Is Less More when analyzing your Revenue data?

Updated: Apr 19


Summary.


Like most of the people, I check my emails several times within the hour, spend 4+ hours on my smartphone, interact with 10+ apps and multitask 40% of the time.


Under normal circumstances (pre-COVID) these activities would slip through the cracks as they blend in quite nicely with my social life. But under the current state of curfew we live in, they can easily wear you down mentally.



This is how I started reading “Digital Minimalism” by Cal Newport who proposes a 30-day plan for digital declutter based on 3 core principles:


  • Choice and intention. You’re still using technology, but only what you want and only what connects to your values.

  • Optimization What you allow into your life needs to work for you.

  • Acceptance that you can’t be everywhere all the time.

In the era of Big Data, this philosophy [data minimalism] is also gaining significant traction among companies at a global level. These companies have excelled in collecting and squeezing more value out of data. As a result, 90% of the world’s data has been created in the last 24 months.


As Publisher side professionals, we are constantly reminded that [first party] data is the most critical asset for pretty much any kind of company, however unlocking the full value of data is easier said than done when you work with 10+ tech vendors shooting your way with all sorts of data, that can range from being unstructured to over structured.


At yieldPass we believe that when it comes to revenue data, More is More. However we also believe in the principles of Digital and Data minimalism to optimize the usage of revenue data.


In this post we are taking a shot on how to apply the core principles of Data minimalism to maximize the impact of Data Analysis on your monetization efforts.


Problem Statement.


  1. A publisher deals in average with 7+ isolated and hardly interconnected data sources (Ad Server, SSPs, Analytics, etc).

  2. The more information the more noise. Lack of standardization or naming conventions lead to data normalization issues and discrepancies.

  3. Costs associated with collecting and analyzing data.

  4. Less than 20% of all available data is used. Just think of all log level data on impression level that could be extracted from an SSP that you don’t even know it existed (Bid price, bid rate, win rate, viewability score, ad unit id, connectivity and banswith, OS, device type, etc).

  5. Rigid dashboards won’t provide the flexibility you need to visualize data your way. As a result you can not understand your data.

  6. B.I. solutions can be complex and require substantial manual customization.


Solution & Methodology

  1. Choice & intention. Create a plan to collect and consolidate your data. You can’t analyze your data in silos, you need to bring it all together to make sense of it. To avoid high costs and complexity select only the data sources that are relevant to achieve your OKRs or similar framework. Learn more about OKRs.

  2. Optimization. Inaccurate data equals no data. Ensure your data is normalized, clean and properly instrumented by harmonizing conventions, formulas and definitions used by all your tech vendors. (e.g. some SSPs distinguish between impressions and rendered impressions, ad calls and ad requests).

  3. Customization. Find ways to visualize only the data you need, when you need it. We said before that we use less than 20% of the data available to us. By customizing the way you consume and analyze your data you can unlock a higher % of usable data.

  4. Harmonization. Align your data analysis with your business objectives and key results by making sure you can easily filter and grant access to your teams to specific data sets relevant to their business units.


Execution.

Our B.I. solution was built under the principles of data minimalism and will allow you to execute the proposed solution through following features:


  1. List down all your data sources and relate them to the layer of the monetization funnel they belong to. Learn more about Monetization Funnel here. Also include the business unit accountable for that data set.





2. Shortlist your data sources by the value they provide towards achieving your OKRs or similar goal-setting framework.


3. Once you have a shortlist of the data sources relevant to your goals, identify the data ingestion mechanisms available to each source. Here are the things you need to consider:

  • Ability to export data via API from your different data sources. Contact your tech provider to know if they have an API and if you can have access to it. Some SSPs make their APIs available upon meeting certain volume thresholds.

  • Ability to export data via CSV files from your data sources.

  • Ability to schedule CSV files to be sent out automatically by your data source.

4. Now you need to level the playing field or at least know where are you standing in terms of how data is presented to you. This point has two angles to be considered:


a. Formatting:

  • Ability to change currency formats by your data source.

  • Type of decimal and thousand separator format use by each data source.

  • Type of date, time and duration formats available.

b. Naming conventions:

  • Naming conventions for metrics (e.g. revenue/earnings, ad calls/ad requests, CPM/RPM, Unique Users/Visits, etc)

  • Naming conventions for dimensions (e.g. Ad sizes: 300x250 / 300_250, Device: Mobile/Smartphones, Geo: Europe/EMEA).




5. Create customized dashboards (as many as you need) to visualize your data through highly configurable charts that tell you what you need to hear in a fast and simple manner. Why settling with one predefined Dashboard, when you can have as many as you want, the way you want them. Every morning you pull a report and sit in front of your computer staring at the loading circles. If you need to add another variable, metric or dimension you will have to repeat the operation over and over again. Even though your paying a high fee for your B.I. tool, you are still spending most of your morning updating a dashboard. Customization not only saves you time, but also gives you an eagle's eye view of your business. Try smart dashboards now.







6. Having the eagle's eye is great but you also need to be able to dive deeper. To achieve that you can group and categorize collections of Charts under separated views customized for every team. Create views to track Ad Monetization performance, Cost of Acquistion performance, Direct Sales performance, Header Bidding performance, etc and take your data analyzes to a whole next level.


Useful Resources.


- Use yieldPass Smart Dashboards to unify all your data sources in one place.

- Use yieldPass Surveillance Software to set customized alerts on the sets listed on the Process & Execution section above.


15 views0 comments

Recent Posts

See All