There are analytical criteria and measurements that can help, and attribution is one of them. Attribution tools for digital and TV advertising have already proved their worth in improving the ad results. Attribution is able to increase marketing efficiency up to 15-30%.
Here at MindK we also see the impact of attribution on efficiency. For example, an intelligent attribution system that we developed for leading direct-response advertising agencies in the United States, helps our client to optimize ad strategies and increase Return on the Advertisement (ROAS) thanks to 360-degree campaign analytics and a powerful parsing module.
Attribution in advanced TV, which includes connected TV (CTV) and over-the-top (OTT) services, is an emerging practice that seeks to meet the same goals – better ROAS and ad effectiveness.
If you don’t fully understand the difference between CTV and OTT you can check out our previous article dedicated to CTV and OTT advertising. It will help you gain a better idea of attribution, which we’ll discuss here.
In spite of the fact that OTT attribution technologies are not yet perfect, some advanced attribution tools and software are already taking the ad measurement to a new level. But HOW? In this article, we’re going to puzzle out how attribution works for OTT advertising and what you need to know before implementing this functionality.
Table of contents:
Before we move to attribution in OTT advertising, let’s make sure we are on the same page about what attribution actually means.
Attribution is a marketing term that means an analytical science focused on specifying the ROI of any marketing investments (including advertising), by receiving valuable data on what specific touchpoint made the potential client convert.
A touchpoint refers to any interaction of the user with the brand that may include website visits, online ads, emails, reviews, company events, referrals, and much more. By understanding which touchpoints are effective, marketers and advertisers can work on adjusting their campaigns.
In essence, the attribution enables advertisers and marketers to:
- optimize spendings: by analyzing both strong and weak points of the strategy, advertisers can use budgets more effectively;
- perfect targeting: attribution provides data about the needs and preferences of the potential customers; and
- improve ROI: a marketing strategy based on valuable attribution insights helps to focus both on customer needs and increasing revenue.
Among a number of drivers for attribution metrics in the advertising industry, including OTT advertising are budget constraints and an ever-growing number of channels and devices.
As we’ve already discussed in the article about Big Data in advertising, advertisers are no longer able to justify ad budgets without providing measurable ROAS. 84% of marketers admit they are under growing pressure to prove the effectiveness of their campaigns. What makes things worse is the fact that people are using more devices and channels during their purchase journeys than they used before which complicates tracking the attribution.
Source: Measuring Success in Digital Advertising: 2020 Report
Emarketer estimates that more than 84% of companies in the United States will use digital attribution models by the end of 2021. Still, to make attribution work for the sake of your business needs, it is very important to understand the ways to track it.
To track the attribution, marketers use a number of various attribution models. Generally, the attribution models fall into single-touch or multi-touch.
A single-touch model takes into account one single touchpoint which leads to the sale. It may include:
- first interaction attribution, which considers the first touchpoint of the user who converted, and
- last interaction attribution, which assigns credit to the last event along the user path to conversion.
The single-touch model is quite straightforward, as knowing the first and the last touchpoints is very important. However, the modern user journey is very fragmented. People use various channels and instantly jump from one to another.
For example, the customer might click on your ad on social media or streaming services, review your brand or offer, then a few days later go to your website, a week later they would get retargeted to then finally make a purchase. In the case of single-touch attribution, you only focus on the first or last touch point but completely miss the entire buying process.
The multi-touch attribution model, on the contrary, allocates parts of the conversion to each of the channels along the customer journey. It involves:
- linear attribution: allocates an equal amount of credit for each touchpoint;
- time decay attribution: gives most credit to the touchpoints closer to the sale or conversion;
- position-based attribution, also called U-shaped: allocates most of the sales credit to the first and last touchpoints; and
- last non-direct click attribution model that assigns credit to a last-click, but eliminates any “direct” interactions (manually entering a website or clicking the saved link) that happen right before the conversion.
It’s important to say that none of the attribution models is one-size-fits-all. A lot of companies follow different, even custom, sales cycles, which is why each business should define an attribution model that will fit a certain environment and set of objectives.
Taking into account the above-mentioned definition of attribution, OTT attribution is an approach to measuring the effectiveness of ad campaigns shown on streaming services.
However, we cannot talk about OTT attribution separately from Connected TV. The main reason for that is that the “large screen” remains the preferred screen to watch video content. It means that most OTT content is watched by means of Connected TV, which influences measuring OTT attribution. For instance, around 70% of the time, Netflix is accessed through Connected TVs, as well as the majority of YouTube TV.
Compared to linear TV attribution, OTT and CTV attribution is much more informative. Traditionally, TV advertisers relied on limited demographics and probabilistic matching to analyze connections between TV spots and consumer behavior. It means that TV advertisers get some standard reports with impressions delivered by demographic based on TV ratings and approximate data provided by Nielsen (that measures almost 50% of the global broadcast, print, and online advertising). However, understanding the influence of impressions requires additional efforts like paid brand studies, incremental lift analyses, and similar.
By means of SDK (Software Development Kit) integrations, Application Programming Interfaces (APIs), tracking pixels, using additional software to analyze the engagement and behavior that resulted in conversions, OTT attribution provides a better understanding of ad performance and viewer behavior.
There are three main tools that provide OTT ad campaign measurements:
- Demand-side platforms (DSP), which, in most cases, aside from enabling advertisers to manage and deliver ads, provide OTT and CTV ad campaign metrics. While using DSP to measure attribution doesn’t involve additional costs, demand-side platforms might be biased. As far as DSPs provide ad delivery, they might do their best to make the ad campaign attribution metrics look as good as possible.
- Third-party ad servers, which require additional costs and can not guarantee protection from ad fraud.
- Measurement and analytical tools, which are unbiased platforms focus on providing the up-to-date technologies of measuring attribution and ad performance. Different providers have various technologies, approaches, and algorithms to track performance.
Despite the fact that different attribution measurement providers suggest various approaches to measuring OTT analytics, there is a general high-level overview of the attribution process for Advanced TV. This was recommended by the Interactive Advertising Bureau (IAB), a nonprofit organization that provides legal support and develops industry standards.
At first glance, the whole attribution process may seem a bit confusing, however, after analyzing steps and members involved, everything falls into place.
According to this attribution process workflow:
- Advertisers based on first and/or third-party data define the target and send it to the Identity Resolution Provider. Among the most popular identity resolution providers are LiveRamp, FullContact, and Wunderkind.
- Identity Resolution Provider, in turn, after anonymizing the data, addresses the target list to different inventory partners (Virtual Multichannel Video Programming Distributors (vMVPDs), OTT publishers, and OTT devices) and DSPs.
- Results of the ad campaign based on data from the ad servers, publisher logs, and Automatic Content Recognition (ACR) data from smart TVs, which is considered one of the most revolutionary ways to measure viewing habits, are directed to the Attribution Report Solution Provider.
- Attribution Report Solution Provider analyses the OTT data and, basing on required KPI, generates a report. The report may also be put together by either the data source provider, identity resolution provider, or an additional third party.
- After the report is ready, it is sent firstly to the inventory provider, who has to ensure the quality, and then to the advertiser or agency.
To measure attribution in the CTV environment, there is also a need to match the exposure data to the household. In this case, the video ad spot is served to households based on target parameters set by the advertiser (like demographic, behavior, interests, geography, and similar) or a more general contextual targeting. By means of third-party tracking pixels, either your media partners or ad server collects the exposure data like the device’s advertising ID, user ID, IP address, timestamp, and others. The more data is collected the more neatly the exposure data will be matched to the household.
As we’ve already mentioned and as stated by the IAB, it is just a high-level overview of the process and there may be exceptions to this ‘typical’ path and that companies may serve multiple roles. Let’s review some examples from real life and see how OTT and CTV attribution providers explain their workflows in practice.
For example, LeadsRx, a marketing multi-touch attribution platform, recently added support for OTT advertising. They use Universal Conversion Tracking Pixel installed on the advertiser’s website. After the OTT ad spot is displayed, LeadsRx gets ACR and advertising data. Based on the place and the way the ad is consumed, the anonymous identification data may involve ad ID, IP address, channel, timestamp, and so forth. After LeadsRx receives this data and evaluates it across all the marketing channels, it can provide the user with attribution metrics of the OTT campaign.
Adjust, another marketing attribution provider, applies SDK and Server-to-Server solutions (S2S) to help advertisers measure app installs from ad campaigns run on CTV devices. When the user views or clicks on the ad spot, Adjust, by means of S2S, receives the ID of the device or user. After the OTT partner registers this activity, it communicates this activity and confirms the attribution with Adjust.
Connected TV attribution for CTV apps by Adjust
As you can see the OTT and CTV attribution workflow does not have to be complicated, one of the main goals it should have is to meet your business objectives. There are a number of attribution providers now on the market and each of them proposes its own approaches to attribution and set of features.
However, this market is not yet overcrowded, so if you need a system to meet your custom advertising goals or you plan to invest in an innovative marketing attribution platform, the MindK team has the knowledge and experience to make it happen.
Connected TV and OTT attribution are an emerging technology, that also has some challenges worth mentioning, not only with the purpose to warn you, but mostly to point out the directions that badly require innovations.
#1 Identity resolution
OTT and connected TV are very fragmented with a number of devices and channels operating in a cookie-less environment. Identity remains one of the main challenges until now. While most devices have Identifiers for Advertising (IFAs), some don’t support it, which complicates identification throughout the attribution workflow.
When IFAs are not available or not used, IP-based matching can be applied, but since IP is not a constant identifier, it sometimes might overstate reach and understate frequency. In addition, as we’ve already discussed, CTV attribution involves matching at the household level, not an individual one.
There are a number of international, federal, and state laws and regulations, which impose constraints on sharing user-level data. New compliances are also expected to be focused on Advanced TV.
For example, recently the ePrivacy Directive broadened the scope of applications that fall under it. Now OTT services are covered in the ePrivacy Directive, although prior they were not mentioned.
Some regulations don’t directly address OTT services, however, they should not be overlooked. For example, in spite of the fact that the California Consumer Privacy Act (CCPA) and General Data Protection Regulation (GDPR) do not cover over-the-top content, ignoring privacy concerns may be risky.
Another unobvious example is Children’s Online Privacy Protection Rule (COPPA). Since advertisers track CTV attribution at the household level, it may cause potential COPPA problems, as children consume video content along with adults.
Nevertheless, there are AdTech companies that managed to build a business on this challenge. Take, for example, OneTrust PreferenceChoice that launched a CTV and OTT consent management platform that helps advertisers and publishes comply with CCPA, GDPR, and dozens of other laws and frameworks from the IAB and Digital Advertising Alliance.
CTV and OTT consent management platform
#3 Data collection and measurement
Due to the fact that attribution of Advanced TV is an emerging technology that is gaining strength, sure enough, it has certain challenges with data collection and measurement.
For instance, not all publishers in the CTV/OTT ecosystem allow third-party tracking. Moreover, due to the fact that CTV provides the data at the household level, it is almost impossible to collect more data about each user and track co-viewing of ad spots.
One more problem here is the absence of standardized measurements and attribution that may result in data distortion and wrong decisions. IAB in collaboration with other organizations like DAA, Coalition for Better Ads (CBA), Media Rating Council (MRC), and others, have a strong intention to develop and distribute a unified approach across the advertising industry. So, fingers crossed!
Develop a custom marketing attribution platform with us
Attribution and measurement in Advanced TV are developing at a breathtaking pace. The CTV\OTT attribution ecosystem possesses a striking potential for those AdTech companies that will venture to solve the current challenges and help advertisers optimize their advertising efforts.
Here at MindK, we work to empower businesses through technology. We have experience in developing robust marketing attribution systems that process large amounts of data and help advertisers analyze and optimize ad spending. Check out our Ad Software Development Services to see how our team can help your business.