Advertising….Part Three – Metrics
So just was IS a metric?
Wikipedia has this to say:
Ok, so what does that mean to the average business person? It means that you MEASURE performance. Huh? That’s where you look at what is really happening, not just what you think is happening.
Metrics can be constructed from just about any data. It can be very broad, such as, “how much did I spend on advertising last year compared to the amount of revenue from last year?”. That’s helpful to know, but it’s also so broad that it has limited utility in helping you achieve efficiency and economy in your business. Other examples include things like, “How many customers came from our Ad in newspaper XYZ?”, or, “ How many people purchased the product that clicked on the FaceBook ad?”. Now we are talking! This is far better.
A/B Testing and Metrics
A special type of metric is called A/B testing. Sounds complicated, but it’s not. It’s simply looking at the different performance between two different events. They can be different, “What does it cost to get a sale from the newspaper versus a sale through FaceBook?”, or two approaches which are very similar with only minor differences. The latter is very common and the A/B test may be between two different ads that are identical, except for ONE item, such as a different headline, different color scheme or different placement of the ‘click here’ button. This is where metrics can be very useful and really allow you to fine-tune your ad-spend.
Types of Metrics
They mention seven different types of metric categories. They are:
- quality, and
Let’s go through these one by one and explore them.
Safety: Yes, it’s important, and there may be an element of “business safety” (i.e. truth in advertising so you don’t get sued), but in general, it’s not a metric that you will be focusing on in relationship to advertising.
Time: This is an important one. There are many different types of time metrics that you could be looking at with advertising: How much time it takes to covert a prospect to a customer. Maybe how much time a person spends on a web site, or a page of the web site. How much time they spend watching a video you produced. It’s all about making sure that the time you anticipate your customer will spend will be identified and your work will look to get your advertising to that level. So what is important to your advertising? It’s something that you need to think about. And once you have that thought…create a metric for it. Measure it at an appropriate interval (you want to be sure to get sufficient data to make it statistically significant) and compare intervals to intervals. Watch the trends.
Cost: Woo Hoo. This is a biggie. Everything is about money, so its important to have great metrics regarding your spend on ads and the overall cost. You have probably heard terminology like “cost per click” or “cost per conversion / sale”. Those are metrics. How much did it cost to get someone to click through your ad, or how much did you spend on the sale? This is just scratching the surface of a huge diamond. One that you will need to get familiar with, after all, it’s YOUR money you are spending. It doesn’t make sense to spend $5 to sell a $2 product.
Resources: This is one connection away from cost. Resources cost money, so it’s just really a surrogate for what it’s costing you to do something. Just like most other metrics…it’s expressed as a ratio, but it doesn’t have to be, or it may not be obvious. For example what about the time spent by a client service representative on the phone. It’s money to pay them. You may see that they spend an average of 5 minutes per call. That doesn’t look like a ratio – but it is Time / call. And don’t forget to put in a sufficient amount of granularity into your metric. Perhaps it’s the time per call for web log in issues. That could be a lot different than time per call for billing issues. See where this is going?
Scope: I don’t know what the Wikipedia authors mean here…so I am just going to use my editorial prerogative and skip it.
Quality: Hopefully by now you are starting to see how these things all go together. It’s all similar, but each metric and variation is different. For quality, we can have a metric of the number of failed products per 100 produced. Or if you want to look at it from the other side…the percentage of good units. Again, don’t forget about granularity in data collection.
Actions: This seems nebulous, but here is a great example. A/B testing. If I put up two different ads with minor differences and saw the response rate (i.e. a form of metric), which would perform better. This is a classic form of metric that is used on-line for continually tweaking the response rates of web forms.
There you have it. My “primer” on metrics. I hope you have found it helpful. It’s time for you to start developing and using metrics in your business to help you optimize your efforts…and revenue.