Marketing data integration: why you absolutely must combine cause and effect
Data integration is one of the hardest challenges marketers face. But it doesn’t need to be.
It never used to be that way. And most likely, it wasn’t something you thought about when you considered a career in marketing. But the world has changed and data-driven insights are the key to the scientific marketing that the very best marketers on the planet are doing. That’s precisely how Singular clients like Lyft and AirBnB and Supercell and Ubisoft have grown so fast.
What are they doing?
Using data to build insights that inform marketing strategy and tactics.
This is truly a marketing superpower. If you can acquire a customer or a user 20% cheaper than your competitor, you win. If you can discover an untapped source of new business, you win. If you can move quicker thanks to real-time data, you win. And if you can optimize your advertising campaigns quicker than your competitors, you gain an ever-compounding advantage that just keeps growing over time.
What that means, essentially, is that if you are looking for outsized growth and marketing success, data integration is one of the key things you need. (Plus data integration tools that make it easy so you don’t have to spend countless hours with data integration techniques or feeding a data warehouse, of course.)
Sounds like a stretch? Read on to find out why it’s true.
Here’s what we’ll be diving into:
- Marketing data sources are exploding
- You have OK insight on your “effect” data
- But … data integration is required: it’s not all in one place
- And … there’s a core chunk of data almost no-one is integrating
- Combining campaign and conversion data is the key to knowing true ROI
- Plus, it’s essential for full optimization of ad campaigns
- So now you have a single source of truth for marketing data
- Next steps
Marketing data sources are exploding
This is no surprise to modern marketers. You have multiple paid channels, many marketing partners, multiple ad networks, and at least four or five organic digital channels, and that’s probably not the full story.
Adding to the mix is that there are now so many near-global platforms that aggregate audiences of your potential users and customers. You’re marketing there, which makes them your data sources.
You know that Facebook, Messenger, and Instagram are billion-user platforms. You know that YouTube and WeChat are as well, and are critically important for different groups of people. Plus, there’s LinkedIn and Snapchat and Twitter and Reddit. And don’t forget QQ, Viber, Pinterest, or Qzone, each with hundreds of millions of users.
All of them are potential source systems for marketers, with more and more real-time data being created than ever before.
“Marketing data is exploding. It’s growing much faster than the industry itself.”
– Singular CEO Gadi Eliashiv
All of that is adding big data: more data sources and data integration challenges for marketers to consider. So many sources with real-time data and aggregated data that marketing technology is one of the fastest-growing industries. It’s currently worth $230 billion/year and growing 20% year over year.
(There’s a reason why data integration isn’t just for IT departments anymore!)
In fact, this is adding so much more data that it’s getting challenging to handle, especially for marketing teams that often don’t have a lot of data engineering support. Separate data points are interesting and, yes, good to have … they’re much better than nothing, of course. But integrated data is much more powerful. To achieve that, you might think you need data integration software, or a data integration platform, or a data warehouse. And, of course, a massive data integration project.
And you do, in a sense.
But it’s better to get your data integration solution wrapped up inside a marketing platform that handles all the data for you while preparing it precisely for the purpose of helping you optimize marketing results.
The alternative, of course, is some combination of massive databases, a data lake, or a combination of data integration systems that collect all the data for you … but don’t present it or integrate it specifically for marketing intelligence.
You have OK insight on your “effect” data
The reality is, you have some fairly decent insight on your “effect” data. Good enough for some business intelligence uses.
That’s the dashboard for your Facebook Ads, the report that the agency sends you, your Google Ads dashboard, and any other dashboard you might have for other ad partners. Plus, of course, other analytics on email, website, and other organic or paid marketing.
You’ve got the results, so you’re all good, right?
You can see your conversion rate on Facebook, how many clicks Google drove, your email open rate in one analytics package, and your website conversions in another. For mobile app marketers, your ad partners will tell you how many installs they think they drove. All good?
Well, not exactly.
Not unless you really like doing the dashboard dance. And not unless being handed a bunch of unassembled puzzle pieces is your idea of seeing the big picture.
That’s not good data quality, even if all the data is available. There are too many disparate sources, and no data integration platform that can accurately inform business processes.
But … data integration is required: it’s not all in one place
Seeing the puzzle is hard when the pieces are all over the floor. Or in 15 different dashboards and reports and analytics tools.
That makes it challenging to get actionable business intelligence.
You’ve got disparate data sets and databases in multiple places. And it’s not just a simple matter of connecting those sources through some kind of ETL tool (extract, transform, and load) or process. There’s also the much harder problem of semantic integration: building ontologies of data that allow you to standardize what a click or tap means, what a video view means, what a conversion means, and so on. That standardization of data is critically important.
And then you’ve got normalization.
If you want a simple answer to who’s scoring the most runs in the MLB, you also need to factor in how many games teams have played. If the Yankees have played 15 games and the Giants have played five, then it’s not shocking if the Yankees have more runs. In a similar way, if you want to know what ad partner is delivering the most ROI or conversions, you need to normalize all your ad partners so that you get truly actionable data.
And let’s not get started about data cleansing … most data “scientists” spend huge amounts of their time doing data “plumbing,” just to ensure they have clean data to draw conclusions from.
This is what a proper data integration process can solve. And that’s what Singular can solve for marketers … even if they don’t have a lot of data engineering support.
(Plus, of course, ensuring that all your data is accurate. And supplying proper attribution for those marketing results.)
One of the first things you do when setting up Singular is feeding the data connections: adding source systems to tell the engine where you’re marketing, where you’re advertising, and connecting Singular to all those locations. That might be partners in mobile advertising, web advertising, or other channels: whatever.
Everything gets connected, added, standardized, normalized, cleansed. The result is a true data transformation and a unified view … no data warehouse absolutely required (although many Singular customers do also maintain a large data warehouse). And with Singular providing the data integration tools and data integration platform, all in one place.
But there’s more, isn’t there?
There’s a core chunk of data almost no-one is integrating
It’s really great to have a unified view of your marketing results. It truly is a major achievement to integrate both raw data and aggregated data about all of your wins and losses. And it’s wonderful to have databases with proven levels of data quality.
But it’s only partial.
And it doesn’t give you a full view of marketing performance.
Results are effects. They’re what happened because of what you did … because of money that you invested. Because of campaigns that you created, creative that you designed, ad copy that you wrote, and offers that you presented to potential users and/or customers.
To fully understand effects, you need to know causes. And then you need to connect them in the big picture … and in the granular puzzle pieces littering your marketing floor.
That’s your costs. Your campaigns. The links you’ve created for offers and conversions. The creative that you’ve designed. If you can connect all of that in a very granular way with the impact each of those components has, then you can answer some of the smartest questions marketers can ask.
All of this, from all your sources, needs to come together to provide the marketing intelligence and business intelligence that you need. This is big data with great data quality, brought together via Singular’s data integration tools.
- What creative works best across all our marketing partners?
- What offers out-performed others?
- What is my true ROI across all my apps/products/campaigns?
- How do different bid types and budget limits change my results?
- And much more …
Without a massive data engineering team — or data integration tools like Singular, you simply can’t answer these questions at scale and in near real time.
Combining campaign and conversion data is the key to knowing true ROI
Ultimately, combining both sides of your marketing data is critical to understanding true ROI.
Source systems like the ad network side supplies all your marketing data: what you’re doing. The attribution side supplies your conversions (app installs and other conversions).
Almost no-one besides Singular has an integration solution with data integration software that can do this in near real time. To enable it, we have two (and sometimes more) integrations with almost every marketing platform.
We call it dual integration: all your cause data … your campaigns, creative, bids, budgets, and so on, combined with all the attribution data that Singular, as a Mobile Measurement Partner, has recorded. They get connected and combined.
The result if customer data integration. And the impact is marketing steroids, without the need for additional integration tools or integration processes.
Plus, it’s essential for full optimization of ad campaigns
When you dual integrate your campaign and attribution data, you get insight out of big data. Intelligence, thanks to smart business processes. And you earn yourself the ability to make smart decisions on allocation and optimization.
You can’t get the cause data without deep integrations with hundreds and hundreds of disparate sources: marketing partners and ad networks. Tracking links don’t cut it, because not all networks support macros, you don’t get creative insights, and data loss/discrepancy is high … up to 31%.
Given that you’re often trying to optimize an extra 2%, or 5%, or 10% out of a campaign, 31% is way too high. That much discrepancy will completely destroy your ability to intelligently optimize: you simply don’t know if you’re getting better or worse, or if the discrepancy is either growing or shrinking.
So now you have a single source of truth for marketing data
Once you have both your campaign data and your attribution data centralized in the same place, you have a single source of truth for marketing data.
And that’s important not only for marketing, but for everyone.
It’s not a great situation when the marketing has the the right numbers but the CFO has the wrong ones. That is just going to get the marketing team in hot water and waste cycles of everyone’s time. The way Singular’s solution has been designed takes into consideration that a critical part of the entire company’s success is having a single source of truth for every team in the company.
And, when every team — and all team members — can access reliable, standardized data within a single platform, everyone is empowered to make smart decisions that drive growth and benefit the business. Plus, talking apples to apples in inter-department communications enables better collaboration between marketing and finance, agency and brand, and channel managers.
Are you ready to have a single source of truth bringing all your disparate sources together thanks to cutting-edge data integration for your marketing campaigns and results? Ready to massively improve your business processes around marketing optimization?