Despite becoming increasingly digital and data-driven, marketers still struggle more than their counterparts in other corporate functions to communicate the business impact of their work.  in a consistent way. Marketing performance metrics should be regularly tracked to measure success and gauge whether goals are being met.

There is clarity to be gained in defining a framework that introduces the concept of marketing measurement levels. The tools and outputs of each level are different, and they add value at different stages in the marketing funnel. Said another way, they answer different questions for marketers. It is important to match your question to the appropriate level of measurement or you will get incomplete, or even misleading results back. 

The levels of measurement are: 

1. Plan
2. Campaign
3. Channel
4. Tactic

Here, campaign means a thematic campaign that may encompass multiple channels, like a product launch, a thought leadership campaign, or a funnel nurturing campaign.

These levels are useful for answering different types of questions. In general, as you move down the levels toward tactical measurement, the more specialized and narrow the marketing performance metrics will be (e.g. CTR, white-paper download-count, intra-channel content optimization, even CPC). And you move closer toward plan measurement, the more strategic and business-oriented (i.e. less specific to marketing alone) the results tend to be, and the easier it is to communicate those metrics to non-marketing colleagues (e.g. What’s our plan ROI? Which campaigns are performing best?). 

Example questions for each level are shown in the table below, to illustrate which types of questions are best answered by which levels of marketing measurement. The best answer to different questions (the rows in the table) lies at different levels of measurement (the columns in the table). The best measurement level for the question is shown in green.

Yellow signifies a qualified benefit or an incomplete benefit from trying to answer the question at a suboptimal level. For example, if we try to measure whether we have identified the best message for our audience using the channel as the measurement level, we can only really answer the question for a given channel. It’s true that the expression of a message and audience is shaped by the channel, but the correctness of the overall audience and message selection must be measured at the campaign level and then tuned for delivery over each channel that will be utilized to deliver the campaign.

Example: Imagine a credit card marketing campaign whose goal is to encourage people who travel a lot to switch from their current card to the new ACME card. The campaign message broadcast over the TV channel (“ACME card gives you double points for hotel bookings, and you can convert the points to air miles”) is different to the message for the direct mail channel (“Congratulations Mrs. Nguyen, you’ve been pre-approved for an ACME card. Fill in the application form and get 6 months 0% interest on any balance transferred”). These are different messaging elements of a broader campaign messaging strategy. If we only measure at the channel level, or we measure channel-first (vs campaign-first) we are going to only see a part of the picture. If we measure at the campaign level, we can draw the threads from disparate channels together into a more informative, complete measurement of performance.

Red signifies a specific risk in trying to answer the question with measurements from this level. You will likely harm your marketing insight overall if you try to answer the question by measuring at this level. 

Example: A credit union is trying to increase the proportion of its members who take out a car loan with them. They implement a campaign over billboards, print ads, local radio ads, and digital. A channel-first measurement approach will attempt to work out the independent contributions of each channel. But how do you measure how much billboards contributed vs print ads vs digital? And how to tell whether someone who clicks the Facebook ad has already heard a radio ad and seen the billboard too? It’s possible to spend a lot of effort gaining an incomplete view focused on just one channel vs measuring results in aggregate at the campaign level and asking the question, “Did more people sign up to auto-loans after we executed the campaign?” That is the first-order question.

You may notice that measuring channel-first is suboptimal for a number of the common and important questions shown above. As a practical matter, it’s important for marketers to understand what is best-measured channel-first because there is a channel-first structure in many organizations that can inadvertently lead to trying to answer too many - or the wrong kind of - questions with channel-level measurement. 

Why channel-oriented measurement is so common, and the risks to be aware of

Consider a few of the key marketing channels: TV, Radio, Print, Billboard, Digital Advertising, Social Media, Events. The tools, skills, consultant networks, platforms, reports, and so on, that are needed to orchestrate and run these campaigns are quite different from each other. 

It is not surprising, therefore, that individuals tend to become experts in executing campaigns over a subset of the channels. Events managers rarely also run digital campaigns, and vice versa. 

This reality leads to teams being structured around channel expertise, too. Since budget allocations tend to follow team structure, many marketing teams allocate budget by channel. Furthermore, there are specialized, channel-specific tools, reports, and metrics for measuring performance within a channel, that cannot possibly encompass everything that went into a campaign. For example, ad platforms report ROI based entirely on ad spend, and don’t capture the other costs that go into creating the ads that are ultimately published on the platform.

In a channel-oriented team structure, individual performance is likely to be managed by the performance of the channel. Team members naturally become focused on the performance of their channel. They want to dedicate funds to that channel. They prioritize reporting on how that channel is performing. 

While there are many good things to measure in a channel, unless the campaign is entirely run through one channel, those measurements do not typically tell the company whether its marketing campaigns are successful or not. 

Such an orientation almost can lead to a local approach to marketing measurement - “which channels are performing best/worst”, “Is social working?” - rather than a global approach to marketing measurement - “Is this campaign achieving its target metrics and ROI”, “Are we achieving our key goals?”

Conclusion

The levels of marketing measurement covered in this blog all add value at different phases of the marketing life cycle, and for different individuals in the team. It’s important for marketers to bear the macro measurement questions in mind (Is our marketing working? Are we generating a compelling ROI? Which campaigns worked best?), and to organize their marketing teams and technologies to answer them. 

Even if you have a channel-oriented organization structure and budget allocation, it is advisable to take a campaign-oriented approach to your marketing measurement, as this will expose on a campaign-by-campaign basis how channels perform, as well as enabling you to report marketing ROI on a consistent basis across all your marketing efforts. 

At Plannuh, we encourage marketers to start with their goals, defined in terms of objective - preferably financial - metrics. Then, build campaigns to achieve those goals, defining your desired outcomes in terms of objective metrics, achieved through messaging to a specific audience, delivered over the most effective set of channels you can define. What matters most is how your campaign performs in the achievement of those goals, given your budget.

-------------------------------------------------------------------------------------------------------------------

Dan Faulkner is co-author of The Next CMO: a guide to operational marketing excellence, and the CTO of Plannuh, where he is responsible for the technical strategy and delivery of the world’s first AI-powered marketing management platform. Dan has 25 years of high-tech experience, spanning research and development, product management, strategy, and general management. He has deep international experience, having led businesses in Europe, Asia, North America, and South America, delivering complex AI solutions at scale to numerous industries. Dan holds a Bachelor’s degree in Linguistics, and Masters degrees in Speech & Language Processing, and Marketing. He has completed studies in Strategy Implementation at Wharton.