Deep Root Blends Ad Buy Data with Past Impressions to Predict Competitor Ad Exposure

[RamanK]

Using their proprietary methodology, Deep Root Analytics leverages data to uncover media-buying insights for their clients.  Deep Root analysts, like Danielle Mendheim who recently joined us on a webinar, prep and blend extremely large, multi-sourced datasets, analyze them, and then produce dashboards and reports for their clients to more wisely assess their media buying options. This process involves bringing together historical set-top box data on TV viewership, and combining with the future TV ad buys as documented by the FCC. By combining these data sets, they can project how many impressions political campaigns can expect to achieve. However, producing such critical insights hasn’t always been a seamless process for the team at Deep Root.  A part of the Deep Root “secret sauce” includes having the right technology in place, and for them, this now includes Alteryx, Amazon Web Services and Tableau.

4 Analytic Hurdles Tackled by Alteryx and AWS

• Volume: Deep Root Analytics handles a lot of data, upwards of 20 billion rows of viewership data alone! As you can imagine, 2 terabytes of data does not easily pass through individual CSVs, nor does it lend itself to easy data blending prior to analysis. Together, Alteryx and AWS made it easy for Deep Root to put the power of cloud computing in the hands of their analysts and to leverage all of their data – local or cloud – thus eliminating data blending issues associated with volume.
• Accuracy: Producing accurate insights means utilizing the correct datasets. Prior to using Amazon Redshift, Deep Root’s analysts had to spend time ensuring that the most-up-to-date data was being used as various versions of datasets, i.e. v1 – v10, were often floating around with different team members.  With Redshift’s cloud-based data warehouse and the ability of Alteryx to connect to it, Deep Root quickly eliminated wasted time and frustration. The team now confidently delivers accurate insights to their clients.
• Speed: Many of Deep Root’s clients are in the political sphere, which requires quick turnaround.  With the speed of Amazon Web Services and repeatable analytic workflows from Alteryx, Deep Root is able to deliver insights in the timely manner the political arena requires.
• Automation: Like many analysts, those at Deep Root had daily tasks that required 5-10 minutes per client, per day.   While 5-10 minutes here or there may not seem like much on its own, the Deep Root team realized that it quickly adds up when you multiply it by 5 days a week, roughly 21 business days a month, and 12 months a year.  Danielle and her colleagues now work smarter and utilize the Alteryx Scheduler to run these recurring processes overnight on EC2 and then automatically update their AWS Redshift tables.  By running processes overnight, the team can also use bigger Redshift clusters in their data analysis without being slowed down.

 

Tips and Tricks

In this webinar, Danielle shares a number of insights on how to best work with Alteryx and AWS.

• Scheduler – Danielle like to run Alteryx Scheduler so she can push much of the work to hours while she is offline.
• Macros – Danielle run a series of nested macros that run sequential steps. The advantage of nested macros is that they enable a step to finish before moving on to next step. In her example, data can finish being loaded into S3 before the next macro that copies it to Redshift is invoked.
• Run Command – Danielle uses the Run Command for a number of reasons, including the fact that she can dynamically assign filenames.
• Dynamic Input Tool – The team at Deep Root Analytics loves to use the Dynamic Input tool to quickly filter out unneeded data.
• Using the Alteryx Direct Input tool from Redshift tables - This tool enables you to maintain your data types, and do SQL type manipulations such as the Direct Select statements and Where clauses, directly in the Input tool.
• S3 Upload and Download Nodes – These are faster and don't require local disk space.

 

Overcome Your Analytic Obstacles:

If you face data issues similar to what Deep Root experienced prior to incorporating Alteryx and Amazon Web Services into their proprietary secret sauce, download your free 14-day trial of the Alteryx Designer and experience how you too can quickly blend and analyze data from all sources, apply advanced analytics and enable your data analysts with the power of Amazon Redshift. Download the trial to start expanding your analytic abilities with Alteryx and AWS today.