ai-agents

Fifteen store automations scored as traditional vs agent — not 15 week-one builds. Where-is-my-order is about 18% of tickets. Pick one read-heavy workflow this week.

Shopper asks for a deal at checkout. Classify no-discount, bundle, small percent, or human — do not mint codes. Baymard abandonment is 70.22%, not a coupon ROI.

Slack says Sales down 15% with no next step. The 15% is a schema fixture. Ship metric, contributors, unknowns, and ranked actions. Baymard 70.22% is not your cause.

Tuesday buyer lands in Winback on Wednesday. Name who is at risk, why, and what action — a person still merges the list. Cart abandonment is 70.22%; that is not a churn model.

Monday, the same order and policy tickets. An agent looks them up. Where-is-my-order is about 18% of the pile. Refunds stay with a person. We are not publishing a tickets-down number.

An associate asks "can we ship this SKU on this order?" and jumps five UIs. Six named read tools, evidence on every fact, row cap 25 — not a warehouse chatbot.

Price, stock, and customer records still drift after publish. Run 6 checks in a worker. The model investigates only if you add it.

Monday, the planning table. An agent explains the number a stats model already owns. Sample horizon is 28 days. A buyer still signs the PO.

Someone logs into Looker at 8:12 because something felt off. Pick one signal domain. Detect in code. Cap the live page list. The daily brief stays at five — this is always-on.

Before ads spend on a stockout, the morning brief names 5 priorities with proof on each. Purchase orders wait for a person. We are not publishing hours saved.

Forty zaps, one chatbot per OMS code, two refunds on one order. One investigator loop this week. Tools per class. Investigate and recommend. The ~$791/mo platform cost floor at 50,000 sessions is planning cost, not holds closed.

Monday, the aisle and the open POs. An agent ranks reorder, wait, stockout, or excess. Sample cover at 5 days is a fixture. A buyer still sends the PO.