John Wannamaker has a famous quote about advertising…
“Half the money I spend on advertising is wasted; the trouble is, I don’t know which half.”
He said this in the early 20th century. This was long before computers, the internet, or most other modern tools to measure media consumption.
The quote, though, is still often recited. It fuels the false belief that advertising and marketing effectiveness can’t be measured.
You can measure marketing and advertising performance. In fact, you must if you expect to stay in business for any length of time.
What not to measure
When many people start measuring, they measure things that don’t matter.
By “don’t matter,” I mean things that have little relationship to the success of the business. Many of these are “vanity metrics” because they make people feel good about themselves.
So, what doesn’t matter?
Here’s a partial list:
- Pageviews
- Likes
- Followers
- Shares
- Retweets
- Clicks
- Connections
- Impressions
- Reach
Why don’t these things matter? Two reasons.
- They don’t link to an actual business result (i.e., selling something). You need to jump through a lot of steps to make any connection between those metrics and your bank account. Even then, the relationship to results is very weak.
- They don’t tell you whether or to what degree your marketing efforts are profitable.
The main point of marketing and advertising is to generate sales, right? Doesn’t it make sense then, to measure how well those efforts do at generating sales?
We can also all agree that spending more than we will make to get a customer is a good way to go broke.
Enough about what you shouldn’t measure. What should you measure?
The metrics that matter
The metrics that matter are your cost per X, where “X” is each key step of the marketing funnel.
We’ll start with the most important, Cost per Acquisition.
Cost per Acquisition (CPA)
Cost per acquisition is how much it costs you to get a new customer.
This tells you how effective your marketing and advertising is.
First, you want your CPA to be less than your Customer Lifetime Value, which I talked about last week. There are certain cases where you might be willing to spend more to get a customer than they’re worth. Those are limited, edge cases. For now, we’ll stick to the idea that you want to make money on a customer.
The first thing you must decide is what proportion of the CLV are you willing to spend to get a new customer. The answer depends on your costs, your desired profit, and your marketing goal.
We’ll use a target of 25% of CLV for our example. So, if our CLV is $1,000, that means our target CPA is $250 ($1,000 * 25%).
The key metric we’re going to use to decide if our marketing is working is whether we can get a new customer for $250 or less.
To calculate the CPA, divide the total amount you spent on marketing and advertising by the number of new customers you got. This simple method gives you the average CPA. It’s a good place to start. It can be quickly obtained from your P&L and customer records.
In practice, you would do this calculation for each individual marketing activity. That means, each different type of advertising or promotion you do. Within each type, you’d also look at individual campaigns, target audiences, and advertisements. This requires more sophisticated record keeping.
Each element gets judged on two things.
- Is it at or below the target CPA?
- How does the CPA compare to the performance of other activities?
If an element of your marketing is costing you more that the target CPA, then stop doing it. Or, change it to try to improve the CPA.
When comparing how different activities perform, to do more of the elements with the lowest CPA and less of the elements with the highest CPA.
Cost per Lead (CPL)
The cost per lead is like the cost per acquisition. Except this applies to leads instead of closed sales.
You want to know about how you’re doing at different points in the funnel is so you can diagnose what isn’t working.
I’m assuming a pretty simple funnel here. Prospects go from becoming a lead to becoming a customer. For many businesses, there’s an extra step or two before they become a customer.
For example, there may be the need for a sales call and/or a proposal because of the type of service you offer. In that case, the funnel has three or four steps (lead, appointment, proposal, sale). You would want to create a “cost per” metric for each step in the funnel.
The calculation is the same for each.
The process for setting a target value for your “cost per” metric starts with the target CPA and the expected conversion rates at each step of the funnel.
The target value for each step of the funnel is the target value of the next step multiplied by the conversion rate. You need to work backwards through the funnel from the sale to the top of the funnel for the method below to work.
Let’s say you have a three step funnel; Lead, meeting, and sale. Your target CPA is $1,000.
You know from past experience you will close 20% of the people you meet with. Your target cost per meeting, then, is $200 ($1,000 cost per acquisition * 20% conversion rate).
You also know that 10% of your leads schedule a meeting. Your target cost per lead is $20 ($200 cost per meeting * 10%).
To track your marketing performance, you look at these steps exactly the same way you looked at CPA.
- Is it at or below the target?
- How does the cost per compare to the performance of other activities?
By looking at each step in the funnel you can diagnose where the problems in your marketing effort lie.
Is your cost per lead great, but your cost per meeting is too high? You need to look at how you are trying to get leads to schedule a meeting.
Cost per meeting looks good, but cost per acquisition is too high? There’s something going wrong when you try to close.
Why is “cost per” better?
I hear a lot of people, especially those advertising online, ask “why can’t I just use click rates?”
Because clicks aren’t always related to making a sale, or to how much money you make.
Which would you rather have?
An ad that gets a 15% click through rate, but only converts 1% of them to sales? Or an ad that gets a 4% click through rate, but converts 10% of those to sales? (I did the math, the 4% click through rate ad gets you 166% more sales.)
Let’s go one step further and look at some revenue and cost figures. Let’s say a sall is worth $1,000 and you are paying $5 per click.
Your 15% click through rate ad will need 100 clicks to get a sale, making your CPA $500.
Your 4% click through rate ad will need 10 clicks to get a sale, making your CPA $50.
If you were only measuring click rates, you’d think you hit a home run the 15% ad, when in reality you struck out.
What are you waiting for? Time to change how you measure your marketing.
Need help? Let me know.