01
Read the context
Elasticity, competitor prices, stock position, sell-through pace, margin floor — all pulled from the retail graph in real time.
Kwanta Pricing
Every price change, scored before it goes live.
Regular prices, markdowns, and promo prices — each one checked against demand, competitor prices, your floor, your ceiling, MAP, and margin before it ships. The AI drafts the move using real numbers from every store. Your team approves — or lets the rules run it automatically.
- Elasticity per SKU × store × channel
- Guardrails: floor, ceiling, MAP, margin
- Markdown cadence tuned to sell-through
How it works
One workflow replaces the weekly pricing meeting — the agent does the prep, your team does the thinking.
02
Propose the move
The agent drafts specific price changes per SKU × store. Each one comes with a score, the expected margin delta, and a short written rationale.
03
Approve on rails
Category lead approves, overrides, or batch-approves inside the guardrails you set. Every decision is logged and reversible.
04
Ship and measure
Prices ship to the till. The agent measures the lift trailing 7 / 14 / 28 days and feeds the learning back into elasticity.
What it covers
List pricing
Regular price, markdown, premium tier — with elasticity models per category and per store cluster.
Markdown cadence
Planned sell-through target per collection or season. The agent proposes when and how deep, with margin trade-offs.
Promo pricing
Deal depth, mechanic, and duration — scored against incrementality before it reaches the calendar.
Competitor signal
Pulled from the marketplace and web sources you authorize — surfaced as a soft input, not an autopilot rule.
Guardrails
Floor, ceiling, MAP, margin %, competitor match policy — set per category, per store, per partner.
Audit trail
Every price change, every approval, every override — logged with the reasoning and reversible on demand.
Works with
Pricing runs on top of Kwanta POS, or reads legacy price-book exports while the rest of your stack catches up.
- Kwanta POS
- 1C
- Shopify
- Kaspi · Wildberries · Ozon
- Marketplace listing feeds
- Looker · Metabase
Common questions
Does the agent change prices automatically?
Only if you set a rule that explicitly says so — for example "auto-match competitor within 5% on category X, weekdays only." For anything else, the agent proposes and your team approves. Guardrails are enforced before the change can ship.
How is elasticity learned?
From your historical transactions in the retail graph, re-estimated continuously as new data lands. The model runs per SKU × store × channel, with sensible pooling for sparse SKUs.
What happens if a competitor drops 30%?
The agent surfaces the signal and proposes a response inside your guardrails. It will never auto-match a move outside the floor, ceiling, or margin rules you set.
Start with Kwanta
Ship Kwanta Pricing across your network.
Talk to the founders. Get early access, honest pricing, and a direct line to the team.