Computer Vision is the Most Effective Path to Incremental Retail Media Revenues. Here’s Why.

The Modern Grocery Retailer’s Conundrum

Grocery retailers in Europe and the US face a unique conundrum today.

On the one hand, shoppers have returned to physical stores in droves. 80% of shoppers in major European marketers prefer to shop in-store and will do so well into the next decade . The numbers are similar in the US, with 81% of Gen Z consumers choosing in-store shopping, including for product discovery.

On the other hand, while customers may prefer in-store shopping, not just any store will do.

Shoppers today expect their in-store experience to match their online shopping experiences. So, they will prefer retailers that provide more relevant and engaging in-store experiences.

Retailers must step up to the challenge to ramp up the in-store experience because shoppers want a blended, channel-agnostic shopping experience . Retailers who don’t upgrade will struggle to engage shoppers or create incremental shopper revenues.

But weak in-store customer engagement is not the retailer’s only challenge.

Grocery retailers also operate in a notoriously low-margin industry. They also need to find sustainable ways to create and grow new revenues, increase profitability, and keep their FMCG advertisers – who fund in-store trade, shoppers, and now, Retail Media budgets – happy.

Are Physical Stores Doomed to Irrelevance?

Not if They Unlock the Power of Computer Vision.

The retail industry is quickly waking up to the power of Computer Vision – a mature AI technology with multiple practical and proven applications in physical retail stores.

Computer Vision adoption is booming across industries, from healthcare to automotive. In physical retail, its versatile front and back-end use cases, from better inventory management to in-store fraud detection, help drive operational efficiencies and lower costs.

But the real “power application” of Computer Vision in retail goes beyond improving efficiencies. It creates an immediate, high-margin, sustainable stream of revenue for brick-and-mortar retailers.

What is Computer Vision?

Computer Vision is a type of artificial intelligence (AI) that lets computers ‘see’, process, and analyze visual information almost as well as humans. The computer can identify and interpret what it sees with a high degree of confidence and can then take appropriate action in real-time based on that analysis. It is the foundational technology for fully autonomous stores such as Amazon Go.

One of the most promising use cases of Computer Vision is solving the challenge of in-store audience ‘segmentation targeting’; in real-time for brick-and-mortar retailers. Computer Vision powered sensors can identify and estimate the age, gender, and composition of a group of shoppers, combine it with contextual data like location and time, and behavioral data such as attention – and within milliseconds, serve the most relevant ad to the shoppers.

In-store Retail Media: The New ‘Power Application’ of Computer Vision

Most retailers are familiar with the Retail Media tidal wave that’s swept through the retail industry and made it the second fastest-growing media ad format after CTV . But much of the revenue has been generated by online audiences, and the bulk of revenues went to retailers with a strong e-commerce presence.

While some brick-and-mortar retailers did see early gains by joining Online Retail Media networks, the real prize – monetizing in-store audiences who form about 80% of grocery retail shoppers – has fallen short of its revenue potential, mainly due to technology limitations.

Though elusive, the revenue potential of in-store Retail Media is too significant to be ignored.?

In the post-pandemic US, brick-and-mortar retail is already growing at a faster pace than e-commerce. In 2023, Insider Intelligence says in-store audiences are an average of 70% larger than the already sizeable digital audiences for leading US-based brick-and-mortar retailers.

Calling it the next major media channel for brand advertisers, the report says in-store Retail Media offers advertisers access to shopper audiences at scale in a brand-safe environment and delivers both – branding and sales outcomes. This means retailers can go beyond DOOH and shopper marketing budgets and stake a claim to national advertising dollars.

With profit margins of 70 to 90% on a retailer’s owned channels and almost 60% to 70% of the projected $100 billion (by 2026, US), Retail Media spends touted as ‘net new spending’ — i.e, over and above historical trade and shopper budgets — in-store Retail Media truly has the potential to skyrocket profitability for physical retail.

So what is the roadblock to realizing this potential revenue? So far, the technology to digitize in-store audiences in real-time was missing. In effect, it’s hard to create a more enticing in-store shopper experience without knowing who is coming into the store or how they are behaving.

Computer Vision technology is changing that. Purpose-built Retail Media solutions powered by Computer Vision technology empower retailers to digitize in-store shoppers like online audiences. This advancement will finally unlock the towering revenue potential of in-store Retail Media.

3 Ways Computer Vision Enables What No Other In-store Retail Media Technology Can

Computer Vision is the simplest way to capture, segment, activate and measure 100% of your in-store shopper audiences. Yes. Every single one of them, in real-time.

An advanced solution like Advertima In-store Audience Creator , purpose-built for in-store Retail Media, harnesses the game-changing capabilities of Computer Vision to unlock the full Retail Media revenue potential of in-store shoppers.

• Don’t miss a single in-store shopper: capture, segment, and monetize 100% of your in-store shopper audience?
• Most in-store audience data sources leave money on the table because they cannot capture or monetize the majority of anonymous in-store shoppers.
• Even if they capture more shopper data, no other in-store audience data source can target them in real-time – during their shopping trip – with the most relevant creative.

Advertima’s in-store Retail Media solution, powered by Computer Vision, can enable both. Capture and digitize 100% of in-store shoppers, and monetize them by serving up the most relevant ad in real-time based on their attributes and in-store behavior. (see box: ‘Segmentation Targeting vs. Personalization’).

• Ensure a brand-safe environment with no-compromise privacy compliance: “profile” each shopper without needing opt-in
• Most in-store audience data sources cannot offer a brand-safe way to capture and activate anonymous shoppers without explicit shopper opt-in or registration.

Only Computer Vision, built with GDPR-compliant privacy-by-design framework, protects shoppers, advertisers, and retailers. Advertima’s proprietary real-time segmentation, built within the privacy-by-design framework, uses edge computing to make addressability possible in-store without violating any privacy regulations or requiring opt-in.

What is Edge Computing?

Edge computing refers to a range of networks and devices which are physically at or near the user. Edge is about processing data closer to where it’s being generated, enabling processing at greater speeds and volumes, leading to action-led results in real-time. (Source )

• Deliver media metrics like Online Retail Media: shift new brand marketing budgets to in-store Retail Media?
• No other in-store technology can deliver verified upper-funnel metrics (impressions, viewable impressions) and mid-funnel metrics ( views, view times).

Only a purpose-built in-store Retail Media solution like Advertima harnesses Computer Vision to capture shoppers at each stage of their journey, “profile” them based on anonymous attributes such as age, gender, group constellation, view, time, location, etc., target them in real-time, and deliver full-funnel metrics to make in-store campaign performance truly measurable.

How Advertima In-store Audience Creator Brings Computer Vision to Life

Advertima brings e-commerce-level campaign performance to in-store Retail Media. This enables retailers to shift away from location-based in-store media and towards an audience-led media offering, which creates new value for all stakeholders.

Here’s what Advertima’s proprietary In-store Audience Creator delivers to retailers:

• Real-time ‘Segmentation Targeting’: optimize campaign efficiency?
• Improve messaging relevance: with real-time segmentation targeting of in-store shoppers (based on anonymized shopper profile attributes at the PoP and PoS).
• Reach or exclude specific audiences: unlock otherwise inaccessible advertising budgets by offering dedicated brand-safe ad formats to regulated industries. For example, exclude minors from viewing tobacco ads while in-store.

‘Segmentation Targeting’ vs. Personalisation

Advertisers can address audiences in two ways:

• Segmentation targeting: anonymized 1-to-1 or 1-to-many targeting based on pre-defined segments with specific relevance for advertisers (buyers)
• Personalization: personalized 1-to-1 targeting based on individual and identified audience profiles

Advertima’s Computer Vision technology enables a unique 4-step process to capture, segment, activate and measure physical shopper audiences. This process makes segmentation targeting accessible in real-time while offering precise addressability in a 1-to-many physical store environment — exactly as is possible online.

• Capture: The advanced ML algorithms using edge computing capture in-store shoppers and assign attributes, such as age, gender, the direction of view, time, screen location, etc., to them. This builds addressable audience profiles in real-time, while preserving individual privacy and remaining compliant.
• Segment: Using edge computing, Advertima matches these audience profiles with the advertiser’s pre-defined target segments in real-time*. The privacy-by-design approach protects shoppers and makes the system faster and more stable.
• Activate: Advertisers can access these real-time segments using their existing buying process. Activation ensures that all relevant ecosystem participants know the segments and addressability work throughout the full value chain.
• Measure: The system generates verifiable top and mid-funnel metrics to optimize campaign performance

*Real-time: with Advertima, the entire ‘segmentation targeting’ process occurs within seven milliseconds. Our technology ensures no single shopper is missed, and each shopper is served the most relevant ad creative based on their attributes in microseconds.?

• Real-time yield management: optimize sellable ad inventory
• Earn more from the same number of screens: the intelligent system serves the right ad to the right shopper when they are most receptive. This minimizes wastage, helps reach campaign goals with fewer playouts, and frees up ad inventory to be further monetized.
• Forecasting: aside from using data for real-time targeting, the dedicated ML algorithm helps calculate the expected reach for each segment to better plan the campaign based on specific goals.
• Real-time ad-inventory management: optimize audience activation
• The industry needs a solution that allows multiple stakeholders to access and bid for in-store signage inventory. Advertima’s In-store Audience Creator allows just that –? multiple internal and external stakeholders to access and bid for the same (limited) in-store digital signage inventory simultaneously.
• With real-time inventory allocation, prioritize the most lucrative demand channels:
• Direct I/O bookings: based on real-time audience data, not assumptions.
• Programmatic bookings: based on real-time audience data with current programmatic standards such as OpenRTB and OpenDirect.
• Real-time ‘Audience Analytics’: optimize advertiser ROAS
• Campaign Analytics (in-flight and post-campaign)
• Includes upper-funnel metrics to in-store Retail Media: impressions, viewable impressions, qualified views, view time
• Increases granularity of performance metrics: relating unit sales and sales uplift to upper funnel metrics (per impressions, per view, etc.)
• Creative analytics: analysis of ad creatives and effectiveness helps sharpen messaging.
• Predictive analytics: analysis of first-party and contextual data progressively improves campaign effectiveness and automation.
• Screen network analytics: performance clarity by screen position, time of day, location, and campaign helps optimize the media pricing model and develop premium offerings.

Advertima metrics are designed for the integrated and performance-based future of advertising. Unlike any other in-store metrics system, Advertima’s verifiable metrics give advertisers proof of Return on Ad Spend (ROAS), just like with Online Retail Media.

The Best of Both Worlds

How Advertima Ensures Real-time Segmentation Targeting and No-Compromise Data Privacy?

Advertima’s technology does not need to store any PII – the Computer Vision extracts anonymized and contextual data without storing anything,? while assigning an impression towards a segment in real-time. The system only holds non-permanent data using edge computing in non-persistent, decentralized storage (RAM). Because Advertima’s visual sensors cannot identify the PII of any individual, it is impossible to collect, store or misuse personal data – even inadvertently.

Advertima’s Computer Vision technology uses 3D stereoscopes as optical sensors, not as cameras.?

• We do not use any facial recognition
• We never store or process any PII or biometric data
• We never film or store any image or video recordings

Ready to let Computer Vision Power Your In-store Retail Media? Make Your Best Move!

To unlock in-store Retail Media revenues, you need more advanced, not more complicated, technology.

Advertima’s Computer Vision powered In-store Audience Creator , a pioneering solution purpose-built to drive incremental in-store Retail Media revenue, is easy to deploy, use and scale anywhere in the world.

No doubt, with 73% of US retailers expected to increase in-store technology investments for 2022, Computer Vision-powered solutions like Advertima will be the logical “next step for the industry” — at least for retailers serious about creating new, high-margin revenue streams that monetize their most powerful asset – in-store shoppers.

It’s time to build a decisive game plan to build your in-store Retail Media network. However, your game plan depends heavily on your unique context and current levels of Retail Media maturity.

To plan your moves and start generating incremental high-margin Retail Media revenues from your in-store shoppers, get your copy of our latest whitepaper, ?? “How to Integrate In-Store Audience into Your Omnichannel Retail Media Network: A Playbook for Retailers. ”