Ask the Expert: Andy Crestodina on the Crucial Metrics for Full Funnel Marketing
December 9, 2015
As Co-Founder and Strategic Director of Orbit Media Studios, Andy Crestodina has 15 years of experience in web design and interactive marketing. Andy is a sought-after speaker on content strategy, search, and analytics. His bestselling book, Content Chemistry: An Illustrated Handbook for Content Marketing, is currently in its third printing.
Andy has a passion for helping marketers choose the right metrics and use thoughtful analysis to improve outcomes. We are extremely grateful that he was willing to give us a few minutes of his time. In this interview, Andy identifies the metrics that are most crucial for measuring and improving your full funnel marketing efforts.
Read on to learn the difference between metrics and analysis, how to determine which metrics are most important for your program, and how to address the attribution problem.
Q&A with Andy Crestodina, Co-Founder and Strategic Director of Orbit Media Studios
LinkedIn: How do you define the difference between metrics and analytics?
Andy Crestodina: It’s a big difference. Metrics are just numbers. Analysis is making decisions. You can get all kinds of metrics on lots of pretty charts, but a metric itself doesn’t affect your marketing. Analytics implies decision-making. It implies that you are actually going to take an action that will make a difference in your marketing.
LinkedIn: Is it a fair statement that Google Analytics should really be called Google Metrics, since the analytics need to be applied by the marketer, not the tool?
Andy Crestodina: It’s a totally fair thing to say. In fact, in the Google Analytics Individual Qualification (GAIQ) test it doesn’t show you how to do analysis at all. It doesn’t even touch on it. But then again, it’s really hard to teach analysis. Analysis is really applying the brain to the metrics—something only a human can do, not a piece of software.
LinkedIn: What is your desert island metric? The one metric your business can’t live without?
Andy Crestodina: An easy answer would be revenue. Businesses literally cannot survive without it. But from a marketer’s perspective, if my Google Analytics boat is sinking and I can only grab one metric to take to shore, I’d actually cheat and take two. Top line total traffic, and then the bottom of the funnel conversion rate. It’s the analysis applied between these two metrics that really drives marketing success.
LinkedIn: What advice would you give to marketers for determining what metrics are the most important?
Andy Crestodina: First, it’s definitely on a case-by-case basis. There’s such variation across businesses, goals, objectives and key performance indicators. For example, let’s say you’re a marketer at a brand new company that invents things that have never been done before. Search traffic isn’t really going to help you, because people aren’t looking for it. It’s not likely you’re going to fill the top of the funnel with tons of search traffic from people looking for what you do.
Similarly, if you’re a B2B company you might see quickly that visitors from Facebook don’t convert into people who take action. So you’re not going to care all that much about that traffic source.
So it’s definitely a case-by-case basis. However, what almost every company learns is that visitors who have subscribed to your blog or newsletter are almost always visitors who have the highest conversion rate. If you’re looking to take a broad brush to paint with, I’d recommend focusing on conversion rates from visitors to subscribers, and try to maximize the traction you get on your blog.
LinkedIn: What would you say are the critical upper funnel metrics every company should be looking at?
Andy Crestodina: Total traffic, but a layer deeper: traffic from each channel. It’s there you can look at the three main channels: search, social, and e-mail.
The best metric for search is total visitors coming from organic. But you also need to be looking at the total number of non-branded key phrases Google is sending your highest value traffic.
For social, look at metrics like follower growth and total shares to understand if people are really engaged with your brand. Vanity metrics often have a bad reputation, but if you’re watching social traffic you’ll want to understand the quality of traffic and follower growth specific to each network.
For e-mail, you’ll need to understand your rate of subscriber growth and how consistently you’re getting traction from that: activity like open rates.
LinkedIn: What would you say are the critical lower funnel metrics every company should be looking at?
Andy Crestodina: Find the metric that correlates a visitor turning into a lead and focus heavily on that. For some, it might be time on site. I find that these are sort of like the Golden Box of metrics that if analyzed can give you actionable insights to optimize your marketing.
For example, the average person visiting a commerce site makes a purchase after seeing six to seven pages. If people are seeing 10 or 12 pages perhaps that means your site is confusing. Or if people are only seeing two to three pages perhaps they’re not finding what they’re looking for, or the source of traffic isn’t delivering the most targeted visitors.
So, the bottom of the funnel metrics will definitely be around the conversion rate from visitor to lead, but the way you look at that conversion will vary from business to business.
LinkedIn: How often are you checking your data and adjusting your marketing accordingly?
Andy Crestodina: I have the Google Analytics app on my phone, so I’m looking at our data two to three times a day for general performance, not necessarily making decisions based on that data.
I’m looking at things like, ‘is this a good or bad day on the site’ or ‘is it good or bad on the blog,’ to understand if there’s some outlier action that’s influencing negative or positive activity.
But generally speaking we’re really doing analysis weekly. Typically, it will begin by asking ourselves a question that then leads us to dig in and find the answer to help better our marketing. As an example, right now we’re doing a rolling content audit for lot of the pages on our site. How do you prioritize which pages to go back and rewrite? Well, we’re looking at the pages that get the highest amounts of traffic. So, if you are going to go back and start editing old pages on your site, you can really do any page. But why not take the time to figure out which pages will give you the greatest benefit by asking your analytics first?
LinkedIn: Attribution can be one of the biggest challenges for marketers. How do you solve for that?
Andy Crestodina: It’s a massive flaw in everybody’s analytics. The fact that you have a single user on multiple devices and vice versa, there’s just not a good way to track how people are moving through a buyer journey.
People are coming from apps, social, websites, and so on. And sometimes tracking from email looks like referral traffic and sometimes direct traffic is really traffic from social. So, not only is attribution a problem but just accurate traffic sources data is a problem.
While I truly believe we’re a long way from being able to get accurate data for attribution, I would say it’s more important to be able to make a good decision than it is to have perfectly accurate data. You don’t need to know the exact buyer journey for every visitor. If you did, would it necessarily change your marketing? It might not. But if you know your most expensive keywords are sending people to a landing page with a high bounce rate—that’s helpful! If you know that people who have visited your site five times are much more likely to buy—that’s helpful!
But the really specific attribution cases people are trying to solve are sometimes beyond the point of diminishing returns. Do your best to check the value of traffic in general from every channel, but don’t ever expect to get perfect data or to ever truly know the exact path for attribution.
To learn more about measuring your marketing efforts to get better results, read The Sophisticated Marketer’s Crash Course in Metrics & Analytics.