Triple Whale DS

All Case Studies  🔍 Advertise With Us

Palms are sweaty, knees weak, arms are heavy. There's vomit on my sweater already, mom's spaghetti. This is me right now. Why? Today’s our first in-person Marketing Examined community event. I’m nervous. Excited. Ready to meet the folks who support us. The people building the tomorrow.

Because my entire thesis is this…

Find dope people. Connect them to other dope people. Doing dope things. And who will do dope things together.

If we do that, then I’m happy.

Ight. But for today. Marketing attribution 101. Because we all know marketing attribution in today’s landscape feels a lot like watching Interstellar.

But before we dive into this playbook, a huge thank you to our partner Triple Whale.

You ever watch Interstellar? And wonder what the hell was going on. Or how it all was connected. You had to read some theories online. Search Twitter. Piece it together. Re-watch it. And now you’re a bit more confident but still kinda lost.

That’s the current landscape of marketing attribution.

Hundreds of touchpoints. Multiple platforms trying to take credit for the sale. You know the drills.

Well, let’s start here…

How to Operationalize Marketing Attribution

Marketing attribution is like assigning credits to each marketing touchpoint that drove a sale.

You do this to figure out which channels/campaigns are the most effective. So that you can make smarter marketing decisions.

Because, if you know half your media spend is going down the drain and getting wasted on Reddit, then with attribution, you’d know to cut it. And how to reallocate.

And we’re going to dive into that soon.

But first a quick overview of the 3 most common types of attribution models.

1. Last-Click Attribution:

   • What it does: Gives credit to the last interaction before a conversion.

   • Pros: Easy to access the data, e.g. Google Analytics

   • Limitations: Overemphasizes low-funnel channels and overlooks the importance of demand generation. For example, brand search taking credit for Facebook ad

2. Multi-Touch Attribution (MTA):

   • What it does: Considers a user's entire history of views and clicks to determine which channels caused a conversion through a statistical model.

   • Pros: Provide a more holistic view of the customer journey vs. the last click

   • Limitations: Can't capture all customer data and struggles to capture non-digital channels like billboards.

3. Marketing Mix Modeling (MMM):

   • What it does: Collects historical data about ad campaigns, budget, and conversions to build a statistical model

   • Pros: can capture non-digital campaign impact and show channel diminishing returns

   • Limitations: Need a large amount of historical data to build the model. This means the model cannot capture real-time events, and conclusion from the models are only valid if conditions for the model remains true (e.g. COVID-19)

So, when I was working with Prep To Your Door here in Austin, we didn’t have a team of data scientists or the budget for external agencies to develop complex models.

Which is why we turned to Triple Whale. And how I learned everything about marketing attribution.

At first, we had no idea where revenue was coming from. All we saw was a sale in our dashboard, with zero knowledge of which dial to turn to get the next sale.

So we launched a total impact attribution model using Triple Whale, adding a First-Party Pixel Data and Post-Purchase Survey Data to track attribution across all touchpoints. All of a sudden, the marketing department had clear eyes on which activities were working.

We understood how revenue was coming in across:

• Channels

• Campaigns

• Ad Sets and Ads

But not only did we know where the last sale came from, we knew the weighted impact of each touchpoint that led to the final sale.

For startups, especially in the e-comm game, leveraging data-driven insights into fast iteration is crucial to make smart marketing decisions.

Lean Analytics

A framework that helps businesses make quick iterations.

Here’s how I combined Triple Whale’s marketing attribution model with the Lean Analytics framework:

• Ideas: Generate ideas for your product or service and establish hypotheses about your target audience. For example, you might think that Facebook and Google Ads are good channels to invest in.

• Build: Create a Minimum Viable Product (MVP) or prototype for testing. In our case, it could be a batch of creatives for your ads to start testing.

• Product: Refine your product or ads based on feedback from early users or in-platform performance data. Adjust your creatives based on how well they are performing.

• Measure: Define key performance indicators (KPIs) to measure success. In the context of channel optimization, your KPI could be Return on Ad Spend (ROAS) to measure the effectiveness of different channels.

• Data: This is where Total Impact Attribution comes in to understand the impact of different channels on your revenue.

• Learn: Analyze the data to inform the next cycle of improvement. For instance, while Last-Touch Attribution shows that Google Ads is more effective with a ROAS of 2.15 vs 0.85 for Facebook, Total Impact Attribution reveals that Facebook and Google Ads have similar ROAS.

This is a perfect example of how a shallow attribution model can lead you in the wrong direction.

So with the right attribution model and a lean analytics testing framework, you can take your learnings to restart the cycle for the next round of optimization.

If you’re looking for an attribution tool that will help you make better decisions, faster, to demystify what parts of your marketing spend are truly driving your results, then you should give Triple Whale a try.