It’s Official: Fivetran and HVR Are Now One. Learn more.

Why Ad Agencies Should Leverage a Modern Data Stack
Why Ad Agencies Should Leverage a Modern Data Stack

Why Ad Agencies Should Leverage a Modern Data Stack

Agencies can use data to show why and how campaigns succeed.

By Greg Wheltle, Max Holley, May 11, 2020

Marketing is tasked with responding to unpredictable times by modifying content and messaging. To monitor the success and impact of marketing decisions and make timely yet informed decisions, data is critical.

A modern data stack, composed of an automated data pipeline, data warehouse and business intelligence tool, enables quick access to the accurate data that marketing requires. A data pipeline ensures that data from disparate sources is automatically integrated into a data warehouse, while a BI tool sitting on top of the warehouse enables businesses to create visualizations, track performance and make projections.

Generally, agencies make marketing decisions for their clients and, to prove their value, track the performance and impact of these decisions to present back to the client. The modern data stack ensures that you’re presenting your clients with accurate, up-to-date data.

Getting at the Whys Behind Campaigns

Reporting on campaign performance is mandatory and expected by clients — but it’s table stakes. What sets an agency apart is the feedback that explains why things happened the way they did. Your ROAS was up 27%? That’s great, but why was that campaign so successful? Your campaign had a low CPC? Why? The whys of campaigns are uncovered and validated by drilling more deeply into the data, allowing you to spot trends and identify nuances within respective data sets. Providing your clients with information shows that your agency is a value-providing partner, rather than a machine just reporting figures.

There are many marketing-focused data extraction tools in the market. Some allow you to extract data from over 500 marketing sources, and it’s appealing to be able to pull data from all the applications your clients are running advertising across. These tools, however, give you marketing statistics that amount to surface-level reporting. They don’t get at the heart of the why. To get that, you need cross-platform reporting and measurement that is tied back to the actual business impact. This requires access to first-party data from your clients’ databases and CRM systems. That data can be difficult to access directly, which is why many marketing-focused data tools stay in the marketing lane. Databases and CRM systems require different permissioning and/or have complex APIs — significant barriers to data extraction. This is where Fivetran comes in.

Fivetran has prebuilt connectors for a variety of data sources, including marketing APIs, databases and CRM systems. We pull data from these diverse sources to help you get a richer picture of your clients’ data and uncover the whys associated with their respective campaigns.

End-to-End Platforms

Some platforms operate as end-to-end data stack solutions, providing a data extraction tool, a warehouse and a visualization piece all in one. This can be attractive for a number of reasons. You only deal with a single vendor that provides agency-focused analytics out of the box in a matter of days. There are a number of advantages to these solutions, ease of use chief among them, but you should carefully consider several factors before implementing them.

1. Weak Connectors

Many all-in-one platforms operate as a bit of a black box, limiting access to the data. How do you know what is being brought across from your sources? If a platform doesn’t document what it’s pulling from sources, how do you know you’re getting all the available data?

Also, keep in mind that you can grow only as fast as the connectors pulling the data grow. One-stop shops have fairly rigid APIs. Much of the product development focus for these companies is building good-looking reports that are client-facing, not building robust data connectors. Say someone clicked on an option for a membership or trial. Your out-of-the-box connector will fail because it doesn’t even know to look for the additional data. Fivetran, on the other hand, natively grabs that schema change. In other words, it’s better to have everything and not need it than to lack something you realize you need later.

2. Security

Who owns your clients’ data? If the vendor is providing a warehouse for you, do you own your clients’ data if they leave your agency? What if you leave the vendor for another provider? Does the vendor hold onto this data? How would your clients feel about this? Do clients want to hold onto their own data and keep it within their own walls?

With a modern data stack, you house and own your clients’ data within your warehouse. If clients want to keep data within their own warehouse, Fivetran allows you to direct it to their destination. From a security perspective, this is preferable, and in some cases mandatory. How comfortable might clients feel if they’re sending you data that an unknown vendor controls or owns?

3. Reporting Only

At the end of the day, all-in-one platforms are designed as a quick and easy way to start reporting, but clients expect much more than that from agencies. You’ll provide true differentiation and value to your clients if you can tell the richer story of why campaigns succeed or fail. What is driving transactions and sales? Clients don’t want to see the same reports they got from their last agency partner, just with a new logo at the top. They want to partner with agencies that can generate insights unique to their business.

Why Fivetran?

Fivetran is an automated data ingestion platform that connects to 170+ data sources natively. With a five-minute setup and no coding experience required, you can easily start collecting and integrating client data from various sources. You’ll be able to tell a powerful, data-backed story about your campaigns.

We’re a fully managed service, meaning we handle all API and schema changes that occur in your data sources. The APIs of social connectors change constantly, but you won’t need a data engineering team to update these changes with every new release. We also handle custom objects within most sources, so if your clients have unique fields within their DBs or CRMs, we will bring those over automatically. No need to wait for updates from us to add these fields to your warehouse.

Our Powered by Fivetran platform is designed for agencies to manage their clients’ data in an organized manner. You’re also able to leverage our API, which makes the onboarding of new client data seamless. And since you can direct data to multiple warehouses, you can extract and aggregate data from client sources and push it to their own warehouses. Clients can even use their own BI tools on top of their data.


The agency space has faced major headwinds over the last several years, as margins have shrunk and the need to constantly evolve has become the norm. With the media landscape continuing to consolidate, more agencies are looking to differentiate themselves by becoming a true business partner to their clients. For many, this has become an imperative for overall growth. Some have adopted a data-driven stance only to fall short with reports and slide decks that clients have seen many times before, from many other agency teams. To truly drive value, agencies need to move away from surface-level performance metrics by implementing a modern data stack. This will allow them to show clients that they are data-driven, instead of just saying they are.

Start analyzing your data in minutes, not months

Launch any Fivetran connector instantly.