Advertisement
SAP Concur

The Martech Playbook for Connecting CTV Campaigns to Commerce Outcomes

Tejas Tahmankar•Nov 16, 2026
The Martech Playbook for Connecting CTV Campaigns to Commerce Outcomes
Advertisement
SAP Concur Post Top

For decades, television advertising asked marketers to accept a difficult bargain. Brands could reach large audiences, build awareness and influence buying decisions, but connecting those exposures to actual purchases remained a challenge. Connected TV (CTV) is changing that equation by bringing addressability, audience data and digital measurement into the biggest screen in the house.

However, having access to more viewing data does not automatically mean marketers understand what drives sales. The real opportunity lies in connecting an advertisement viewed on television with the website visit, product discovery or purchase that follows.

This playbook explores how CTV advertising measurement can turn that connection into a measurable business outcome.

The Core Technology Behind Identity Resolution and the Living Room Device Graph

The Core Technology Behind Identity Resolution and the Living Room Device Graph

A consumer might watch a product advertisement on a smart TV, search for the brand on a smartphone and eventually complete the purchase on a laptop. From a marketer’s perspective, these look like separate interactions. Connecting them requires a system that can recognize relationships between devices and their users.

This is where a living room device graph becomes important. It maps relationships between connected TVs, smartphones, tablets and computers using available identity signals. Household IP matching is one method. When a television and a mobile device share a household network, an advertising platform may associate their activity and connect an ad exposure with a later website visit or purchase.

However, an IP address is a household-level signal, not proof that the same individual used both devices. Shared networks, changing IP addresses and device movement can introduce errors. Therefore, marketers need stronger identity signals wherever possible.

Google’s Ads Data Hub offers one example. Its RDID tables include Connected TV Identifiers for Advertising, which advertisers can join with hashed first-party identifiers to analyze conversions, revenue and customer lifetime value. This gives brands a way to connect advertising exposure with their own customer records.

The wider shift is towards privacy-conscious identity resolution. Probabilistic signals can help establish possible device relationships, while authenticated first-party identifiers can provide stronger matching when available and permitted. Effective CTV advertising measurement depends on knowing not only which signals connect devices, but also how reliable those connections are.

Three Core Attribution Models for CTV Commerce

Once marketers establish how devices and audiences connect, the next question becomes more important. What actually happened after someone watched the advertisement? CTV attribution provides several ways to investigate that journey, although each model answers a different question.

Post-view attribution and household IP matching

This is often the starting point for CTV conversion tracking. Since viewers cannot always click an advertisement on their television, platforms associate an advertising exposure with subsequent activity on connected devices. For example, someone watches a CTV advertisement and later visits the advertised brand’s website through a smartphone. The platform uses available household or device signals to associate the two events within a defined attribution window.

However, a subsequent visit does not automatically mean the advertisement caused it. A consumer might already have intended to purchase the product. Marketers should therefore treat post-view attribution as an indication of an association, not definitive proof of causation.

Multi-touch attribution

The customer journey rarely begins and ends with one advertisement. CTV might introduce a product, Google Search could capture the resulting interest, and a retargeting campaign might bring the consumer back to complete the purchase. Multi-touch attribution distributes credit across these interactions instead of assigning the entire conversion to the last recorded touchpoint.

This gives marketers a better view of how CTV contributes to a wider channel strategy. Nevertheless, the model depends on which interactions the measurement system can observe and how it distributes credit. It cannot independently prove that CTV generated additional sales.

Also Read: Programmatic Curation in Advertising: How It Is Simplifying Open-Market Buying and Transforming Media Strategy in 2026

Incrementality testing through holdouts

Incrementality testing addresses that limitation. Marketers divide eligible audiences or geographic markets into exposed and holdout groups. They then compare the outcomes to estimate whether the advertising generated additional visits, conversions or purchases.

This is particularly important when a campaign receives credit for conversions that might have happened through organic search or existing customer demand anyway.

Samsung’s Playtika case study offers a practical example of performance measurement. Playtika used AppsFlyer server-to-server postbacks, independently verified installs and Day-seven ROAS to assess its Bingo Blitz campaign. Within three months of launching, the company moved entirely to performance-based buying.

The lesson is straightforward. Strong CTV attribution needs reliable conversion signals, clear reporting rules and a measurement approach that distinguishes observed activity from genuinely incremental outcomes.

Connecting Audience Data and Retail Media Networks

Connecting Audience Data and Retail Media Networks

CTV advertising becomes more commercially useful when brands can bring their own customer data into the measurement process. Most businesses already hold valuable information through CRM platforms, purchase records and loyalty programmes. However, this information often sits separately from the audience data used to plan and measure advertising campaigns.

First-party data can help brands build more relevant CTV audiences. For instance, a company can use existing customer records to exclude recent buyers from an acquisition campaign, identify potential prospects or create segments for precision retargeting. Consequently, CTV audience targeting becomes more closely connected to actual business priorities.

Data clean rooms help make this collaboration possible. They provide controlled environments where advertisers, publishers and retailers can analyze permitted datasets without giving each participant unrestricted access to another party’s underlying customer information.

AWS provides one example through AWS Entity Resolution and AWS Clean Rooms. Its Retail and Commerce Media Monetization architecture describes how advertisers and retailers can combine first-party datasets for audience building, campaign performance analysis and privacy-enhanced post-campaign measurement.

This becomes particularly relevant when Retail Media Networks (RMNs) enter the picture. Platforms such as Amazon Ads and Walmart Connect give brands access to retailer environments that can connect advertising activity with purchase information. For example, a brand could run a CTV campaign and then work with a retail media partner to understand whether exposed audiences subsequently purchased the product through the retailer.

The opportunity is significant, but marketers must understand the distinction between attributed purchases and incremental purchases. A clean-room match can help identify transactions associated with an advertising exposure. It does not automatically establish that the advertisement created additional demand. That distinction should remain central to any serious retail media measurement strategy.

Building a Commerce Analytics Dashboard That Tracks What Matters

A CTV campaign dashboard should explain more than how many people watched an advertisement. It should show whether the campaign reached the intended audience, how efficiently it delivered impressions and what commercial activity followed.

Start with delivery quality. Household-level frequency helps marketers understand how often a household encounters an advertisement. Without proper controls, repeated exposure can waste budget and create ad fatigue. Video Completion Rate (VCR) shows how often viewers watch the full advertisement, while Invalid Traffic (IVT) monitoring helps identify fraudulent or non-human activity that can distort reporting.

Reach and frequency reporting also needs to account for shared viewing. Google announced that its Total Co-view definition would be adopted for video reach metrics from June 2, 2026. The change affects unique-user reach and average frequency reporting by accounting for co-viewing on connected TVs. This matters because a single television exposure can reach multiple people, making device-level reporting an incomplete picture of audience exposure.

However, delivery metrics only explain part of the campaign. Commerce analytics should track what happens after exposure through several outcome measures.

Metric

What it tells marketers

Site Visit Lift

Whether website visits increased following campaign exposure

Cost Per Acquisition (CPA)

How much the campaign spends to acquire a customer

Cost Per Add-to-Cart

How efficiently advertising generates product interest that reaches the shopping cart

Return on Ad Spend (ROAS)

How much attributed revenue the campaign generates against advertising spend

Together, these measurements create a more useful view of CTV performance. Still, ROAS needs careful interpretation because attributed revenue does not account for every business cost or prove incremental sales. A dashboard should bring delivery, exposure and commerce outcomes together without treating them as interchangeable measures of success.

Three Steps to Launch a Measurable CTV Commerce Campaign

A measurement strategy needs to be established before the campaign goes live. Otherwise, marketers risk collecting plenty of data without having a reliable baseline against which to evaluate performance.

Step 1. Establish the baseline

Audit organic website traffic, branded search activity, existing conversion rates and baseline CPA before launching the campaign. These figures provide a reference point for evaluating changes during and after CTV advertising. Where possible, establish a suitable holdout group to separate campaign-driven activity from existing demand.

Step 2. Define the attribution window

Choose a window that reflects the product’s buying cycle. Impulse FMCG purchases may suit a shorter seven-day window, while high-ticket consumer goods may require a longer 30-day window. These are starting points rather than universal standards. Adjust them according to customer behavior and validate the results against actual purchase data.

Step 3. Optimise pacing and frequency

Monitor campaign performance throughout delivery. If the same households receive repeated advertisements without generating additional engagement or conversions, shift spending towards audiences that offer incremental reach. At the same time, review frequency, site visits and acquisition costs to identify where the campaign creates genuine commercial value.

The Future of Shoppable TV

CTV is moving closer to the point of purchase. On-screen QR codes, interactive formats and TV-to-cart integrations are giving viewers more ways to act on an advertisement without relying entirely on a separate search journey.

Samsung TV Plus has introduced remote-enabled interactive video advertising using Amazon Ads technology. Eligible Amazon product advertisements allow viewers to add products directly to an Amazon cart using their television remote. Other formats can direct viewers to their phones or signup experiences.

This changes what marketers can measure, but it also raises the standard of proof. An interaction or cart addition is not a completed sale. Brands still need to connect these actions with verified purchases and incremental revenue. CTV is becoming a serious pillar of performance marketing, but its long-term value will depend on how reliably it can demonstrate business outcomes.

Advertisement
SAP Concur Post Bottom