Marketing might just be the most measured dimension of business performance. There is a herd of popular marketing metrics that dominates the landscape of marketing and sales plans. Just a small sample:
- Online advertising cost-per-click
- % leads from social media
- New website visitors
- Active subscribers
- Promo email open rate
- Promo email click-through rate
- Marketing cost per lead
- Lead-to-customer conversion rate
- Market share
- Churn rate
- Customer retention rate
- Customer lifetime value
- Net Promoter Score
But the existence of this herd doesn’t mean that every marketing goal is easy to measure in a meaningful way. Some of the challenging goals relate to rather intangible marketing results like these:
- Direct mail effectiveness
- Heighten brand strength
- Build customer loyalty
- Enhance market trust/rapport/credibility
- Improve reputation
- Maximise cost efficiency
Something very specific makes goals like these harder to measure than goals like ‘increase market share’ or ‘increase sales conversion’ or ‘keep customers coming back’ or ‘decrease the cost per sales lead’. Something that has to change before any forward progress toward meaningful measures can be made.
It starts with the language.
What does the goal really mean?
If direct mail is to be effective, exactly what does ‘effective’ mean? Effectiveness is not measurable, because it can mean lots of different things to different people, and in different contexts. Does direct mail effectiveness mean that lots of recipients take the call to action? Or does it mean that lots of recipients become customers? Or does it mean that more recipients are now aware of the company and its products? Or does it mean all of these things collectively? It’s going to depend on the specific purpose of the direct mail campaign, of course.
If a brand is to be made stronger, exactly what does it mean to be stronger? What is a weak brand? How would someone recognise the difference between a strong brand and a weak brand? Is a strong brand one that the target market can recall without prompting, like the maker of iPods, iPads and iPhones? Is a brand strong when its promises match what customers actually experience, like Starbucks or Subway? Is a brand at its strongest when the company or product name is adopted as part of the lexicon, like ‘googling’ to find a good vegan restaurant in Paris? Or does brand strength mean that no matter where you are, you can find it, like Coke in the remote Congo village of Conkouati? Or, is a brand strong when all of these things come together?
Stop hitting your head against a brick wall. When you have a goal that you’re struggling to measure, rather than continuing to ask the question “how could we measure that?” you need instead to ask the question “what exactly does it mean to us?”
If you feel stuck with how to measure a goal, the starting point is to make sure the goal has a specific and understandable meaning. Rewrite it if you have to, practice explaining it to others to test their understanding of it. Make it simpler, make it more specific. You might even need to turn it into several smaller goals.
When you have a truly clear and specific goal, you can then start seeing in your mind’s eye what achieving that goal might actually look like.
What is the observable evidence of achieving this goal successfully?
Let’s assume we had decided that ‘direct mail effectiveness’ for us meant that lots of recipients responded to our direct mail’s call to action. Sure, it would be nice if the direct mail campaign also generated more customers and increased awareness of our product in the market. But if the primary result of the direct mail campaign is to get recipients to take the call to action, then to measure our success, we’d have to be able to detect when they took that call to action.
If the call to action was to visit a webpage, we’d need to know when someone landed on that webpage. If the call to action was to call a specific phone number to book in a free consultation, we’d need to be able to identify when someone did make such a booking.
Say, for the ‘heighten brand strength’ example, we decided this goal really meant that our target market can recall our company name without prompting, when asked about the nature of services we provide. We’d therefore have to somehow find out whether or not members of our target market think about our company when they think about the type of service we provide.
What the achievement of a goal looks like is the magic ingredient to finding good measures for it. Performance measures are evidence of results. That evidence must be articulated before you can easily find potential measures.
It’s hard to hold back from brainstorming measures. But it’s critical that you do hold back. Unless you are very clear about the evidence first, you can easily end up with measures that aren’t really measures, measures that are not relevant enough for your goal, measures that are infeasible to implement. Or you might still fail to find any good measures at all.
The evidence you’d observe if your goal was being achieved is the best starting point for building a list of potential measures.
How could this evidence be quantified and routinely monitored (i.e. measured)?
If we meant that ‘direct mail effectiveness’ was about responses to our call to action, and our call to action was to visit a webpage, then some potential measures are:
- Unique visitors to campaign landing page
- Unique visitors to campaign landing page, as a percentage of number of mailouts
- Campaign landing page conversion rate
For our goal of ‘heighten brand strength’, it’s a little bit more complex, but the approach is the same. We’d be looking for evidence of whether or not members of our target market think about our company when they think about the type of service we provide. Some potential measures might be:
- Number of times our company name has been googled
- Percentage of our target market who name our company as the first brand recalled, when prompted to think about our service
- Percentage of our target market who name only our company, when prompted to think about our service
- Percentage of our target market who recall our most recent advertising
- Percentage of our target market who accurately describe the service we offer when prompted with our company name
Some of these measures might seem better than others, and that’s very typical of a list of potential measures. That’s because they are still only potential measures. Again, it’s important to hold back from rushing to a decision about which measure to choose, because you’ll risk taking a deliberate enough approach to sifting out the best measure.
Don’t underestimate the power of writing measure ideas down first, before assessing their pros and cons.
Which of the potential performance measures is best?
Each of the potential measures for ‘direct mail effectiveness’, where the call to action was to visit a webpage, are not equal in power. ‘Unique visitors to campaign landing page’ is certainly easy to measure, but it’s hard to meaningfully interpret without knowing how many mailout recipients there were.
‘Unique visitors to campaign landing page, as a percentage of number of mailouts’ solves this interpretability problem, and is thus a stronger measure for the goal.
We might also be tempted to choose the measure of ‘Campaign landing page conversion rate’ because that’s ultimately what we want people to do: go to the landing page and then buy the product or download the report or register for the event (whatever we sent them to the landing page for). But that would be a mistake. The conversion rate is a measure of how good the landing page is, not how good the direct mail campaign is. It probably is a measure we’d use, but we’d need to link it to a goal about the landing page, and not to the goal about the direct mail campaign.
So ‘Number of unique visitors to webpage as a percentage of number of mailouts’ is probably the best measure because it is both a strong indicator of the specific result of ‘direct mail effectiveness’, and it’s easy to measure too.
To measure the goal of ‘heighten brand strength’, not only do we need to make a call on which of the potential measures most accurately indicate whether or not members of our target market think about our company when they think about the type of service we provide. We need also to consider how the data would be collected, because these measures are based directly on feedback from what could be a very large target market. Good sampling and survey design would be needed, and that costs money.
There are two important features that any measure needs to have in order to be worthy of adopting. The first one is strength, and this means it’s going to provide very relevant and convincing evidence about achievement of the goal. The second feature is feasibility, and this means it’s going to be worth it to invest in the data collection and analysis to bring the measure to life.
A powerful measure, a measure worth adopting, is one that has high strength and at least moderate feasibility.
You’re not finished yet.
Say we did in fact choose the measure ‘Number of unique visitors to webpage as a percentage of number of mailouts’ to monitor our goal of ‘direct mail effectiveness’. What would we do next? Perhaps we’d set up a Google Analytics campaign to track unique visitors to our target webpage. And perhaps we’d get the direct mail team to email us each week on the number of mailouts they’d posted. At the end of each week, we’d get the number of unique visitors and the number of mailouts, and calculate our percentage.
And we’d probably be wrong.
There are a couple of concerns we’d first need to sort out. Firstly, when we calculate the percentage, we may be counting unique visits from recipients of last week’s mailouts and not this week’s. So our calculation isn’t accurately telling us about the response to any particular week’s mailouts. To make it more accurate, we might need to incentivise people who visit the webpage to put in a code that identifies the week of their mailout.
Secondly, depending on how accurate our mailing database is, some of those mailouts might be returned to us because of invalid addresses or people who have changed address. Do we want to include these in our calculation? Probably not.
One of the assumptions that people often make, that renders their measures completely useless, is that it’s okay to select their own samples to make data collection cheaper. But without getting professional advice from market researchers or survey statiticians, those samples end up being biased, too small to be reliable, or sometimes even more costly than they could have been.
The sample design for measuring most of the potential measures for brand strength, above, would need to be deliberated created by a sample design professional. Then the instructions for ongoing data collection and analysis really need to be defined in detail for each of the chosen measures.
To be sure your measures don’t fall prey to lazy assumptions about their calculation and data sources, it’s imperative to put in a little bit of work to make it clear exactly how to bring them to life.
It takes a little bit of discipline up front to define the details of your measures, to avoid the pain of ending up measuring the wrong thing and making ill-informed decisions.
Unlearn the bad habits and practice the new habits and your marketing measures will rock.
At the root of the struggles we have with measuring marketing performance (or any aspect of business performance, for that matter) are a few bad habits.
We use vague and weasely language to articulate our marketing goals, rather than being clear and specific. We brainstorm measures for our goals, rather than deliberately designing them as the best evidence of our goals. We make assumptions in implementing our measures that risk their misinforming and misleading us, rather than taking the time to define them thoroughly.
Performance measurement isn’t hard. It’s just that we have some ingrained bad habits that are making it hard. Unlearn the bad habits, and adopt what is actually quite a simple and logical approach, and measurement will become one of the most powerful things you do to master your marketing success.