Score suppliers on total cost, delivery and quality together, agree the weights first, then move each weight by 10 points. If the winner changes, it is a close call for a person. Simulated suppliers.
13
What is Heizen Studio?
A workspace that carries a software project from the first client conversation to the final release: requirements, stories, tests and the context behind them in one place.
AI made coding faster. Studio is built to get the requirement right first.
The cheapest supplier won, then the line stopped twice. Score cost, delivery and quality together, fix the weights first, and test the ranking.
24
Exact rules catch obvious copies. AI reads descriptions for meaning and checks supplier, quantity, date and approval limits to catch near-duplicates, then a buyer decides. Simulated requests.
20
Same pallet wrap, two plants, both approved. Exact rules catch copies. Meaning and context catch near-duplicates. A buyer decides before approval.
27
Three questions: is the workflow the same everywhere (buy), do the data and approvals already live in SAP or Coupa (extend), or is it specific to you and across systems (build a thin layer)? SAP stays the record.
1
34
Buy what is standard. Extend what already lives in SAP or Coupa. Build only the narrow piece in between, and keep SAP as the system of record.
30
A blank quality score was counted as perfect, so the wrong supplier ranked first. Put every field on one scale, keep blanks honest, show a confidence level, and send anything uncertain to a buyer. Simulated scorecard.
25
A blank quality score counted as perfect put the wrong supplier first. Normalize every field, keep blanks as unknown, and show the confidence next to each score.
26
AI tracks who owes a quote, classifies every incoming email, sends a polite reminder and escalates to the buyer before the deadline slips. Track, classify, remind, escalate. Simulated inbox. #procurement #supplychain #AI
30
A quote was due Friday. It is Tuesday. Nobody noticed. Track every reply that is due, classify what arrives, remind politely, escalate in time. (Simulated inbox, illustrative data.)
18
Here’s to new beginnings, bigger dreams and brighter days ahead 🪔💫 Happy Diwali from all of us at Heizen! 🎆
1
48
AI pulls the contract, past prices and your volume and drafts the reply. People set the ceiling and rules first, and nothing reaches the supplier until a person approves. Every step is logged. Simulated example. #procurement #supplychain #AI
25
Can AI negotiate with suppliers? It can prepare, draft and follow up. It should never commit on its own. Put an approval gate before every supplier message. (Simulated example.)
14
SAP holds the data, email talks to suppliers, Excel compares quotes. A thin AI layer closes the gaps between them, reads three things in SAP, writes one, and hands anything unclear to a person. Simulated workflow. #procurement #supplychain #AI
30
You do not need to replace SAP to automate RFQs. Put a thin AI layer in the gaps between SAP, email and Excel. It reads three things in SAP and writes one. (Simulated workflow, illustrative data.)
22
Five suppliers reply to one RFQ in five formats. AI extracts every field, converts units and currencies, adds freight and compares landed cost, then a buyer makes the call. Simulated workflow, illustrative data. #procurement #supplychain #AI
28
Five supplier quotes, five formats. The lowest unit price came third once freight was added (simulated example). That is why bid normalization matters.
1
30
Walmart fines suppliers for showing up early. 3% of the cost of goods, same as late.
On time doesn't mean fast. It means exactly when you said. Does your scorecard count early as a miss?
26
