There’s method in the metrics
A strategist’s defence of online advertising
by Leith Strategist, Jamie McAdam
Online advertising feels under siege at the moment.
The root of both their points is that online advertising doesn't work as well as we first thought. They argue that you can't accurately measure its success, or allocate it to the bottom line.
Whilst online advertising certainly has its problems, I disagree with the sweeping generalisations in both articles. There are many positives to online advertising.
The challenge is - how do we define and demonstrate these effectively? Let’s take a closer look and see if we can come to a solution.
Short-term impact. Long-term memories.
Traditional advertising is great for brand building. Big reach. Emotional connections. Long-lasting influence.
Everyone remembers a really good TV campaign. One of my favourites from the '90s was Guinness' White Horses or BRU's Snowman. They're ingrained in me now, much more so than a Facebook advert ever would be.
But that's not what online advertising does best.
It's great for brand activations. It can be used as an extension of a brand-building campaign. To echo key messages. Trigger consideration thought processes. Encourage instant purchases with a few taps of the thumb. Short-lasting influence.
It should be regarded as an ally of other channel communications. Not a replacement for.
The challenge with measuring online advertising performance
Here's a reality check though: there’s no doubt that digital metrics are convoluted. Variables are everywhere and we're peppered with confirmation biases.
People can get fixated on numbers that justify the activity, rather than the impact of it.
“This campaign had a higher engagement rate than the industry benchmark, so it was a huge success.”
But these metrics don't translate up the food chain. To the decision-makers.
Why would a Financial Director care about how many laughing emojis were on a post?
Even benchmarks can be misleading, particularly in social media. For example, they don’t incorporate how much promoted content has an impact these days. Yet there is still only one engagement rate benchmark per industry.
"How is this channel contributing to the bottom line?"
Online advertising's biggest problem is when people answer that question wrong. Trying to validate their campaign hypothesis by showing numbers with no real meaning.
This is a challenge shared by many in the industry. In the 9 years I've been working in digital marketing, we have evolved the way we measure success multiple times.
As a result, there has, quite rightly, been a growing pressure to translate tactical KPI's into more upstream and strategic business goals.
A methodical approach to demonstrating impact
How to do this? I had a moment of clarity when I found the Impact Matrix by Avinash Kaushik.
Kaushik’s approach acknowledges that conversations will change depending on who you’re dealing with in the food chain. A Campaign Manager is likely to be interested in tactical performance metrics. Whilst a Marketing Director is likely to be focused on the strategic impact to the business.
It was framing the problem in a new way. I could now see how I could translate the story to every relevant stakeholder.
You need to match up the right metrics to the right business objectives. Which are the short-term tactical ones that will progress into long-term strategic goals? More importantly, which ones will allow you to make better business decisions?
Let's say I work for the digital agency for a well-established DJ hardware company. They sell a range of music equipment for DJs like mixing decks, turntables, and speakers. Their Financial Director started to quiz me on why I'm asking for more budget for the next campaign.
"How do we know this channel is contributing to sales?"
I might say:
- We recently launched a new mixing controller exclusively to DJs with large followings on their social accounts.
- We hand picked them for testing and knew they would generate some high levels of reach.
- The feedback was a bit mixed. Mostly very positive with lots of excitement about the launch date but there was something that needed to be addressed.
- We could see from comments on their posts that there was a consistent theme emerging about where one of the buttons was.
- The controller went back into production for a minor adjustment before launch, based on this social feedback.
- The new version was sent back to the same social influencers.
- Hundreds of videos of DJs playing sets with the new controller were published all over Instagram and Youtube.
- There was mass excitement from all their fans and thousands of comments across all their social posts.
- We published a series of targeted ads with a link to pre-order the new controller on the website.
- Site traffic peaked at new all-time highs.
- Social channels were the highest referring traffic source.
- It was the highest-selling controller the company has ever had.
I'm not going to show the FD what the most liked post was. Or what the highest engagement rate was. They were all useful to me during the campaign but not to the overarching measure of success. Not when the FD wanted to know why I needed more investment to do the next one.
Here's how to approach the answer using an adaptation of the Impact Matrix:
Tactical metrics and strategic goals
There are three main metric groups to focus on. They build the foundations with tactical metrics that lead to strategic ones.
1) Visibility - Focused around impressions or reach.
2) Engagement - Focused around interactions with content e.g. comments, view rate.
3) Action - Focused around intent or purpose e.g. Click through rate, conversion rate.
Visibility is straightforward enough. Is the right audience seeing our ads? What variations could we test? Whilst there is still room for improvement, targeting has definitely improved in the age of online advertising.
Engagements take the brunt of the criticism these days, and are dismissed as vanity metrics at best. But they can also be a pawn in a chess game. They can test a strategy out. They can be functional. Measuring interactions allows me to get a sense of one particular thing at a glance; is this content resonating with my target audience? If it isn't, what do I need to do to change that?
Action metrics are the next level up. What do I want them to do now that they’re engaged? Link clicks could lead to a library of conversion opportunities, such as product pages, sign up forms and long-form content. Variations and priorities will differ by brand. I'm interested in the fundamental change in the actions that people take. Measuring a positive shift in the actions they take is a confident stride forward.
Let's bring it back to the example scenario now.
Here are the top three metrics to pull out to build into the matrix:
1) Visibility - 1,000,000 people matching the target audience's profile are reached through DJ influencers and targeted ads.
2) Engagement - Invaluable feedback that leads to simple product adjustments.
3) Action - 300% increase in pre-orders of the new controller on the website. Membership sign-ups increased by 600%.
The focus is on which tactical social metrics lay a solid foundation. Now I’m starting to show how tactical metrics can contribute towards strategic goals. I can start to have conversations about how to model this activation into revenue. I’m speaking the same language as the Financial Director.
I’m not denying that mistakes can be made with online advertising. But not everything is black and white - traditional or online.
We shouldn't paint with a broad brush or dismiss entire channels if, with the right approach, there’s demonstrable value to be gained.
There’s plenty of evidence that online advertising can be great for brand activations. It can be a tactical weapon in the armory that builds up to strategic goals. It can contribute to the bottom line.
And it’s on us, as digital marketers to prove it, which is why a framework like the Impact Matrix is so useful.