Marketing measurement in 2024

Winners and losers in a changing measurement landscape

TL;DR cookies are going away and we're not sad about it (they weren't as effective as they promised to be anyhow) but modern marketers still need to adjust their measurement strategy. As of right now, organizations can rely on experimentation, mixed media modeling, qualitative surveys, and monitoring business performance (in addition to cookies, as long as we have them).  We believe marketing teams that fully embrace experimentation and other statistical measurement techniques in 2024 and beyond are going to win against teams delaying the transition away from cookies. A culture of experimentation is the only way for marketers to win going forward.

I’ve been in digital marketing for 15 years and have seen the rapid rise of cookie-based attribution — I’m here to say that I can’t wait for them to be dead. 

Sure, it’s going to be a big year of changes for marketers but it’s for the better. Cookies weren’t great at measuring performance marketing anyhow. 

But it does leave the question, how do organizations measure their advertising spend going forward? That’s exactly what we’re going to dig into.

What are the problems we’re facing now?

The earth is shifting under marketers in 2024. Third-party cookies, a simple and ubiquitous marketing measurement technology, will be blocked by the largest web browser (Google’s Chrome) in 2024. 

With the death of third-party cookies so too dies the dream of hyper-granular digital marketing measurement and the foundation for how many marketing teams today measure the results of their advertising campaigns.

Thankfully, the tools and approaches replacing cookie-based attribution are much better anyhow. Why? These tools put the focus on incrementality, getting away from short-term thinking and creating a better understanding of the levers driving real business goals. 

The marketing teams that embrace this change will set themselves up for long-term sustainable growth while the rest struggle to find performance in an industry bent on granularity.

How exactly cookies work (and where they don’t work)

Let’s start with some of the reasons why marketing measurement isn’t as easy as installing a pixel, setting up a campaign, and watching sales go up and to the right.

Tracking people across the internet is hard

Website tracking today mostly relies on cookies - little bits of data created in your browser to tell advertising platforms and analytics tools what you’ve done and who you are.

And they work amazingly well! They help websites remember you. They help analytics tools know which pages you’ve visited for analysts to understand behavior. They can even connect a user back to their profiles on Facebook, Google, or other platforms.

But things get complicated when you go past just a single web session. Cookies expire, are cleared, or are blocked by browser plugins (or browser settings). People use multiple devices, their partner’s device, their office device, a public device, or any device they can get their hands on. Users also SPEAK to each other, both in real life and in closed garden platforms like Slack, Discord, and many more.

Basically, there are a million ways cookies can’t track users effectively across their interactions with a brand. So the moment your advertising is disconnected from when a user viewed it to when a user takes an important action - cookies fail to make the connection.

We call this a “known unknown” - cookies aren’t tracking the whole user journey but we don’t know what pieces are missing. This is a huge problem for accurately measuring the impact of your marketing on sales. Take this example from our VP of Product & Partnerships, Phil Ware.

What are the ways of measuring marketing?

Ok, so we know tracking people across the internet is difficult but where does that leave marketing teams who are trying to demonstrate the ROI of their campaigns and to justify their marketing budgets in 2024?

Here’s a high-level picture of the ways to measure marketing right now:

Cookie-based attribution

• What - Attribution ties a conversion event (purchase, lead form etc.) back to a specific ad in a specific advertising campaign using very specific rules - the most common type of cookie-based attribution is Last Click Attribution.
• Pros - fast feedback, simple to set up and understand.
• Cons - doesn’t measure incremental impact, “greedy”, contains a ton of known unknowns, no estimate of uncertainty.

Experimentation

• What - Experiments are designed to measure the effect of a change to your marketing program. This could be turning on a new channel, turning off a channel in a few select GEOs, or testing a new type of creative to only a percentage of your audience.
• Pros - measures causality (incrementality), estimates uncertainty.
• Cons - difficult to set up well, requires statistical thinking and rigor, can be difficult to interpret, takes time, measures only a few things at a time, potential opportunity cost from pausing channels.

Media Mix Modeling

• What - Media Mix Modelling (MMM) is a statistical model using your historical data to estimate the impact your various advertising tactics have had on your business. Most MMMs integrate some type of lag and saturation effects to estimate the dynamics of your marketing program.
• Pros - can measure all major channels simultaneously, can provide an understanding of advertising lag and saturation effects, doesn’t have an opportunity cost.
• Cons - Very difficult to set up well, requires statistical thinking and rigor, can be difficult to interpret, susceptible to bias.

Qualitative surveys

• What - Qualitative surveys typically ask your customers where they found out about your business. This can happen at the time when a customer signs up for a demo, makes a purchase, or even during a sales call with your team.
• Pros - can reveal insights important to your target audience, often measures the most important touch point of the customer journey.
• Cons - can be biased by the customer’s experiences, qualitative data is unstructured and requires cleanup, only includes one touch point, and is directional.

Business Performance

• What - Business level output is comparing your marketing efforts to the top-line performance of the business over time. If your marketing budget increases, so too should your top-line business metrics.
• Pros - correlates with the real performance of the business, measures the entire marketing program, directionally estimates incrementality.
• Cons - no breakdown by channel or tactic, is a lagging metric.

Where is the industry today?

If all of that seems complicated, that’s because it can be - but it doesn’t HAVE to be.

Every organization is going to have different capabilities to measure marketing, but in general, this is where we see many marketing teams today.

1. Cookie-based-attribution is still the most prevalent way of measuring digital marketing performance today, but with changes to privacy laws and software settings, an increase in word of mouth and “Dark Social”, as well as a steep decline in 3rd party cookie effectiveness, cookie-based-attribution is becoming less effective to measure marketing at scale.
   
   
2. We see marketing teams are often still disconnected from business-level outcomes, with some marketing teams focused entirely on driving low-intent leads instead of high-intent sales meetings.
   
   
3. In the past few years, we’ve seen an increase in interest in both experimentation and Media Mix Modelling, but these tools are still not being used to their full potential.
   
   
4. We still see a gap in how marketing teams are actually using these measurement tools to make strategic decisions. Despite measuring incrementality, advertising lag, or marginal effectiveness (diminishing returns), many marketing teams are still relying on week-over-week attribution reports to make most of their decisions.

The future of measurement and how your organization can get ahead

We believe marketing teams that fully embrace experimentation and other statistical measurement techniques in 2024 and beyond are going to win against teams delaying this transition.

Here are some reasons why:

• It takes time to develop an experimentation culture, so building momentum is a huge long-term benefit.
• Media Mix Modelling takes a lot of time and effort to get right. Bringing on a good partner who can work with your team to help them understand the ins and outs of MMM builds a culture around real incremental impact to the business.
• A focus on incrementality will drive much stronger business results long term as it will force teams to turn off channels that look good in attribution reports but don’t drive actual business outcomes.
• Marketing teams that don’t embrace incrementality and experimentation will have a harder time demonstrating the ROI of their marketing programs, forcing them to spend more marketing dollars at the bottom of the funnel - this will lead to a “race to the bottom” in the industry with more dollars going after fewer conversions.

We believe any marketing team should be running experiments in 2024.  But specifically, those spending over $2mm on advertising per year as they can't afford not to — with so many elements at play, having a clear path to measurement and understanding of results is imperative. Ultimately we want you to get the most out of your spend and measurement across channels becomes particularly difficult if you aren’t running those experiments. 

I’ve covered a whole lot on this topic here so if you have any specific questions, I’d love to chat! Feel free to reach out to me on LinkedIn.