tl;dr: A well-architected martech dataflow infrastructure simplifies your data flow and makes it easier to engage users at key moments. Here’s how to build it.
Those of you in the digital marketing technology (martech) space are familiar with the ever-growing number of tools available for customer relationship management (CRM), digital advertising, email marketing, marketing automation, analytics, and countless other functional areas.
The Chief Marketing Technologist blog, in their 2018 annual survey of the martech space, identified and classified 6,829 martech solutions from 6,242 unique vendors, up from roughly 1,000 different solutions in 2014. Their martech landscape diagram is both exhaustive and exhausting; clearly there is a tremendous amount of activity in the martech space.
There are 3 Approaches to the Stack
Faced with an explosion of combinatorial possibilities for assembling their companies’ own martech stacks, you can take one of several approaches to building your martech stack:
- All-encompassing solution vendor: Large SaaS vendors continue to develop and enhance their suites in an attempt to deliver every digital marketing solution that you need. What used to be a simple CRM is now a “marketing cloud.” These large vendors often grow their suites by acquiring previously independent point solution vendors and attempting to integrate those tools with their existing software. The downsides of this approach are cost, inflexibility, and missing out on capabilities when the all-encompassing solution vendor’s suite falls short of your needs.
- Best-of-breed: You can, of course, assemble a best-of-breed suite of tools by choosing those individual solutions that best meet your particular needs. The main challenge of this approach is integrating your chosen tools together. The tools almost always require data from other sources, and they produce data to be sent to yet more tools such as analytics, reporting, and dashboards. This approach also requires you to learn many different user interfaces.
- Hybrid: This is a fairly common approach: you may choose to use certain components of an all-encompassing solution, but supplement with best-of-breed point solutions in certain areas where the core suite’s offering falls short. Integration challenges still remain, but will be less than a pure best-of-breed approach.
Integration is the main challenge marketers face when choosing the best-of-breed or hybrid approaches.
Some tools come with built-in integrations. But when a tool lacks integration capabilities with another tool in your stack, IT has to develop a custom integration. These custom integrations may rely on periodic exports of data to spreadsheets or CSV files and nightly batch jobs to load new data into tools. There is rarely a single data repository that functions as the overall system of record. If user data is not flowing in real time to all the appropriate platforms, your view of the current state a user or audience will always be somewhat out-of-date.
Secrets of Highly Successful Martech Stacks
Real-time data integration is especially critical to an omnichannel martech stack that spans all of your sites, mobile apps, advertising, email campaigns, social media, points of sale, and other touch points.
The most important source data is the actual data: what sites a user visited, what pages or screens they viewed, what actions they took on those pages or screens, what data they submitted via forms, what actions they took after receiving an email, etc. These individual user actions are properly called events. A well-integrated martech stack will ensure that each event flows in real time to all the tools that can make use of that event.
A new solution solution to this integration challenge is the “customer data infrastructure” tool or “user data infrastructure” tool. Data infrastructure tools simplify and improve your overall martech stack architecture by eliminating other integrations.
The best data infrastructure tools available can perform four key functions:
- Connect to a wide variety of user data sources (sites, apps, cloud services, emails, advertising) to collect user events
- Distribute events in real time to a variety of other martech tools that can make use of them
- Collect all user events in a data warehouse that functions as a system of record for all user event data
- Identify users, when sufficient data exists to do so, to allow events to be attributed to actual known contacts
Why Every Marketer Needs a Data Integration Tool
Adding another tool to your stack may seem counterintuitive when trying to solve the “swivel chair” problem of too many platforms. But good data infrastructure will deliver significant benefits:
- Faster, more accurate view of users: By delivering event data in real time to connected downstream tools, a data infrastructure tool will provide a more accurate, up-to-date view of user activity. Subsequent actions on different platforms, such as triggering an email or other user interaction, can happen faster.
- Improved data reliability and quality: Most martech tools are cloud SaaS tools with APIs for receiving data. Sometimes those APIs experience outages or momentary periods of overload, which can cause events to be lost simply because they cannot make their way into the tool. A good data infrastructure tool will retry events over a time period to improve the delivery of event data to downstream tools.
- Centralized, proprietary source of insight: All user event data will flow into a data warehouse, which is owned by your company. This provides a system of record, controlled by you, which allows data to be analyzed and aggregated into reports and dashboards via business analytics tools.
- Reduced burden on your IT team: Built-in data infrastructure tool integrations eliminate the need for internal IT resources to develop custom integrations, and also simplify the client-side software deployed to sites and mobile apps. One piece of code from the customer/user data infrastructure tool is placed on your sites, apps, and other properties — instead of multiple pieces of code from different tools in your martech stack.
- Platform flexibility and data replay: If you choose to change or upgrade a martech tool in the future, it’s as simple as turning off the old integration in your data infrastructure tool and turning on a new one for the martech tool you’re adding to your stack.
Furthermore, since historical user event data is stored in your data warehouse, that data can be “replayed” into the new tool to build up the new tool’s database.
There’s no data integration tool that can connect together all 6,829 martech solutions in the 2018 martech landscape, nor is there actually a need to do so.
But for those tools in your stack that can act upon real-time user events, a good data integration tool can make the constantly evolving martech space that much easier to navigate. With the explosion of tools available, now is the time to hedge against chaos with an integrated data flow across your entire architecture.