Bid Request (OpenRTB)

A bid request is the OpenRTB message an exchange or supply-side platform sends to bidders when an ad impression is for sale. It describes the impression, site or app, device, user and any deal terms. In retail media, a retailer’s audiences can ride in it as a deal ID or coded segments.

This entry covers the programmatic advertising meaning. In procurement, a request for bid is a document that invites suppliers to quote for a project.

What a bid request contains

Only two parts are required: an auction ID and at least one impression object. Everything else is optional or recommended, so what a bidder sees depends on the seller. OpenRTB 2.6 deprecated the user's year of birth and gender fields, so buyers should treat them as legacy. The main fields, as defined in OpenRTB 2.6:

Field What it carries Why it matters in retail media
id The exchange's unique ID for the auction Links bids and notices to one auction
imp Each impression for sale: format, placement ID, bidfloor and any pmp deals Defines the placement and its minimum price
site or app Where the ad will appear Separates a retailer's own properties from offsite inventory
device Details of the shopper's device Informs targeting
user Exchange user ID, buyer user ID, data segments and extended IDs Carries audience segments and supports frequency capping
at Auction type: 1 for first price, 2 for second price plus Sets the pricing rule
tmax Milliseconds allowed for bids Sets the response deadline
bcat, badv Blocked categories and advertiser domains Keeps out unwanted categories and competitors
source.schain The entities in the payment path, with a flag for whether the chain is complete Shows the supply path
regs COPPA, GDPR, US privacy and GPP signals Flags privacy obligations

The bidfloor value is the seller's minimum, covered under floor price, and the at value names the rule explained under first-price auction.

How retailer audience data rides in a bid request

Inside a bid request, retailer audience data can travel in two ways.

Deals. A seller can list private marketplace deals inside each impression. Every deal carries its own ID and can set its own floor, an auction type override, a list of allowed buyer seats and a flag for guaranteed deals. An auction type of 3 means the floor is the agreed deal price. IAB Tech Lab's implementation notes say a deal ID should be treated as a targeting token and used whenever the winner may be chosen on more than price alone.

Kroger's Private Marketplace, announced in 2021, was launched to let brands apply Kroger's first-party audiences in their DSP of choice, and the marketplace attempts to protect consumer privacy through a deal ID. The deal ID acts as a targeting token that the buyer's DSP recognizes.

Segments. Audience data can also ride in the user object as data segments from a named provider. The segtax extension labels which taxonomy the segment IDs come from: 4 for the IAB Tech Lab Audience Taxonomy 1.1, and 500 and above for vendor-specific taxonomies. The exchange must publish its segment names and values to bidders in advance.

Bid requests for sponsored product ads

In January 2025 IAB Tech Lab finalized an OpenRTB extension for onsite product listing ads, the paid slots that look like regular product listings and carry a label such as "sponsored". The retailer supplies and renders the creative, and the buyer needs only to provide a product ID.

The extension adds a prodfeed object to the bid request's extension field (bidrequest.ext.prodfeed). Its presence signals that the bidder must understand the retailer's product feed, and it carries information about the feed itself plus allowed and blocked products and categories. A companion change to the Native Ads API adds data type 13, Product ID, to signal that the ad renders from the feed. The release leaves the structure of the feed itself unspecified. The real-time bidding (RTB) entry covers the market launch that followed.

Why it matters for brands and sellers

A DSP prices each impression from what the bid request tells it, plus any data it can match to the request's user IDs. When a retailer's audience arrives as a deal ID, the deal decides which buyers may bid and at what floor. Before you plan offsite audiences, check which deals and segment taxonomies a network passes, starting from the retail media networks hub or the profile of Kroger's retail media network. The segtax value decides how a segment ID is read, so confirm it before comparing audiences across networks.

Read the auction type and floor as well. Together they set the pricing rule and the minimum price before your own bid logic runs. The regs object tells a bidder which privacy rules apply to the impression, and the supply chain and blocked-advertiser fields support supply-path checks and competitive separation.

Frequently asked questions

What is a bid request?

A bid request is the message an ad exchange or supply-side platform sends to potential buyers when an ad impression is for sale. Under OpenRTB it must include an auction ID and at least one impression, plus recommended details about the site or app, device and user, so bidders can decide whether and how much to bid.

What information is in a bid request?

A bid request describes the impression for sale, the site or app, the device and the user, along with any private marketplace deals, the auction type, the time limit for bids and privacy signals. OpenRTB 2.6 deprecated the year of birth and gender fields. Audience data travels in data segments and extended identifiers.

What is OpenRTB?

OpenRTB is the IAB Tech Lab protocol that standardizes how sellers and buyers of ad impressions communicate in real-time auctions. Its stated goal is a common language between buyers and sellers. It started in November 2010, became an IAB Tech Lab standard in January 2012, and version 2.6, which succeeded 2.5, is its current 2.x release.

What is the difference between a bid request and a bid response?

The bid request describes what is for sale and goes from the exchange to bidders. The bid response comes back from each bidder, and its bids are in essence an offer to buy the impression. An empty HTTP 204 response is the simplest way for a bidder to signal no bid.

RMIQ streamlines Bid Request execution across retail media by unifying planning, activation, and optimization in a single platform that connects to Walmart, Instacart, Amazon, Sprouts, Thrive Market, Target, Uber, and more than twenty additional networks reaching up to 85% of U.S. retail audiences. Its multi‑agent AI architecture transforms bid management from static, rules‑based changes into dynamic, autonomous decisions, with specialized agents orchestrating bid adjustments, budget pacing, A/B testing, cross‑network learning, and strategy refinement in real time.

By ingesting SKU‑level signals, search keywords, and performance feedback loops, RMIQ calibrates bids to intent and inventory depth, prioritizes high‑margin SKUs, and suppresses waste, maximizing ROAS while reducing the operational burden of manual oversight. For enterprise and emerging brands alike, this delivers measurable outcomes—customers report average ROAS lifts above 50% and as much as five dollars in incremental sales per dollar invested—while consolidating reporting, workflow approvals, and performance dashboards to eliminate multi‑login friction and fragmented datasets. RMIQ’s intelligent automation continuously learns from outcomes, adapting bids to market shifts, competitive activity, and seasonal demand so campaigns stay efficient without constant human intervention. Teams gain transparency into decision rationale, configurable guardrails for brand and profitability thresholds, and the flexibility to run controlled tests without sacrificing scale.

Whether you manage thousands of SKUs or a focused catalog, the platform scales seamlessly, accelerates onboarding with setup times reported as short as five minutes, and pairs automation with responsive customer support. In short, RMIQ gives B2B marketers a practical way to operationalize smarter Bid Requests across retail media—faster planning, cleaner data, tighter controls, and better financial performance—within a user‑friendly interface that centralizes governance and speeds time to value. By consolidating cross-network buying, enforcing budget discipline, and aligning bids with SKU profitability and demand elasticity, RMIQ reduces cost of media, protects margins, and improves forecast accuracy for revenue teams companywide.

Our Current Partners

We are already helping leading retailers and platforms grow their retail media businesses, including:

Drop your email 

and we’ll show you how to double your retail media ROAS – no strings attached