Description: Advertisers have long heard about the power of DCO, but thinking about and executing the right strategy isn’t always obvious. We’ve developed a couple of quick questions and guidelines to help ensure that you deploy the most effective trial for testing the viability of dynamic creative. This article will cover:


Questions To Frame Your Strategy

  1. Do you have existing audiences or targeting? You’re likely already using some data to segment or target your audience via your media buy. One of the easiest ways to layer in DCO is to match that targeting with personalized messaging. Have a complex targeting strategy? Test out what’s simple first - can that audience be grouped into a few more general categories?
    • Example: If you’re already targeting a segment of young professionals vs. new parents, you should be using messaging that aligns with the differences between those two audience groups.
  2. What’s the objective, and how will success be measured? This may sound like putting the cart before the horse, but if we know what action we want to drive, experiences can be reverse-engineered to achieve that result. 
    • Example: If awareness is the goal (CTR, completion rate) focus on broader personalization that adds relevance to the moment. This could be as simple as weather data signals to show a hot beverage on a cold day. For lower funnel activity (revenue, conversions), lean into strategies that directly inspire purchases like a product feed featuring up-to-date price points.
  3. What assets do you have available? This can be used to inform what’s easy for you to execute. If you only have one or two main assets, Innovid can personalize both video and display experiences to the viewer through changes in copy or imagery - don’t feel like you need multiple pre-made ads to run DCO. Conversely, if you already have multiple versions produced, you can think about how those match up to a targeting strategy without having to rebuild. Additionally, with the growth of generative AI, platforms like ChatGPT and Midjourney can help you generate the assets needed to version.
    • Example: A standard :15 or :30 second video spot can be personalized by creating additional real estate around the video asset which can then feature dynamic product images, offers, copy, and CTAs.
  4. What signals do you have available? If you’re already collecting 1st party data, this is a great place to start for dynamic messaging. Generally, we recommend using data that relates to consumer interests or past activity rather than specific user information like someone’s name - which can be creepy. If you don’t have 1st party data - signals that are commonly available and easy to activate are time/day, weather, and geolocation.
    • Example: 1st party signals like past purchases or site activity offer better dynamic opportunities than anything that could be viewed as PII. 


Parameters For A Test

After thinking through the above, we recommend running a test that starts small. The reason we suggest this is because DCO shouldn’t live and die on a single campaign and should be an evolving effort. Below is a general suggestion to be changed per your needs:

  • Impressions: The more, the better here because smaller sample sizes = greater volatility of results. We recommend running at a minimum of 10,000 impressions per day for each creative version you’re testing.
  • Versioning: 2-5 different creatives that are differentiated by 2-3 elements. 
  • Data: We recommend using one data strategy on your initial campaign to keep mapping your strategy simple to start.  
  • KPI: Are you using DCO for creative or trafficking efficiencies, to personalize messaging to users, or something else? It is best practice to identify what the success of a test looks like and tailor your campaign to that.


Glossary (DCO Strategy Terms)

General Terms

Dynamic Creative / Personalization - The automatic generation and customization of ad content based on various data signals and consumer interactions. The goal of dynamic creative is to deliver personalized and relevant advertising experiences to consumers.

Creative Optimization - The automated process of selecting the best-performing creative or creative elements for your specific KPI and increasing the distribution of that particular version or element.

Data Feed - The data set being used for a particular dynamic campaign which could come in a variety of formats or delivery mechanisms (e.g. CSV file, XML file, API connection, JSON feed, Google Spreadsheet)

Decision Tree - A visual tool to map overall strategy, data points are shown as branches of a chart that connect to the different creatives associated with each. 

Data Signals

1st Party Data - Data sources provided by the brand. 

3rd Party Data - Data sources that can be purchased from an outside vendor.

API (Application Programming Interface) - A program that allows applications and systems to communicate with one another. Building an API connection allows Innovid to automatically communicate with data sets that might update in real-time. (Example: Sports scores or stock prices.)

Audience Segment - A subgroup of consumers defined by an advertiser based on criteria such as product usage, demographics, psychographics, communication behaviors, and media use. 

Geolocation Data - The identification or estimation of the real-world geographic location of an object, such as a mobile device or computer, using digital information. In digital advertising, geolocation is often used to target viewers based on their physical location, allowing advertisers to deliver relevant content or advertisements based on a viewer’s current or past locations.

Publisher Macro - Dynamic variables or placeholders used in ad tags or creative templates that are filled in by the publisher's ad server with specific values at the time of ad serving. These variables can include information like the viewer’s geographic location, device type, or any other relevant data. Publisher macros enable the customization of ad content based on various parameters, providing a personalized experience for the audience.

Retargeting - Also known as remarketing, is a digital advertising strategy that involves displaying ads to viewers who have previously interacted with a brand's website or mobile app. It works by using cookies or other tracking mechanisms to identify viewers and show them targeted ads across different websites or platforms. The goal is to re-engage viewers who have shown interest in a product or service but may not have completed a desired action, such as making a purchase.

Technology Data - The various devices used to deliver and consume online content. This includes web browsers, mobile devices, operating systems, and other technologies that influence how ads are displayed and interacted with. Advertisers may consider the technological landscape to optimize their creatives for different devices and platforms or to inform viewer-specific preferences like language. 

Time/Day Data - Delivering ads based on specific times of the day or days of the week. Advertisers can schedule their campaigns to reach target audiences during optimal times for engagement or when specific events or promotions are taking place. This strategy allows for more precise and timely delivery of ad content to maximize its impact.

Weather Data - Adjusting ad content based on current weather conditions in a viewer’s location. Advertisers can use weather data to tailor their creatives to match the current weather, promoting products or services that are relevant to specific weather conditions. For example, an ad for a clothing retailer might showcase warm clothing during cold weather or beachwear during hot weather.

Was this article helpful?
0 out of 0 found this helpful