The point of measure marketing is to determine if a marketing strategy is working and how the strategy is performing. Marketing metrics (and the marketing industry) is filled with acronyms, ratios, buzzwords – thing that obscure what they mean and what they measure.
In this series, were going to take a look at what marketing metrics are, why they are important and how they can be used to help optimize a broader marketing strategy.
Cause and effect
Before I dive into the metrics of marketing, I think its important to share my definition of marketing. I believe marketing is the attempt to answer the following question with different creative, answers:
What can I do to influence someone to behave in a certain way?
In marketing, the goal is to design a system where the actions I take influence the behavior of my target audience. I want my marketing effort and actions to be the cause that produces a desired effect.
Marketing cause and effect examples
As part of a marketing plan, I might try a number of different strategies to influence my audience.
Below are a few example strategies drawn from typical marketing playbooks that help illustrate the cause / effect relationship of marketing.
- Write a blog post so that people are better informed of how my product fits into the industry.
- Pass out free samples of a new chip flavor on the street so potential customers buy this new flavor when they see it.
- Run an AdWords campaign to get more people to buy my product.
- Run a display campaign to raise my brand awareness.
In each of these scenarios, I take an action with the goal of influencing my target audience to act or think in a particular way.
What should I measure?
This is often one of the hard parts in marketing – knowing what to measure. With the industry so full of metrics and numbers, it can be confusing how to contextualize what specific metrics measure and why they are important.
At marketing measurement’s core, the concepts are pretty simple.
We want to measure the input and the output of our marketing efforts.
Input measurements evaluate the actions I control and take when marketing to someone and output measurements evaluate the actions / thoughts / behaviors of the audience after being exposed to my marketing.
With these two measurements, we can determine the impact that our marketing strategies have on the desired behavior we are trying to influence.
Measuring the inputs
Often times, the metrics for measuring an input come in two forms: cost and reach.
Cost is easy to articulate – its simply the amount that has been spent executing a strategy. Reach is the number of individuals a particular strategy has the opportunity to influence.
For example, when handing out samples for a new type of chip, some candidates for input measurement include:
- time spent passing out chips (cost)
- value of product given away (cost)
- number of people who were given an item (reach)
- number of people who saw us giving away free items (reach)
In the case of a blog post, we might be similarly be interested in measuring:
- cost of producing the post (cost)
- cost of distributing the post (cost)
- number of people who arrive on the page (reach)
- number of people who read the post (reach)
Measuring the outputs
For some marketing strategies the output is easy to measure – for instance, how many sales were driven by people who clicked on my AdWords campaign – but oftentimes, it is difficult to quantify the impact of a marketing strategy.
Why? Because its not something that I directly control (like the cost) or have direct insight into (like the number of people reached by a campaign).
Unlike the cost of a running a marketing campaign or counting the number of people I have reached, it is often difficult to know if my campaign caused a user to have a positive affinity toward my brand. Or if the consumer who tried the sample of my chips will buy them later.
However, even with this ambiguity, it is even more important that I get this measurement right because output measurement is the only thing that empirically lets me know if my marketing strategy is successful or not.
Getting output measurement right
Now this is the hard part — fighting through the ambiguity of measuring audience behavior to ensure the metrics we choose measure the output.
So how do we figure out what the right value to measure number is? A trick I often use is to start at the ideal metrics and work backwards.
For instance, in the chip example, I want to measure the number of people who buy our new chips after trying them.
On the surface it seems straightforward, but logistics often prevent the proper collection of data. For the correct data to be collected to measure the impact, I would have needed to do something like hand out stickers to people who took a chip sample, find the people who wore the sticker at checkout and count the number of people who purchased the new chip. This data collection is likely intrusive and costly and still misses a percentage of consumers who may purchase on another day.
Instead, in marketing measurement, proxy metrics are often used. As an example, if I work backwards from the act of purchase, I can come up with a handful of proxy metrics that could help us measure the effectiveness of this chip giveaway campaign.
One of these proxy metrics might have observed the number people who finished the chip bag samples. Our assumption is that, if a person liked the chips, they are more likely to purchase the new chips than someone who did not. With this proxy metric, I could then estimate the number of purchases based off a measurable piece of behavior that occurred in response to my marketing effort.
If I’m not able to collect these metrics, I would move further upstream. At each layer we move upstream, we lose accuracy in our estimate.
The major thing we want to try to avoid is measuring our campaign based solely on the input metrics.
For instance, using the above method (moving upstream), I could estimate the number of people who like chips on average and the number of people that buy the chip who like the chip, then all I need to do is count the number people I pass out samples to to get an estimate for the number of new customers I have generated.
On the surface this seems good because it makes measurement easy – I only need to track one metric, but robs us of the ability to measure the impact of any changes I might make to the strategy.
For instance, do more people buy chips from our giveaway if we change the color of the bag or the positioning of the giveaway booth. If we only track the number of items distributed, we lose the ability to run these experiments, gather meaningful results and iterate on our strategy.
Measuring the inputs and outputs of a marketing campaign enable us to optimize our marketing strategies and properly allocate resources. In future posts, we will dive into how we can setup good marketing measurement and how the allocation of resources across strategies could work.