Launching a FAANG Marketing Campaign

With a degree in marketing and working in the field for over a decade, I have had experiences with companies small (< 50) and large (50,000+). None of this truly prepared me for the technical program management of a marketing campaign in a FAANG environment.

To give a high level, here’s a list of the teams/skillsets represented — each of these being filled by a different individual and many times multiple are needed for a single program:

  • Project Manager
  • Marketing Manager
  • Analytics
  • Data Engineer
  • Software Engineer
  • Marketing Operations
  • Sales Enablement

…and this is just the core team on day-to-day needs.

In my past few weeks as part of this core team, I have learned much about the process, dependencies, and hand-offs. I thought it would be helpful for others interested in working within a large tech company (FAAANG or not).

First thing is first. The Marketing Manager and their greater team (referred to as “the business”) will write up the campaign they want to launch. Along with a general description of the campaign, the brief generally covers:

  • objectives and customer behaviors — why is this campaign being run? what is it driving — revenue? product adoption?
  • the problem is the campaign going to address — what customer pain point is will the messaging speak to?
  • the message or call to action for the campaign — what are we asking customers to do?
  • KPIs and targets — how will the campaign be measured and what is our goal?
  • target audience — who is the campaign and its messaging meant for?
  • marketing channels — how will the target audience be reached? email, social, SEM/programmatic ads?
  • launch date and reporting date expectations — what’s the deadline for this campaign?

Once the executive powers that be approve the campaign, we are off to the races from the data perspective.

The business has documented who they want the campaign to reach as their target audience and the next step is to speak with the Analytics resource to build the campaign audience. This is an iterative process. What is in the campaign brief will oftentimes not be detailed enough to get to the final audience.

For instance, the business can ask for clients who have not yet adopted product X; but there may also be data that tells us of clients who were pitched product X and are unable to adopt it because of current technical issues. The Marketing Manager doesn’t want to message these clients to pitch product X again if the client a) already knows about it and b) they are struggling with technical implementation. This will only frustrate these clients.

In my experience, this is very much an ongoing process as business discussions around the campaign happen every hour of the workday and adjustments based on those conversations are often needed. The important thing here is that Analytics knows the best data sources available to understand these audience features and they have time to model out and analyze the data. This analysis can be focused on clustering the population defined to understand how to segment and create messaging for external purposes of the campaign, or even focus more internally on how to efficiently test a campaign within a certain region or Sales team (i.e. instead of piloting a campaign with 1000 clients belonging to 1000 Sales Reps around the world, how do we pilot in a single region with 50–100 Sales Reps). In this way, the data work creates the foundations for the Marketing Manager, Marketing Operations, and Sales Enablement to move forward to create a more personalized and organized campaign with Marketing and Sales alignment.

In a large tech company, there are endless data sources and internal wikis on what to use — but what is the best source of truth? Fortunately, the Analytics resource had to dig into this already in the audience building step; but now we have to work with Data Engineers to build the infrastructure (data tables) to support the campaign.

There are plenty of discussions to be had here including: planning out and constructing the data table(s) schema, reliance and dependencies on upstream tables for certain data points, feature engineering needs, and ultimately deciding what value type the data needs to be stored as especially when planning to push it into another system (hello data transformation steps!).

These infrastructure conversations bring another consideration to the table — timing. This is because in many instances, this data is triggering marketing messaging that aims to reach the right audience at the right time. Thus, the foundation building in this part of the process is a significant driver of how a campaign can be executed in the real world.

As another example, there can be a campaign around onboarding new clients. The onboarding messaging needs to go out on the first day of the quarter to introduce clients to their account manager, but some systems don’t fully receive and store this information until two weeks into the quarter. If we need a table to store the data and get it into a Marketing system to kick off a welcome email day 1, then getting that data on day 15 won’t work.

Timing conundrums like these can lead to a reassessment of data sources. This may impact the initial audience build from Analytics and just like that we continue with the data iterations supporting the campaign. Ultimately, there will be a point of compromise where priorities are restated and the data source decisions align with those priorities instead of meeting all of the possible requirements.

To continue with the onboarding example, Marketing and Sales alignment in this type of campaign is a priority. Both teams need to know the account managers and the clients who are assigned to them on a given start date. Thus, the most important thing is to build the audience for this campaign based on what Sales is using day-to-day to manage their client assignments. This may also mean that Marketing cannot get this information until day 15 instead of day 1; but since it is more closely aligned with the priority of the campaign, this should be used as the source of truth for the audience build and stored in the tables supporting this campaign ongoing.

Once there is an understanding of the campaign audience size and a plan of attack on creating the data infrastructure to support the campaign, the Software Engineer can understand how to best get this data from the data tables into the system.

Many times these API integrations in my experience are for pushing data into a marketing automation platform like Eloqua or Marketo.

This is the point where Marketing Operations and Software Engineering need to work together. The Marketing Operations resource can assess what is needed to manage the campaign in their system including any field additions that are needed, checking that the current data formatting in the data table is supported by their system so they get the expected values (especially if planning to use them in client facing messaging like plugging in a client name), and maybe most importantly, making sure that duplicate contacts are not created in this process of importing the campaign audience.

As Software Engineering makes progress on establishing a pipeline into the marketing automation system, Marketing Operations can help them verify data in passing in as expected and in a timely fashion. This can be extended to testing automated emails across segments (based on our Analytics clustering) that would get triggered when the data is pushed in/updated in the marketing system.

Across all of these teams is the Project Manager who is understanding each of these puzzle pieces needed to complete the campaign execution picture. This person is mapping out the the level of effort, time required, and dependencies for all the tasks associated in launching the campaign. With this higher level view in mind, they can reverse engineer timelines and assign deadlines to each of the tasks. This helps to establish the urgency and speed that tasks need to be completed, how much allowance there will be for hiccups that undoubtedly turn up, and ultimately manage the expectations of the stakeholders outside of the day to day progress.

Having a confident hold on calendar deadlines and resource availability for tasks helps in holding people accountable as well as driving efficiencies for campaign launch.

Congrats! The campaign made it to launch!

Even though everyone mentioned has been part of QA’ing the campaign before it went live, everyone stays on alert for any unexpected challenges. There will always be at least one as no one is perfect and that is okay! Just make sure the team has accounted for anything high impact that could go wrong. ;)

MarTech, CRM, automation and data nerd. Managing a small zoo of 3 cats and a dog in Austin, TX.