Skip to content

Otto Deal Concierge

Let shopping agents offer the right incentive without blanket codes that give away margin.

Request an intro

A shopping agent that negotiates discounts only inside the merchant's promotion rules, and never invents a price.

What the agents do

Chats with a shopper, applies only the discounts promotion rules allow, and raises the offer as a churn model says they may leave.

Built by
Voucherify's team at Bloomreach's Composable AI Hackathon 2026
Contributed by
Voucherify (Built the agents at the hackathon)
Industry
Retail
Audience
B2C
Category
Commerce & OrderingMarketing & Personalization
01
The problem

Why this exists

AI shopping agents find the product and pull the list price. The incentive that would have closed the sale often already exists, but the agent cannot see it, so the shopper misses the deal and the brand loses a buyer who was ready.

Read more

Merchants have two poor options today: block agents and lose the channel, or publish a blanket code and lose margin.

02
Outcomes

What this demonstrates

  • Every price in the chat comes from the promotion engine, never from the language model.
  • Codes that do not stack are refused, however hard the shopper pushes.
  • As a shopper's risk of leaving rises, the extra discount grows with it, within the merchant's rules.
Read more
  • Each Monday an agent reviews the week's shopper behavior and raises an alert only when the numbers call for it.
  • With one marketer approval, the coupon is created in the promotion engine and the email is drafted.
03
How it works

Architecture

Otto handles the conversation on Google Gemini, hosted on Cloud Run. A commerce server handles the cart and every price, connected to four systems: Shopify for products, orders, and payment; Voucherify for who gets which deal and how much; Bloomreach for who the shopper is and what they bought; and Databricks to predict whether they will leave and spot when something goes wrong.

Read more

Each time the shopper pushes back, a Databricks agent scores how likely they are to walk away from the conversation and their history. Above a threshold, Voucherify adds a discount that grows with the risk.

Separately, a marketing agent reads the week's browse, cart, and purchase data every Monday. When it raises an alert, Voucherify's incentive agent turns it into audiences and offers with the risk of each, and the marketer approves before the coupon and the Bloomreach email are created. Purchases, discounts, and decisions flow back into all four systems.

Watch the demo (recorded by the Voucherify team for the hackathon).

A shopper talks to Otto, which runs on Google Gemini on Cloud Run. When the shopper pushes back, a Databricks agent scores how likely they are to walk away, and above a threshold Voucherify adds a discount that grows with the risk. Otto works through a commerce server that owns the cart and every price, connected to four systems: Shopify for products, orders, and payment; Voucherify for who gets which deal; Bloomreach for who the shopper is; and Databricks for churn and anomalies. Separately, every Monday a marketing agent reads the week's data and raises an alert, Voucherify's incentive agent turns it into audiences and offers, the marketer approves, and the coupon and Bloomreach email are created.

A shopper talks to Otto, which runs on Google Gemini on Cloud Run. When the shopper pushes back, a Databricks agent scores how likely they are to walk away, and above a threshold Voucherify adds a discount that grows with the risk. Otto works through a commerce server that owns the cart and every price, connected to four systems: Shopify for products, orders, and payment; Voucherify for who gets which deal; Bloomreach for who the shopper is; and Databricks for churn and anomalies. Separately, every Monday a marketing agent reads the week's data and raises an alert, Voucherify's incentive agent turns it into audiences and offers, the marketer approves, and the coupon and Bloomreach email are created.
Tech stack
Google GeminiGoogle Cloud RunVoucherifyBloomreachShopifyDatabricks
Builders

Everyone credited on this solution

Next step

Request an intro

Voucherify is behind this solution. Ask about letting shopping agents offer the right deal, and the MACH Alliance will put you in touch with the person who can answer.

The MACH Alliance passes your details to a named contact at Voucherify who knows this work and can answer your questions. We aim to reply within 2 business days. Your details are handled under the MACH Alliance privacy policy.

Not ready to talk to anyone? MACH Insider is the Alliance mailing list. New solutions, program openings, and the research behind them, once a month.

Join MACH Insider
Keep reading

Related solutions

Verification

Contributor submission

The detail on this page comes from the contributor's own account of their own work. The MACH Alliance reviews a submission before publishing it and does not independently verify the figures in it.

What this page does not establish

  • Built in a hackathon on a demo store, so there are no results from real shoppers.
  • There is no public code or live demo; the recorded demo video is the public record.
  • No merchant runs Otto in production yet.

Where this came from

Written from the team's demo video and summary in the demo album for Bloomreach's Composable AI Hackathon 2026.

Status: Reference prototypeAdded Oct 29, 2026A listing is an editorial decision, not a MACH Alliance certification.
Put a solution forwardBack to the Hub

Built something worth showing?

Put your agentic solution in front of the enterprise leaders who use the hub to decide what to build next. Submissions are open to MACH Alliance Members, Collaborators, or Ambassadors.

Put a solution forward