3.1 How do activity metrics work? - Video Tutorials & Practice Problems
Video duration:
13m
Play a video:
Video transcript
<v ->We're gonna look at activity metrics.</v> These are the things that people do, your prospects, your customers, and they do these things before they come to your website while they're on your website and sometimes even after they're on your website if you're a B2B company or a company that sells offline. And so there's all sorts of metrics listed here, don't worry about them, we'll cover all of them. But what I want you to think about is that before they come to your website, they're using tracking codes or other types of traffic metrics to say here's where people came from. Now, once they get to your website, it's your web analytics system like Google Analytics that tells you what they're doing there. And then after your website, it might be your customer relationship management system, or maybe a point of sale system that tells you what they're doing with a coupon. And so all of these things are things that you can measure with the right systems, and these are all activity metrics. And so let's look at activity metrics for a display ad. So before they click, you might know how many impressions there are so how many times that display ad was shown, how much you have paid for the campaign, what the click is, the URL that you put in, and after they come to your website, you're gonna have a tracking code that says that this person came from this ad and after you come to the website there's all sorts of things that we're gonna see the same for almost every different type of tactic. So there's gonna be a session or a visit. There's gonna be a page view. You're gonna be able to measure time on site, you're gonna be able to measure whether they added something to the cart or submitted a form these are all gonna be the same metrics on the website for every tactic. What about for social ad? Well, again look and see on the website all the things are the same, but for a Facebook ad, they're actually pretty much the same as well. And so all sorts of different ads are gonna have pretty much the same analytics. What about for a paid search ad? Well, now it's getting a little bit different, and so you might see the ad on Google and you might know the keyword volume. So how many times do people enter that keyword? You would know what your bid was for that keyword, what the average rank was, so where it would appear on the screen, number one, number five and your quality score that's Google's way of telling you how good the ad is and how good your landing pages. And again, once you get to the website, all of those metrics are the same. Let's not just list off all these metrics, let's take some examples of what to actually do with these metrics. So here's an example of a paid search campaign that's showing performance for your campaign over a three month period, and your goal is to determine whether there are underperforming campaigns, are there places that you can optimize because maybe there's high costs but very low engagement. So can you find inefficient spending, well, what would you do? Well, the first thing you might do is to sort by cost descending, so the highest costs are at the top, why? Well, because if you have inefficient spending but it's only spending pennies a week, you don't really care. You wanna look to see if you have inefficient spending for the things you're spending a lot on. Then what you might wanna look at is your click-through rate your CTR and your average cost per click your CPC. If you have a low CTR and a high CPC, that's kind of a bad combination. Now, when you look at this immediately it jumps out at you that the remote site management campaign has a very low click-through rate compared to the other campaigns. And so what kind of questions do you wanna ask now? Well, you might wanna ask which ad group within the campaign because if you're familiar with paid search campaigns are made up of ad groups, which ad group within the campaign is using up most of the budget. And maybe even break that down because within the ad group, there are keywords. And so which keywords are using the low click-through rate and using up all that budget. I don't know if you noticed, but breaking down your campaign by ad group and breaking down your ad group by keyword that's segmentation. This is the key to getting insights from your data, that segmentation step that we talked about before. So what do you see when you do this? Well, if you sort by cost descending at the ad group level, now all of a sudden you see the remote site cellular group is by far the most expensive, because this helps you answer the question of which ad group within the campaign is using most of the budget. Now, what about what we're seeing here. So this is now looking at the keywords within that ad group. And you can see that the 4G router and LTM modem keywords have a lot of impressions, but the ads only clicked a small fraction of the time and why is that? Look at that quality score one out of 10. Geez, it couldn't be worse than that, it's terrible. And so you might wanna pause those keywords and move them to a separate ad group so that you're now starting to improve the quality score of this ad group, and maybe turn your campaign around. So let's look at SEO, organic search. This has some of the similarities of paid search, you get keyword volume as before you can find out how many times your keywords are being entered, but you can also check what your average rank is. So in Google search console for example, you can see where your page tends to show up for that keyword. And again, once they get to the website, same metrics is always for understanding what they did after they got there. So let's look at an SEO example. So what can we do about this problem it looks like there is a big dip here. And so when we're trying to analyze how the traffic has performed, it looks like over a long period of time it went down for several months. So what's the big story here, it's that downward trend. And you don't see performance recovering until December after that trend of dropping in August. Clicks and bounces follow a similar pattern, although the bounces have a drastic drop off in August. And so what do we wanna do with this data? What questions do we wanna ask? Well, we might wanna ask what happened in August that caused the decrease, and we might wanna ask are there any specific pages that lost a ton of traffic? And so what if you don't have any data for that? How would you find data to figure out what happened? Well, maybe it's not in your analytics. So maybe ask the content owners, were there changes that were made to the site at that time? Is it possible that this was an anomaly that was a tracking error? Can you ask your metrics people was there a problem with the JavaScript tag that does your analytics tracking, or maybe there was a change in Google search algorithm? Maybe that's something you can research, and so sometimes the answers to your question aren't in dashboards, the answers are in asking people who might know and researching things outside of the analytics on your site. But what can you look at in your analytics? Well, you could look at that last question which is whether there are any specific pages that lost a ton of traffic, have you figured out what you're doing here, you're segmenting, you know the whole site lost a ton of traffic but now you're asking, hey, were there specific pages can I segment by page to take a look at that, so I can sort by sessions descending and see the pages that had a lot of sessions, but their sessions decreased significantly in that period. And once those pages have been identified, then maybe see if they have anything in common or if any changes were made to those pages. And how about email? So if you're looking at activity metrics for email, there's a bunch of things you can look at. So you can look at reach, so how big is your email list, you can look at opens, how many times people opened your email. Then you obviously know when they clicked and when they get to the website same old cast of characters for the metrics on the website. There is something that you would probably love to know about your email that you probably can't figure out. And that is how much of your email went to spam versus went to the inbox, but was merely unopened. I wish we could figure out how to tell you that, but you really don't know. All you're gonna know is how many you sent out and how many you got open. What you can tell though is can be very interesting, 'cause you can actually know how long the email was opened for. And you can classify them, you can segment the times and say, hey, I'm gonna call glance delete under two seconds, skim between two and eight seconds and read is over eight seconds, and maybe you wanna find out, hey, how many people actually clicked on it, and which categories do they fall into? Probably almost nobody fell into the glance delete category, but are there some people that skimmed it and still clicked on it, or did everybody read it. Those are questions that you're gonna ask by looking at the time that the email was open. And so let's look at activity metrics for web pages. So you have content analytics that tells you entrances, so which pages are the first session on the site, exits, which pages were the last in a session on the site and bounces, which were the both the first and the last, you can also look at how much time they spent on pages, scroll depth, how far down the page did they go. You can look at social shares, how was the page shared and how many other websites are linking to that page. And so let's look at some metrics for a landing page. So what can we do with this data? So this data shows a site's most visited landing pages and their performance metrics, and would you wanna do perhaps is to identify and optimize underperforming landing pages so that you increase your conversions. Well, how do we begin that type of analysis? Well, first sort by the number of sessions, because you wanna see the landing pages that had the most traffic. Why do you wanna do that? Because why bother improving landing pages that had hardly any traffic. And so you're gonna have the focus on landing pages that actually require an interaction landing pages that could actually lead to a conversion rather than say, a blog post where people tend to come to the page and quickly leave. And so if you identify landing pages that have a high bounce rate, and a low conversion rate, they might be the ones that you wanna improve. So what questions do we start asking in the face of this data? Well, the first thing we might wanna know is what the source of the traffic is. And maybe we also wanna know whether there are on page elements that we can improve to decrease the bounce rate and therefore increase the conversions. So we start by segmenting, are you getting tired of that? I'm telling you that is your key, figure out how to segment thing and you all of a sudden get your data telling you some insightful information. So what's the source of the landing page traffic that's an example of segmenting. So what you wanna do is to look at the source of data for the landing page, and you're seeing that most of the sessions are actually coming from Google. So that means it's organic search that's really driving most of your traffic. And so different sources of traffic might require different strategies, but we're gonna focus on SEO. And so how do we optimize the landing page? Well, maybe you go drag out a checklist and say, hey, here are things we need to do, strong headlines, convincing copy, clickable buttons, let's go through our checklist and make sure that we really hit them all here. So as we look at all of these different types of activity metrics, there's really a way of thinking about them, even though they may all have different names, you really breaking them down into three categories. You wanna know if they saw your marketing, if they engaged with your marketing, and whether they converted. Now, the simple one is converting because we call that conversion in every situation. But depending on the type of tactic, we've come up with all sorts of confusing names, and so what you wanna do is to think about what it's telling you. So display advertising, social advertising, we call it impressions that just means they saw it, email its reach and opens, for organic search it's impressions and maybe the rank, for paid search again, impressions and rank, and for a web page it's page views, all of those things mean the same thing, did they see your marketing? The next column is about whether they engaged, now, mostly that's clicks, but sometimes you can think of time open as being part of that for an email or referrals for organic SEO. But for a web page, there's a lot more engagement information. It's the time on page, the scroll depths, click on a new page from that page and entrances exits and bounces, and whether they're being shared in social and whether you're getting inbound links. So by thinking about answering these questions with the analytics, did they see the content, did they engage with the content, and did that content caused them to convert. That's gonna help you to understand most of what you need to know about your activity metrics.