Five ways to solve the marketing data challenge

No. The “marketing data challenge” is not the latest dance trend on TikTok, but rather a tricky and intricate balance that performance and digital marketers face. 

You see, the sheer volume of data and marketing analytics that can be collected from an ever-increasing number of platforms is enough to boggle the mind. In fact, less than half of all digital marketers trust the data they're working with. 

Adding to this crisis of confidence, and backed up by Funnel’s own VP of growth, is the fact that marketing data is being used to inform and steer high-level strategic decisions for entire businesses. 

Digital marketers need to gain control and solve the challenges of data-driven marketing. But how?

Tactically speaking, there are four main ways to approach your marketing data — from basic solutions for beginners, to dynamic tools that are ready to scale. 

1. Manual copy and paste
2. Plugins
3. Building a data stack
4. An all-in-one tool
5. A marketing data hub

Let’s examine each, so you can determine the best approach for you. 

1. Good old copy and paste

Sometimes, nothing beats manual data entry — especially if you’re just starting out in your career. By manually exporting the performance data from each of your platforms, then pasting it into some sort of spreadsheet software, you can gain a real hands-on perspective that can be essential for good analysis later on. 

An example

We can very easily envision this methodology by thinking of a straightforward digital marketing example. Let’s say you run digital ads across Facebook, LinkedIn, and Google Ads. In order to start making sense of your entire digital advertising ecosystem, you’ll want to move your analysis out of each platform. 

To do that, you’ll need to export the performance data from each of the three platforms and paste it into a spreadsheet,. 

From here, you’ll need to perform some manual data transformation and cleaning to ensure that you are comparing “apples to apples” across the multiple tabs. (For instance: making any necessary currency conversions) Then, you can start performing some basic data analysis. 

The limitations

As you can probably guess, this solution doesn’t really set you up for long-term success. There are a couple reasons for this. First, the more times you “touch” your data, the more possibility for human error you are entering into the broader equation – thereby leading to an even greater marketing data management challenge. For instance, you might not export all of the intended data, a value or row may be truncated when pasting into the spreadsheet, or a whole host of other unintended consequences may occur. 

Also, while manually copying and pasting into a spreadsheet may provide some immediate help, think about how you might approach complex marketing operations – or even how or an agency may need to approach the marketing data challenge. That single spreadsheet would get really “heavy” quickly. Plus, as the data becomes more complex with more platforms and campaigns represented, the risk of formal and sheet breaks increases almost exponentially. 

However, manually copying and pasting data can still be employed by the most advanced digital marketers — particularly for ad hoc analysis. So, don’t dismiss it entirely. Just don’t base your reporting stack on it. 

2. Plugins, baby

When your marketing strategy necessitates a move from manually copying and pasting data into spreadsheets, you may start venturing into the world of plugins to automate some of your work. Specifically, we mean plugins for spreadsheet and visualization tools. 

An example of how plugins work

To get a feel for how plugins can help you solve the marketing data challenge, let’s say that you want to visualize some of your data in a tool like Looker Studio. (Check out our Funnel Tip about the Looker suite here.) You also don’t want to be stuck continually copying and pasting all of your marketing data into a Google Sheet before connecting it to your Looker Studio report. Instead, you select one or several compatible plugins for Looker Studio from its data source library.

Depending on the plugin, they can automatically collect and import data from your various marketing platforms to Looker Studio. From there, you can build dashboards and reports as needed, and the data is automatically refreshed when you open the report or change date ranges. 

The limitations

There are three key limitations to plugin products: data retention limits, quota/rate limits and limited data cleaning & transformation capabilities. Let’s break them down one-by-one:

Data retention refers to how far back in time a particular platform stores your data. Up until the last couple of years, this wasn’t an issue most marketers would run into, since data-retention limits were either non-existent or very long. However, this has started to change. Facebook Ads have a 37-month maximum retention, Google Analytics 4 maxes out at 14 months for certain reports, and Amazon Ads limits data retention to just 60 days. Needless to say, storing data in an intermediate destination has become essential to maintain a proper, historical view of your data. 

Then, there are data quotas. In November 2022 Google Analytics started enforcing its Analytics Data API (GA4) quotas which led many Looker Studio users to experience issues with their reports. This is because (when you use a plugin) you are fetching data directly from the API, and platforms put limits on how much data you can fetch per hour or day. Once that quota is exhausted, there’s nothing to do but wait for it to be refreshed. That’s less than ideal if your report is used by other stakeholders or customers. 

Finally, plugins don’t offer the most comprehensive data cleaning and transformation capabilities. This can become a sticking point when you fetch data directly from the API, since it’s not necessarily presentation ready. Any data cleaning or data transformation that you need to apply to your data will have to be carried out in the report or spreadsheet. This means that the transformation and data cleaning logic is less transparent, and it’s unique to that specific report or spreadsheet. This means you have to perfectly reproduce any transformation logic for any new reports. 

3. Building a data stack

Data stacks are a series of technologies and tools that are assembled to manage the full flow and scope of your data management needs. To use a little metaphor, you can think of a stack of pancakes. Each individual pancake represents a different piece of software to manage a given data process. 

An example of a data stack

For an average data marketer, their data stack may use an extract / load / transform (ETL) tool to gather the required data from multiple sources. That data could be sent via the ETL to a singular data warehouse for storage to be later cleaned and transformed by a service like dbt. All of this data would then need a visualization tool for easier analysis and interpretation. 

The limitations

Yep. Data warehouse. While this approach gives you a large amount  of flexibility and the ability to mix-and-match tools for your specific applications, it will also require significant developer or IT resources to set up and maintain the stack so that data flows smoothly. 

In order to apply necessary data transformation, you’ll need one or several resources with expertise in writing SQL with domain knowledge of the data that is being worked on. Even the ETL tool may require expert involvement to ensure connections are up and running. 

The other limitation you’ll face here is how complexity continues to drive the marketing data challenge. Think back to our pancake metaphor. As you add more pancakes (our metaphorical tools) to the stack, it starts to get awfully tall and will eventually wobble. See, the more tools you need to employ, the higher the likelihood of breakages. 

And since this approach requires BI or IT experts, that may mean your ability to perform marketing analysis may be stuck in their endless queues of requests from other parts of the business. 

Additionally, if you need to use a few different tools in a single data stack, those costs can start to add up rather quickly. 

4. An all-in-one tool

To avoid five different monthly charges from different service providers to your data stack, why not seek out a single tool that can handle everything? One solution to rule them all! Muahahaha. 

Sinister laughs aside, many marketing organizations that have a lot of complexity built into their own data-driven marketing challenges may want a single tool that does everything they need. Imagine taking all of the individual abilities from that data stack and integrating them into a single powerful tool for your business. 

An example

One great, standout example of such a tool is Microsoft PowerBI. While it may be a data visualization tool, it is so much more. It allows teams across an organization to manage and analyze their data as needed. While BI teams often head the management of this platform, it can provide robust analysis tools for marketers that can integrate data from other parts of the organization. This can be especially powerful in companies that rely on marketing data to inform strategic business decisions — a core part of the marketing data challenge. 

The limitations

There are two glaring drawbacks for tools like these. First and foremost, these tools tend to be built to serve large and complex enterprises. That means their capabilities go well beyond the scope of the marketing department. These tools and their configuration often require sophisticated BI teams and developers. 

Just like complex data stacks, this can leave marketers at the mercy of the service queue if they are looking to create new dashboards or fix bugs. 

When we talk about tools designed for enterprise clients, we can usually also expect a higher price tag. It’s for good reason, though. These tools are incredibly powerful and complex. 

5. A marketing data hub

A what? Hear us out. 

The marketing data challenge presents us with reams of data from loads of sources. Marketers don’t have time to constantly clean and normalize the data. Instead, we need it ready for analysis so we can find those golden nuggets of insights. We need a reliable tool. A single source of truth. And, ideally, no complex code requirements. 

That’s where the concept of a marketing data hub comes in. Such a tool (like Funnel) allows you to connect to all your data sources to extract the data. It’s much more than just an ETL or data pipe, though. A marketing data hub also lets you perform transformations while also storing the underlying source data. It also shares your data anywhere you need it, and stores data to mitigate the risk of breaking reports and dashboards. 

It’s a fairly new concept within the marketing data space, but also one that makes logical sense when viewed through the eyes of a digital marketer. 

It’s a tool that can be “owned” by marketing teams so that they don’t get stuck in IT bottlenecks while still being able to integrate with complicated data stacks that are maintained by the BI teams. It’s a single source of truth that’s also rock solid. And if it’s a tool like Funnel, you don’t even need to write a single line of code. 

Which solution is right for you?

While it has the same underlying problems, the marketing data challenge presents itself differently to every marketing team. The answer lies in how much complexity you face in your data-driven marketing strategy. A straightforward and smaller marketing operation can be quite comfortable with the copy-and-paste approach. However, this won’t even last a day for a medium or large digital agency. 

Plugins can help you take on more complex levels of the marketing data challenge, but there comes a day when they won’t be sufficient anymore. Instead, you’ll need a data stack or a marketing data hub. 

The most interesting approach is that of the marketing data hub. It can adapt to suit many levels of complexity. It can help those considering the plugin approach all the way up to enterprise marketers who need a single source of truth.

It all comes down to your own preference and business needs.