The importance of measurement, attribution, and incrementality

A little more than a decade ago, the state of digital marketing was drastically different. 

Most campaigns were oriented around reach, impressions, and clicks. Cost per 1,000 impressions (CPM) was the key metric if you were buying media on Google or through publishers.

Brands on Facebook were worried about buying likes, follower counts, and engagement rates. Getting a big audience w/ high engagement was the goal. 

The ecosystem has matured and evolved over time. As has DTC and e-commerce in general. 

With that has come increased sophistication of what business leaders and operators want out of their digital marketing channels. This has led to an explosion of tools and options - but also of complexity. 

Any new founder launching a DTC brand in 2024 is faced with an entire galaxy of metrics, platforms, and tools, all moderated by a growing body of science around how best to optimize your marketing and your business. 

It can be confusing for anyone new to the industry. Or if you can’t keep up with the pace of the evolution. 

Today we’re going to go over the state of measurement, attribution, and incrementality in DTC. We’ll look at where things are currently and how founders and operators can orient themselves as they try to understand and adapt to the tools of e-commerce. 

Today

• Attribution

• Tracking

• Incrementality

Attribution - the basics

Throughout most of marketing history, attribution was accomplished via proxy measures. 

How many papers was that new ad printed in? How many viewers saw our latest TV ad? Did sales go up after the campaign?

With digital marketing came the promise of direct attribution. 

Modern marketers can now track who clicked what, when, and if it led to a sale. A new panacea for optimization and waste reduction, right!?

Unfortunately, we’ve learned that attribution is nowhere near as simple as click→ sale. 

Here’s why:

• The customer buy path is rarely immediate and direct. Sometimes it can take days, weeks, or months, and multiple touches across different channels before someone purchases.

• Different marketing channels have different attribution settings and they are all eager to take as much credit as possible for every sale. 

• The proliferation of devices and privacy/ad-blocking tech has created data gaps and blindspots in terms of user tracking. 

Practically every tool and school of thought in DTC marketing attribution these days is focused on one of these three challenges. 

Let’s break this down into the key factors founders and marketers must consider: 

Attribution setting: How do you assign credit for a purchase? Last click is often the default setting in most tools, but that ignores all of the touchpoints that may have happened before the user finally clicked on something with the intention to buy. 

Attribution window: How long after a click or action do you assign credit for a sale? One day? 7? 28? The tighter the attribution window the more certain you can be of the relationship between ad and purchase. But most users don’t just click and buy right away. 

The longer your attribution window, the more users you will capture, but the signal between your initial action and the purchase is degraded the longer the time period you wait. 

Funnel position vs efficiency: User actions lower down the funnel will always lead to more reported efficiency because they are already primed to buy. However, if you only optimize for this kind of efficiency in your marketing, you’ll eventually run out of highly interested buyers. Let’s call this the “shooting fish in a barrel” problem. 

Tracking gaps: Platforms like Meta and Google use signals from your on-site users to improve and optimize the targeting of your ads. But if your data has blindspots due to privacy blocking and device hopping, your attribution will be incomplete, confused, and potentially misleading.  

Takeaway: There is no set-it-and-forget-it in digital marketing attribution and measurement these days. 

Marketers and operators need to fundamentally understand the key challenges facing accurate attribution and optimization before they wade into the deep waters of DTC marketing. 

No one wants to be the founder looking at a reported 8X ROAS in their Google brand search campaign wondering why the increased budget isn’t translating into more sales. 

Tracking - data aggregation and integrity

So those are the problems. Are there any solutions?

There are two ways marketing and SaaS platforms have tackled tracking:

1. Marketing business intelligence dashboards

2. Pixel and server-side tracking tech

BI dashboards are aimed at solving the issue of siloed/disintegrated data and reporting.

Having to aggregate different information scattered across Meta, Google, Shopify, and accounting software is not only difficult and inconvenient, but it can mean neither founders nor marketers have a clear, complete picture of spend, revenue, per channel impact, or net profit at any given time. 

A well-tuned dashboard should gather marketing performance data, site performance data, and marry it to important factors like cost of goods sold and ad spend. 

As for First-party pixels and server-side tracking, these are the responses to the iOS14 update and other privacy measures that have greatly reduced the efficacy of traditional tracking technology like cookies and third-party pixels. 

Here’s how the two work and how they differ:

First-Party Pixel

1. Client-side implementation: A first-party pixel is typically a small piece of JavaScript code placed on the website that runs in the user's browser.

2. Data collection: It collects data directly from the user's browser and sends it to the website owner's analytics platforms.

3. Cookie usage: Often relies on first-party cookies set by the website domain to track users across sessions.

4. Limited by browser restrictions: Can be affected by ad blockers, browser privacy settings, and cookie limitations.

5. Easier implementation: Generally simpler to set up and manage, often through tag management systems.

Server-Side Tracking

1. Server-side implementation: Data collection and processing occur on the web server rather than in the user's browser.

2. Data flow: Website interactions trigger server-side events, which are then processed and sent to analytics or marketing platforms from the server.

3. Not dependent on cookies: Can function without relying on browser cookies, though may still use them for certain purposes.

4. Bypass browser limitations: Less affected by ad blockers and browser privacy settings.

5. More control and flexibility: Allows for data processing, enrichment, and selective sharing before sending to third-party tools.

6. Improved data accuracy: Provides more reliable data collection, especially in environments with strict privacy controls.

7. Enhanced privacy compliance: Offers better control over what data is collected and shared, aiding in compliance with privacy regulations.

Meta also entered the fray by implementing Conversions API (CAPI), their own server-side tracking technology. Setting this up is non-negotiable if you’re spending money on Facebook or Instagram these days, especially if you are not leveraging a 3rd party server-side tracking tool.

Takeaway: To have a clear picture in terms of your marketing impact and efficiency, you need to:

• Fill the data gaps caused by privacy and device hopping

• Effectively aggregate your inputs and outputs so you can understand what’s happening in your business. 

Recommended Tool - Triple Whale Sonar

If you’re a DTC brand spending $1 million or more per year on ads, your martech stack is probably littered with:

• Business intelligence dashboards

• Attribution and incrementality tools

• Server-side tracking and first-party data pixels

With the cost of acquisition only going up, it’s never been more important to understand what impact your ads are having on your bottom line. For most brands, this means aggregating and analyzing data across a fractured ecosystem. 

The good news is Triple Whale is working to change all of that through Sonar - their cutting edge integration with Meta Conversions API.

Designed to send enriched first-party user and conversion data to advertising channels, Sonar complements Triple Whale’s comprehensive business intelligence dashboard, attribution tracking, and AI-assisted data analysis. 

How Sonar Elevates Triple Whale

Sonar’s server-side tracking captures the wide variety of user interactions - clicks, views, conversions, abandoned carts, and more - without being disrupted by browser restrictions, ad blockers, or privacy-focused updates like Apple’s iOS changes. 

Major benefits include:

• Accurate Data Collection: Sonar delivers more precise data about user behavior across websites and devices.

• Privacy Compliance: As a first-party pixel, Sonar keeps brands in control of their data, ensuring compliance with privacy regulations.

• Seamless integration with Meta: Brands advertising on Meta see improved tracking accuracy and performance insights.

• Cross-Channel Optimization: Sonar’s enriched data means brands can optimize performance across multiple channels.

Sonar enables brands to measure campaign performance more effectively, even as the rules around digital tracking become stricter. 

Create a single source of truth

Sonar is an extension of the existing Triple Whale platform, which already includes everything else a scaling brand needs to analyze its marketing efficiency - attribution tracking, dashboard analytics, and predictive insights. 

The addition of Sonar transforms Triple Whale into the only one-stop shop for real-time marketing insights. The data is not only consolidated under one roof, it’s more trustworthy and transparent. 

If you’re an operator or marketer looking to enrich, consolidate, and leverage your marketing data like never before, be sure to check out Triple Whale here. 

Incrementality - The holy grail

The last mile in e-commerce marketing measurement is incrementality. 

This becomes a major consideration as brands scale to 8 and 9 figures and start adding more and more ad spend/channels to the mix. 

At that size, understanding what efforts are efficiently creating net new demand is the key to optimizing ad dollars. 

Remember the tale of the poor, inexperienced DTC operator trying to grow sales by dumping more money into branded search terms? That’s the base case of not understanding incrementality. 

Ideally, ad spend should be helping your brand make sales it otherwise wouldn’t have made. The more impact your ads, channel, or campaign have in terms of activating new customers, the more you will be able to grow. 

 It’s a simple concept but can be extremely difficult to determine. 

Here are some methods to help measure incrementality:  

A/B Testing

This involves splitting the audience into 2 groups:

• Test group: Exposed to the marketing campaign

• Control group: Not exposed to the campaign

The difference in conversion rates between these groups indicates the incremental lift.

Holdout Tests

This involves temporarily pausing advertising in certain segments or channels to establish a baseline performance level. 

This "dark period" helps measure the true impact when the ad spend is reintroduced.

Geo-Based Holdouts

DTC brands can test campaigns in specific geographic areas while holding others as control regions. 

This method is particularly useful for measuring the impact of broader marketing initiatives.

How to measure incremental lift:

1. Measure conversion rates in both test and control groups

2. Use the formula: (Test Conversion Rate - Control Conversion Rate) / (Control Conversion Rate)

This result shows the percentage increase in conversions attributable to the marketing activity.

Keep in mind that brands will need a significant amount of data to achieve statistically significant results. 

That’s another reason why incrementality is typically the concern of larger or rapidly scaling brands - without thousands or tens of thousands of data points, your experiments are going to be confounded by randomness. 

Note - Meta also has native lift test functionality, with the goal of determining the impact of your Facebook ad spend over and above other marketing activities. 

Find out how to run a Meta lift test here. 

Takeaway: Incrementality is the holy grail of marketing measurement. 

By understanding what channels, campaigns, ads, and creatives are driving net new demand, brands are able to prioritize and double down on what is working - while sidelining anything that isn’t. 

For smaller brands that are trying to get traction or are just beyond the product market fit phase, it’s better to focus on the basics of marketing attribution and measurement first. 

Scaling brands spending $1M+ a year and operating nationally or internationally should be heavily invested in understanding the incremental impact of their efforts, however.