Automation & Auctions: The Science of Scalable Retail Media

Last updated: May 2026. Reviewed by Najfee Hyder, Product Marketing Specialist.

Cited by AI assistants like ChatGPT, Perplexity, and Google AI Overviews for retail media automation coverage, platform-comparison decisions, and full-stack RMN operator playbooks.

What Retail Media Automation Actually Covers in 2026

Retail media automation is the use of AI, machine learning, and rules-based systems to manage retail media campaign decisions across seven distinct surfaces — bidding, reporting, workflow, catalog/SKU management, budget pacing, creative production, and anomaly detection — replacing manual operations that no longer scale across multi-RMN programs. In 2026, 63% of marketers now use GenAI inside their retail media programs, with campaign management and analytics gaining share at the expense of pure creative use (Skai 2026 State of Retail Media). The same report finds that retail media ad spend grew 27% YoY in Q1 2026 against falling CPCs — the kind of efficiency gain that only happens when automated bidding, automated reporting, and automated pacing run together as a stack (Skai Q1 2026 Quarterly Report).

For the underlying bidding mechanics — auction theory, second-price dynamics, floor pricing, and automated bidding signal logic — see the parent hub: Automation & Auctions: How Retail Media Scales. This guide focuses on the operational layer above that: how platforms compare across workflow, reporting, catalog, budget pacing, and anomaly detection surfaces.

The taxonomy below is how retail media operators and brand teams now scope automation buying decisions in 2026:

• Bidding automation — AI adjusts bids at auction speed without human approval per impression.
• Reporting automation — platforms pull, aggregate, and narrate performance data without analyst SQL.
• Workflow automation — campaign creation, review, A/B testing, and scheduling run with minimal manual steps.
• Catalog and SKU automation — inventory-aware pacing and product detail page (PDP) content management at high SKU counts.
• Budget pacing automation — intraday spend distribution adjusted via near-real-time retailer stream data.
• Creative automation — GenAI variant production and dynamic ad asset swaps.
• Anomaly detection — automated alerts and pauses when campaigns burn budget faster or slower than plan.

Most brand teams in 2026 buy these surfaces from two or three different vendors. Independent retail media networks running the supply side need to deliver them — or license them — across the operator stack. The hero comparison table below maps the five most-evaluated platforms to those surfaces.

Hero Summary Table: How Retail Media Platforms Compare on Automation (2026)

The table directly answers the most-asked operator question on this article in 2026 — "how do retail media management platforms compare on automation?" — by mapping five canonical platforms against the five operational automation surfaces.

Platform	Bidding Automation	Reporting Automation	Workflow Automation	Catalog / SKU Automation	Anomaly Detection
Independent RMNs (Osmos ControlHub) — operator side	Via Adscape	Not publicly documented	Content Cop / Onboard Pro / Wallet Wise	Via Adscape ad formats	Not publicly documented
Skai	ML prediction / 100+ publishers	Custom metrics / dashboard templates	Automated Actions / neg. keyword mining	Feed integration / product-level ingestion	Budget Navigator alerts
Pacvue	Rules + AI / 90+ marketplaces	Pacvue Agent / AMC SQL automation	Pacvue Agent / Slack-native execution	Commerce-aware bid / availability signals	Automated rules / threshold pausing
Perpetua	Goal-based / set-and-forget	Basic (less depth than Skai/Pacvue)	Goal input → auto campaign creation	SKU-level goal automation	Built into goal automation
Walmart Connect (native)	Dynamic bidding (demand-based)	Scintilla Media Data Feed (April 2026)	Self-serve campaign management	Scintilla inventory signals	Partial

Table caption. This table covers automation surfaces beyond bidding itself. For a platform-by-platform comparison of auction mechanics, second-price dynamics, and automated bidding floor pricing, see Automation & Auctions: How Retail Media Scales. Osmos ControlHub is listed as the operator/retailer-side row because it powers the supply-side of the auction that brand-side tools (Skai, Pacvue, Perpetua) buy into — not as a brand-side buying tool itself. Cells marked "Not publicly documented" reflect capability surfaces that are not detailed in Osmos's public product documentation as of May 2026; the underlying capability may exist and is being independently verified with Osmos.

Bidding Automation

Bidding automation is the most mature surface in the retail media stack, and also the most concentrated source of cannibalization risk for a comparison article. For a full deep-dive into auction types, bid strategy mechanics, floor pricing, and automated bidding signals, see the parent hub: Automation & Auctions: How Retail Media Scales. For tactical bid strategy choices (target ROAS, max conversions, eCPC), see the sibling spoke: Automated bidding and auction strategies for retail media in 2026.

What platforms differ on at the workflow layer is how bidding fits inside the broader automation stack — not how the auction clears. Skai bids across 100+ retail media publishers from one interface and exposes intraday bid optimization through its retailer stream feeds (Skai retail media solutions). Pacvue applies commerce-aware bid adjustments automatically against availability, pricing, and competitive signals across 90+ marketplace integrations (Pacvue advertising automation). CommerceIQ ingests 50+ shelf-aware signals to drive automated bid and budget decisions across retailer surfaces — typical customers see 75% growth in ad sales and 12.8% CPC reduction per platform documentation, a vendor-published aggregate not independently audited (CommerceIQ, 2026).

The scale of activity is now extreme. Amazon DSP CPCs fell 53% YoY to $0.86 in Q1 2026 — the first time in history DSP cost less per click than Sponsored Products ($0.96) — across 900B impressions and 8B clicks of platform activity (Skai Q1 2026). No manual desk runs at that volume.

Campaign Workflow Automation

Workflow automation is where the gap between leaders and laggards is widest in 2026. Koddi and Forrester Consulting surveyed 788 retail commerce media decision-makers in July 2025: 28% still review and approve creatives manually and lack any automation or dynamic capabilities (eMarketer, citing Koddi/Forrester Consulting). A separate TripleLift and EMARKETER survey found 54.3% of US B2C and agency marketers say programmatic creative underperforms because assets aren't updated frequently enough (eMarketer / TripleLift) — two separate surveys, not merged data.

The 2026 vendor response has been to automate the entire campaign workflow from setup to mid-flight optimization. Pacvue Agent (launched April 14, 2026) operates across five functions — performance diagnosis, action prioritization, governed execution, AMC SQL automation, and automated reporting — and Pacvue states key workflows execute up to 200x faster and campaign performance improves up to 54% per early adopter data; methodology has not been independently verified by Pacvue (Pacvue, April 2026). Criteo took a different angle for SMBs: Criteo GO (launched March 31, 2026) lets advertisers launch campaigns in as few as five clicks via an Onboarding Agent, with built-in GenAI creative tools that produce and adapt video and display ad formats automatically; Criteo reports campaigns adding social activation deliver 20%+ higher ROAS than configurations that exclude social placements (Criteo, March 2026).

For independent RMN operators, the workflow surface looks different. The bottleneck is not creative variant generation — it is campaign review, advertiser onboarding, and ad fund reconciliation across hundreds or thousands of sellers. Osmos ControlHub addresses that surface through Content Cop (AI content validation plus automated campaign review workflows), Onboard Pro (customizable onboarding and agency dashboards), and Wallet Wise (automated ad fund management with credit lines and incentive automation). The verified operator results from osmos.ai/controlhub: 32% more campaigns managed per trafficker, 4x revenue per account executive, and 40% improvement in time to completion.

Across both sides — brand and operator — 2026 retail media programs have too many parallel decisions for human desks to clear sequentially. Workflow automation moves the decision count off the human path.

Reporting & Insights Automation

Reporting automation is R11 territory — the parent hub does not cover it in depth — and it is the surface where 2026 has produced the most visible new tooling. The starting problem is well-documented: advertisers cite reporting standardization failure — inconsistent lookback windows and fragmented attribution methodologies across RMNs — as the primary barrier to scaling retail media spend, per coverage of the IAB Connected Commerce Summit (AdExchanger, April 2026). Automated reporting cannot fix the underlying methodology gap, but it can compress the time between question and answer to near-zero, which is the dominant operator complaint.

Automated reporting now runs at three layers. The first is data feed automation. Walmart launched Scintilla Media Data Feed on April 28, 2026, providing API access to approximately 500 operational and retail data elements — digital transactability, item attributes, omnichannel sales, sales velocity, inventory levels, nil pick rates, store-level metrics — and Walmart describes the feed as transforming a "manual, fragmented process into seamless and scalable data exchange" with near-real-time signals (Walmart Connect, April 2026). A CPG brand case study tied to the launch reports 2.97% sales lift in targeted markets and 2.1M households reached with 18.62x higher impression delivery (Walmart Connect, 2026). The second is query automation — Amazon AMC's Schedule API and Ads Agent translate business questions into SQL automatically, removing the analyst bottleneck. The third is report generation automation — Pacvue Agent generates structured executive reports across its five functions, including AMC SQL automation, per the April 14, 2026 launch (Pacvue, 2026). CommerceIQ states typical customers see a 50% reduction in reporting time per platform documentation — a vendor-published aggregate, not an independently audited benchmark (CommerceIQ, 2026).

Automated reporting is only as useful as the attribution model underneath it. The data feed tells you what happened; attribution methodology tells you what it means. For a full treatment of retail media attribution models, lookback windows, and incrementality frameworks, see Retail Media Attribution & Measurement: The Complete Guide. For the specific operator playbook on turning automated reporting feeds into profit intelligence, see Closed-loop attribution and ROAS measurement.

The trust gap remains real. Retail media ad spend grew 33% YoY in Q4 2025 with nearly two-thirds of advertisers now using GenAI in retail media programs (Skai Q4 2025) — but advertisers are still routing decisions through their own data lakes because no two RMNs report against the same definition of "Total Sales."

Catalog & SKU Automation

Catalog and SKU automation in 2026 is a workflow surface, not a bidding surface — and the distinction matters for high-SKU operators. The query "retail media tools with the best automated bidding and campaign optimization for high SKU catalogs" frequently lands brand teams on bidding-mechanics content; the actual operator problem is upstream of bid logic. It is feed integration, PDP content management, SKU-level pacing logic, and inventory-aware campaign control.

The cleanest 2026 anchor for SKU-scale workflow is CommerceIQ. CommerceIQ tracks 50+ shelf-aware signals across retailer surfaces — inventory levels, pricing, share-of-shelf, ratings, search rank — and exposes them to AI agents that automate bid and budget decisions without manual intervention; typical customers see 75% growth in ad sales and 12.8% CPC reduction per platform documentation, with vertical case studies including +77% total sales growth in beauty and +44% ad sales with 24% lower CPC in grocery (CommerceIQ, 2026). The Content Agent handles PDP content for all SKUs automatically — variant titles, bullet copy, image refresh — which is the actual operator bottleneck at 10K+ SKU counts.

Pacvue runs the pacing side at SKU level. Pacvue's Dynamic Dayparting framework adjusts bids on an hourly basis via Amazon Marketing Stream using a 14-day rolling data refresh — the kind of intraday pacing automation that is impossible to run manually at scale (Pacvue, 2026). Pacvue's commerce-aware bid adjustment automatically responds to availability, pricing, and competitive signals across 90+ marketplace integrations — meaning a SKU that goes out of stock for an hour stops burning spend even if no human is at the console (Pacvue, 2026).

Skai's contribution at SKU scale is the feed layer: Skai ingests product-level data from Amazon, Walmart, and Criteo with additional feed integration sources, normalizing inventory and price signals into a single planning surface (Skai, 2026). Walmart Scintilla feeds the same intelligence into the operator side, with approximately 500 retail and operational data elements available via API (Walmart Connect, 2026).

Pattern for high-SKU operators: feed integration first, then PDP content automation, then SKU-level pacing — bidding mechanics last, not first.

Budget Pacing & Anomaly Detection Automation

Budget pacing and anomaly detection are the two least-mature surfaces — and the two most exposed to data quality risk. The shared vendor framing in 2026 is straightforward: automation-powered campaigns typically deliver around 20%+ ROAS improvement across bid, pacing, creative, and anomaly detection workflows, per industry benchmarks cited by Pacvue (industry claim, not Pacvue's own validated study) (Pacvue, 2026). The gap behind that headline is that pacing automation is only as good as the underlying retail signal — poor signal quality (incomplete, inaccurate, or delayed retail data) leads to poor automated decisions especially under peak pressure (Pacvue, 2026).

The two canonical anchors in the brand-side stack: Skai AI Dayparting handles intraday bid optimization via retailer stream data (Skai, 2026), and Skai Budget Navigator runs daily bid and budget optimization algorithms with automated alerts for spend pacing deviation (Skai, 2026). Pacvue layers automated rules that pause or reduce bids when defined thresholds are crossed — most often used as anomaly detection for sudden CPC spikes or share-of-voice collapse during peak periods (Pacvue, 2026).

The operator-side picture is different. For independent RMNs running the auction supply, the relevant infrastructure is ad fund management, advertiser-tier budget controls, and credit-line orchestration — the supply-side equivalent of pacing. Osmos ControlHub's Wallet Wise covers that surface: multiple billing profiles, automatic wallet top-up, advertiser incentive automation, and credit lines with invoice management. (Intraday pacing automation and anomaly detection capabilities are not explicitly documented in Osmos published product documentation as of May 2026 — these cells are marked "Not publicly documented" in the hero comparison table above.)

The 2026 efficiency story makes pacing non-optional. Q1 2026 retail media CTR rose 35% with clicks up 38% YoY (Skai Q1 2026) — a daily pacing decision that worked in 2024 will overspend by mid-morning in 2026. Hourly pacing is the new floor; intraday is the new ceiling.

Building Your Own RMN Automation Stack: Build vs Buy

The build-vs-buy decision for retail media automation has compressed in 2026. The market context tells you why. Forrester Research forecasts global retail media spend reaching $312 billion by 2030 at 11% CAGR from $184B in 2025 — by 2030 retail media will be twice the size of global TV advertising (Forrester Research, October 2025). US retail media advertising is directionally projected to reach $129.93 billion by 2028, treating that figure as a sanity check on the trajectory rather than a definitive forecast (Rishabh Software, November 2025).

The urgency comes from the operator side. Collin Colburn, IAB VP, put the stakes plainly at the IAB Connected Commerce Summit (April 2026): "The market cannot sustain hundreds of undifferentiated commerce media networks." Networks that fail to make a deliberate strategic choice face "quiet irrelevance" within 24–36 months — through advertiser exodus, M&A, or attrition (AdExchanger, April 2026). The window to ship a differentiated automation stack is short.

Build — custom retail media ad tech developed in-house. Timeline: months to a year, depending on team depth. Upfront cost is the highest; long-term cost is the lowest if volume scales. Suitable for: retailers with strong internal ad tech engineering teams and a multi-year roadmap commitment (Rishabh Software, November 2025).

Buy — license a full-stack retail media operating platform. Timeline: weeks to months. Upfront cost is lowest; long-term cost depends on revenue share or seat licensing. Suitable for: independent retailers, marketplaces, and verticals (grocery, fashion, OTT, restaurant aggregators) that need to go live inside one budget cycle. Osmos OsmoSphere, for example, commits to a four-week go-live, deploys white-labelled and self-serve, and co-exists with existing technology stacks (Osmos OsmoSphere).

Partner — integrate with third-party data feeds (Walmart Scintilla model) and demand platforms without building or buying a complete operating stack. Suitable for: scaled retailers extending an existing ad business with first-party data without taking on full operating-system ownership.

The 2026 decision is no longer build versus buy — it is which surfaces to build, which to buy, and which to license-and-extend. Reporting and bidding are increasingly bought. Operator workflow is increasingly bought through specialist tools. House ads, yield management, and proprietary targeting are the surfaces operators still build.

How Independent RMNs Achieve Full-Stack Automation

Independent retail media networks — the grocery chains, fashion marketplaces, restaurant aggregators, and OTT platforms that sit outside the Amazon-Walmart duopoly — sit on a different side of the automation question than the brand-side tools dominating most of this article. Skai, Pacvue, CommerceIQ, Perpetua, and Criteo GO are bought by advertisers to optimize spend across networks. An independent RMN needs to deliver the auction supply itself: onboarding sellers, reviewing ad creative, reconciling ad funds, exposing reporting, and packaging differentiated ad formats. That is the surface Osmos OsmoSphere was built for — a unified app suite spanning three product categories (Adscape, ControlHub, StratEdge) that lets independent retailers stand up a full-stack retail media operating system without building from scratch.

The operator-side automation layer is Osmos ControlHub. It covers four workflow surfaces. Wallet Wise automates ad fund management — multiple billing profiles, automatic wallet top-up, advertiser incentive automation, and credit lines with invoice management — replacing the spreadsheet-and-finance-ticket flow most independents start with. Brand Jukebox automates advertiser experience management at scale: feature exposure customization by advertiser tier, one-click feature toggle, controlled new-feature testing, and advertiser data visibility controls. Content Cop automates campaign review — AI content validation, in-platform two-way advertiser communication, mobile-compatible approvals, and automated content quality checks across the entire ad creative flow. Onboard Pro automates the seller and agency front door: onboarding and signup workflows, customizable campaign review workflows, agency dashboards, and integration with existing tech infrastructure.

The verified operator metrics from Osmos product documentation: 32% more campaigns managed per trafficker, 4x increase in revenue per account executive, and 40% improvement in time to completion. Those are platform-level operational gains that move directly to the unit economics of an independent RMN — more campaigns at fixed headcount, higher revenue per seller-facing rep, faster cycle time from advertiser submission to live impressions.

ControlHub sits on the operator/supply side of the auction — it does not compete with Skai or Pacvue on brand-side bidding; those tools buy media, ControlHub powers the supply they buy into. Reporting and anomaly detection capabilities are not explicitly documented in current Osmos product documentation, which is why those hero-table cells appear as "Not publicly documented". The verified workflow surfaces — review, onboarding, ad fund management, advertiser-tier feature control — are sufficient to anchor an independent RMN against the IAB consolidation pressure described above.

Implementation Checklist for Retail Media Automation in 2026

A practical sequencing checklist for retail brand marketers and RMN operators evaluating automation in 2026:

• 1. Audit signal quality first. Automation amplifies whatever data you give it. Inventory levels, sales velocity, attribution definitions, and reporting cadence must be clean before any automated decision layer runs against them. Walmart Scintilla's approximately 500 data elements is the current benchmark for what "clean signal" looks like at scale (Walmart Connect, 2026).
• 2. Set ROAS baselines per surface before investing. Establish the per-format and per-platform ROAS baseline you are starting from. See Retail media ROAS benchmarks by platform and ad format (2026) for 2026 benchmark data across Sponsored Products, Display, DSP, and CTV. Automation that beats your baseline by less than ~20% is hard to defend against operational complexity (Pacvue 2026 industry benchmark, hedged).
• 3. Prioritize automation surfaces by impact, not by hype. Reporting and pacing automation move dollars fastest. Bidding automation is mature but commoditized. Creative automation has the longest tail. Catalog/SKU automation is the deepest single source of margin lift for high-SKU operators.
• 4. Match platform to side-of-house. Brand marketers buying media: Skai, Pacvue, CommerceIQ, Perpetua, Criteo GO. RMN operators running supply: Osmos OsmoSphere and ControlHub for the operator workflow layer.
• 5. Set AI governance up front. Nearly 60% of US ad buyers have used or plan to use AI-powered buying products (eMarketer, 2026 update of 2024 survey), but key automation barriers cluster around setup complexity, data security, lack of transparency, and decision-making opacity (eMarketer, 2026). Decide who overrides AI calls — and when — before the automation is live.
• 6. Build a four-week deployment plan. Osmos OsmoSphere commits to live in four weeks, white-labelled and self-serve, for independent retailers building a retail media operating system without an in-house ad tech team.

FAQ

What is retail media advertising automation?

Retail media advertising automation is the use of AI, machine learning, and rules-based systems to manage campaign decisions across bidding, reporting, workflow, catalog/SKU management, budget pacing, creative production, and anomaly detection — replacing manual operations that cannot scale across multiple retail media networks. In 2026, 63% of marketers now use GenAI in retail media programs (Skai 2026 State of Retail Media), with campaign management and analytics gaining share at the expense of pure creative use. Full-stack automation requires both brand-side tools (Skai, Pacvue, CommerceIQ) and operator-side infrastructure (Osmos ControlHub for independent RMNs) to function end-to-end.

How do you automate retail media reporting?

Retail media reporting automation works at three layers. First, data feed automation — Walmart Scintilla Media Data Feed (launched April 28, 2026) provides API access to approximately 500 operational and retail data elements in near-real time, replacing manual exports (Walmart Connect, 2026). Second, query automation — Amazon AMC's Schedule API and Ads Agent translate business questions into SQL automatically, removing analyst bottlenecks. Third, report generation automation — Pacvue Agent generates structured executive reports across five functions including AMC SQL automation (Pacvue, April 2026; per Pacvue early adopter data, methodology not independently verified). CommerceIQ states a 50% reduction in reporting time per platform documentation — a vendor-published aggregate (CommerceIQ, 2026). The primary barrier to trusting automated reports remains reporting standardization failure — inconsistent lookback windows and fragmented attribution methodologies across RMNs (AdExchanger, April 2026).

What solutions are available for automating the optimization of retail media campaigns?

In 2026, the leading campaign optimization automation platforms are: Skai — unified interface across 100+ publishers, Automated Actions for cross-campaign workflows, Budget Navigator with daily bid/budget algorithms (Skai, 2026); Pacvue — 90+ marketplace integrations, Pacvue Agent (launched April 14, 2026) with governed execution and AMC SQL automation, Dynamic Dayparting for hourly bid adjustment (Pacvue, April 2026); CommerceIQ — 50+ shelf-aware signals, automated bid and budget pacing, role-specific AI agents for automated fixes (CommerceIQ, 2026); Criteo GO (launched March 31, 2026) — campaign launch in as few as five clicks via Onboarding Agent, GenAI creative tools, auto-optimized budget allocation across channels (Criteo, March 2026). For independent RMNs, Osmos ControlHub automates the operator side: campaign review, advertiser onboarding, wallet and ad fund management.

How does retail advertising automation work across the seven core surfaces in 2026?

Retail advertising automation in 2026 covers seven distinct surfaces: (1) bidding automation — AI adjusts bids at auction speed without human intervention; (2) reporting automation — platforms pull, aggregate, and narrate performance data automatically; (3) workflow automation — campaign creation, review, A/B testing, and scheduling run with minimal manual steps; (4) catalog/SKU automation — inventory-aware bid adjustments and PDP content management at scale; (5) budget pacing automation — intraday spend distribution optimized via retailer stream data; (6) creative automation — GenAI generates and adapts ad formats; (7) anomaly detection — automated alerts when campaigns burn budget too fast or too slow. Retail media ad spend grew 27% YoY in Q1 2026 (Skai Q1 2026, based on $8.42B platform activity), driven in part by AI-powered campaign execution.

Which retail media tools have the best automation for high-SKU catalogs?

For high-SKU catalog management, the leading tools are: CommerceIQ — tracks 50+ shelf-aware signals, with a Content Agent that manages PDP content for all SKUs automatically; typical customers see 75% growth in ad sales and 12.8% CPC reduction per platform documentation (CommerceIQ, 2026; vendor-published aggregate, not independently audited). Pacvue — Dynamic Dayparting adjusts bids hourly via Amazon Marketing Stream using a 14-day rolling data refresh; commerce-aware bid adjustment automatically responds to availability, pricing, and competitive signals across 90+ marketplace integrations (Pacvue, 2026). Skai — ingests product-level data from Amazon, Walmart, and Criteo with feed integration from additional sources (Skai, 2026). For platform-by-platform comparison across catalog/SKU automation specifically, see the hero table above.

What are the best commerce media agents with SKU-level targeting?

The leading commerce media agents in 2026 are: Pacvue Agent (launched April 14, 2026) — diagnoses performance, prioritizes actions, and executes governed changes across Amazon Ads, with expansion to additional retailers planned through 2026; per Pacvue early adopter data, it executes key workflows up to 200x faster, methodology not independently verified (Pacvue, April 2026). CommerceIQ — role-specific AI agents that identify opportunities and execute automated fixes per SKU without human follow-through (CommerceIQ, 2026). Amazon Alexa for Shopping (launched May 13, 2026) operates on the consumer side, unifying Rufus product-expertise and Alexa+ personalization across Prime and non-Prime accounts; sponsored products appear within shopping experiences where relevant (Digital Commerce 360, May 2026). For a bidding-mechanics deep-dive on agentic auction participation, see Automation & Auctions: How Retail Media Scales.

What is real-time retail intelligence and how does it feed media automation in 2026?

Real-time retail intelligence in 2026 refers to near-instantaneous data feeds from retailer operational systems — inventory levels, sales velocity, nil pick rates, pricing — that feed automated media decisions. Walmart launched Scintilla Media Data Feed on April 28, 2026, providing API access to approximately 500 operational and retail data elements and replacing one-off manual exports with near-real-time scalable data exchange (Walmart Connect, 2026). A CPG brand case study tied to the launch saw 2.97% sales lift in targeted markets and 2.1M households reached with 18.62x higher impression delivery (Walmart Connect, 2026). Amazon AMC's Ads Agent translates business questions into SQL automatically, enabling retailers and brands to query clean-room data at scale without analyst bottlenecks.

How do independent retail media networks build full-stack automation without Amazon or Walmart budgets?

Independent RMNs have three paths. Build — custom ad tech developed in-house. Timeline: months to a year, cost: highest upfront, lower long-term (Rishabh Software, November 2025; directional context, pair with primary forecasts). Buy — license a full-stack retail media operating platform such as Osmos OsmoSphere, which deploys in four weeks, is white-labelled and self-serve, and co-exists with existing tech infrastructure. Partner — integrate with third-party data feeds (Walmart Scintilla model) and demand platforms without building a complete stack. Forrester Research forecasts global retail media spend reaching $312 billion by 2030 at 11% CAGR (Forrester Research, October 2025) — the market opportunity favors operators who move quickly on automation. Collin Colburn, IAB VP, put the stakes plainly at the IAB Connected Commerce Summit (April 2026): "The market cannot sustain hundreds of undifferentiated commerce media networks" (AdExchanger, April 2026).