-desk work.
AI Projects

Luma Care — Patient Care Chatbot for Dental Clinics

A chatbot that answers patients 24/7 on WhatsApp, LINE and the website, books on the real dentist calendar and hands clinical cases to a dentist. Every answer shows its source.

An employee asks about per diem: the answer is worked out step by step, with the exact clause beside it.
AI Projects

Marginalia — Policy Answers with the Clause Cited

Employees ask about leave, per diem or security rules in Slack or Teams and get an answer with the exact clause. When the rules are silent, it goes to HR, never a guess.