Amazon Marketing Stream: Near Real-Time Ads Data
Amazon launched new data to supercharge your online marketing strategy
As of June 2022, the Amazon Advertising team released the Amazon Marketing Stream service beta. The service delivers near real-time performance data to advertisers. The Amazon Marketing Stream beta is only available to registered developers.
Rather than call the Amazon Ads reports API, the Amazon Marketing Stream delivers push-based data feeds multiple times daily. Direct access to the advertising stream helps avoid API throttling that can occur with reports.
The new service will allow those on Amazon Seller Central, Amazon Vendor Central, or Agencies to unify and analyze hourly feeds of advertising data in a private, trusted data lake or cloud warehouse, allowing you to use tools like Tableau, Power BI, Google Data Studio, and many others.
What Are The Benefits Of Amazon’s Marketing Stream?
Amazon Marketing Stream allows Amazon Advertisers to understand better how campaigns perform by pushing the delta values instead of aggregates. This helps you see how campaign performance changes with the target audience over time and allows you to refine your digital marketing strategy efficiently. The new Amazon Marketing Stream API service offers benefits when compared with API-based reporting:
- Improved data granularity: Traffic and conversion data are summarized hourly, compared with daily granularity if you use the reporting API. The data is streamed automatically to your AWS account, which you’ll provide when subscribing to the Amazon Marketing Stream data set.
- No throttling: With Amazon Marketing Stream data streaming, API users no longer need to make frequent API calls for intra-day updates. This helps to prevent throttling while also increasing data accuracy.
- Near real-time messaging: Get notified about budget consumption, such as when campaigns run out of budget.
- Deep Dive: Show all dimensions available in a single dataset, such as performance at the target ID, placement, or ad level. This includes granular data such as keyword performance for a given placement.
What Types Of Data Is Available In Amazon Marketing Stream?
Currently, Amazon’s Marketing Stream includes two categories of data: reporting and messaging.
- Reporting data: Reporting includes an advertiser’s processed (non-raw, summarized) performance data, including traffic and conversions sent hourly and data dimensions necessary for analytics.
- Messaging data: Amazon sends real-time notifications for state changes and other events. Currently, this messaging includes budget consumption and will notify you whenever your budget allotment changes by 5% or more. We plan to expand the messaging to include notifications about product eligibility, bid recommendations, and other events.
How Can Amazon Marketing Stream Retail Sales?
AMS allows various digital marketing disciplines to fuse performance data from media investments with orders, returns, inventory, and other Selling Partner feeds.
All the retail, reporting, and messaging data is unified in a private data lake or cloud warehouse technologies like Amazon Redshift, Amazon Redshift Spectrum, Google BigQuery, Snowflake, Azure Data Lake, Ahana, and Amazon Athena.
Turbocharge reporting, analytics, and business intelligence tools like Google Data Studio, Tableau, Microsoft Power BI, Looker, Amazon QuickSight, SAP, Alteryx, dbt, Azure Data Factory, Qlik Sense, and many others.
With data unified and using your favorite analytic tools, you can start to ask questions like;
- Do we have an effective ecommerce marketing strategy? How can we optimize based on AMS?
- What impact on e-commerce sales are we seeing based on our optimizations?
- Do we need to refine our broader digital strategy? Do we need to adjust messaging or create off Amazon in our social media marketing campaigns?
- Do our marketing goals align with the performance we see in our Amazon business? Any adjustments to the mix in our online marketing channels
What Are The Requirements For Amazon Marketing Streams?
- Status: Beta
- Campaign type: available for Sponsored Ads & DSP (see below).
- Region: Available in North America only (Canada, Mexico, United States)
- Cadence: Hourly
- Technology: You must have an Amazon Web Services account and a data lake or cloud data warehouse to store all the data
- History: Since this is a real-time data stream, there is NO historical data. Data starts flowing from the point of activation. Amazon does not support replaying events that happened in the past.
- Data volumes: The volume of information is significant. As such, you may have increased storage costs and query costs.
Understanding The Costs Associated With Amazon Marketing Streams
The Amazon Ads team does not charge for the service. However, if you are considering AMS, you must factor in how the scope and frequency of Amazon Marketing Streams data may impact overall costs.
Like any near-realtime data feed, streams mean more data is coming at you quickly. All that data has to be stored somewhere. A cloud warehouse or data lake is required.
The warehouse or the lake will constantly have new information for processing and loads. Near-real time means your system will always be “busy” processing and loading data. This activity has costs.
For example, let’s assume you use Snowflake. Snowflake’s pricing model is based on two consumption-based metrics: usage of computing resources and data storage. The more Snowflake is “active,” the more your consumption-based costs increase accordingly.
Let’s compare the cost variation between typical Amazon Ads reports and Streams;
- API Reports: API reports are processed daily and use about 60 minutes of computing time per day. The reports total about 200 GB of data per month. Your Snowflake costs will be about $140 a month.
- Near-Realtime Streams: Streams are processed every hour, every day, for about 5–10 minutes. The amount of data generated is about 2 TB of data each month. Your Snowflake costs will be about $550 a month.
While this is a hypothetical example, it is meant to highlight choosing to use AMS will have cost implications.
Don’t want to use Snowflake? Even if you decide to use Redshift from Amazon Web Services or BigQuery from Google Cloud, the same cost and performance considerations apply. They may have slightly different implications, but they exist nonetheless.
The key takeaway is that the streaming service should not be viewed as a different type of report. The velocity and scale of the data far exceed your regular reports. As a result, even though Amazon Ads does not charge for AMS service, you should be mindful of the broader total cost of ownership when using near-real-time data service.
Amazon Marketing Stream Datasets
Currently, there are three core datasets that are part of streams;
- NEW: Amazon DSP campaign management: Provides a snapshot of all your Amazon DSP campaigns, flights, ad groups, and targets. Link
- NEW: Sponsored ads campaign management: Provides a snapshot of all your campaign objects — including Sponsored Products, Sponsored Brands, and Sponsored Display. Intra-day campaign management and optimization. Building responsive applications to drive optimization. Link
- NEW: Sponsored ads budget usage: Provides the budget status for sponsored ads (Sponsored Products, Sponsored Brands, Sponsored Display) campaigns and portfolios. Link
- Sponsored Products traffic (sp-traffic)
- Sponsored Products conversions (sp-conversion)
- Sponsored ads budget usage (budget-usage)
- Sponsored ads campaign recommendations: These contain recommendations to optimize existing ad campaigns or create new ad campaigns to improve the performance of your sponsored ads portfolio. Keep local reporting in sync with Amazon systems with near real-time insights. Link
- Sponsored Products performance: Provide details about traffic and conversions for Sponsored Products campaigns. Develop push notifications within your tool’s interface. Link
- Sponsored Brands performance: Contains click and conversion metrics related to Sponsored Brands campaigns. Link
- Sponsored Display performance: Contains click and conversion metrics related to Sponsored Display campaigns. Link
We go into a few of these below, as currently defined by Amazon. Please note these are subject to change, given the service is in beta. The marketing streaming service applies to Amazon shoppers (not off Amazon) and those merchants using Seller Central or Vendor Central.
Sponsored Products traffic (sp-traffic)
Offers key Amazon Ads campaign metrics. The sp-traffic dataset contains click, impression, and cost data related to Sponsored Products campaigns. Initial click data is available within 12 hours of the ad click. As part of the traffic validation process, clicks can be invalidated during the following 72 hours after the initial report. Amazon Marketing Stream will automatically deliver any adjustments to your queue.
The total cost of all clicks is expressed in local currency.
Sponsored Products conversions (sp-conversion)
The sp-conversion dataset contains conversion data attributed to Sponsored Products campaigns. Conversions for Sponsored Products campaigns are reported based on the hour when the click they are attributed to occurred.
Conversion data is reported daily, weekly, and monthly, and you should expect to receive revisions to conversion data up to sixty days after the initial click.
The total number of units ordered within 30 days of ad click where the purchased SKU was the same as the SKU advertised. Will assist with analysis of ad spend and impacts on net sales.
Sponsored ads budget usage (budget-usage)
The budget-usage dataset provides the budget status (spend compared to budget) for sponsored ads (Sponsored Products, Sponsored Display, or Sponsored Brands) campaigns and portfolios. Advertisers receive notifications at every 5% interval of budget consumption.
Amazon Marketing Stream pushes budget usage updates in near real-time. You should receive a notification soon after your budget usage increases by 5%. Due to Amazon’s traffic validation process, you may receive the same notification twice. For example, invalidating a click may reduce your budget usage under the 5% threshold.
Summary
With the new service, you can access all of the data related to your product ads. As mentioned, the Amazon Marketing Stream “pushes” updates rather than the current “pulling” of report snapshots.
While Ad reports are still critical to understanding performance, streams offer additional views into how well your media investments impact marketing plans and broader business outcomes.
Here are just a few of the benefits of code-free, convenient access to the data will enable:
- Improve operational efficiency by using real-time information to understand the campaign budget, extend campaign reporting with hourly performance metrics, and timely insights to fuel deeper campaign optimizations.
- Automated, code-free API access ensures you get direct campaign data sets without the messy or complex technical operations of building your own Amazon Ads API integrations.
- Create intra-day insights, daily performance, or real-time analytics snapshots of campaign performance metrics that support advanced campaign optimization efforts.
- Unify eCommerce campaigns on and off Amazon to get a complex picture of campaign efficiency and campaign impact linked to your campaign goals.
- Increase the velocity in understanding conversion metrics, calculated conversion rates, or conversion potential for different audiences and campaigns.
Get a 30-day free trial so that you can try Amazon Marketing Stream API automation for yourself.
Openbridge is an Amazon Ads verified partner, Amazon Selling Partner, and part of the AWS Partner Network. We support the complete suite of Amazon Sponsered Advertising reports, campaign metrics, and performance data from the Amazon Demand-side Platform (Amazon DSP) and Amazon Attribution. If you are a Brand, Digital Agency, or something in between, we offer a free 30-day trial so that you can unlock your data.
The Openbridge Amazon automation for Advertising and SP-API data pipelines deliver metrics that matter via fully automated feeds.
Fuel your insights and BI dashboards with direct connections to your seller data. This means no technical resources to manage implementations, no coding, and no need to worry about API versioning or updates.
