Presented on May 30, 2013
In this webinar, Search Mojo’s Janet Driscoll Miller discusses the search engine marketing metrics that matter, and the variety of ways to measure them using tools like Google Analytics and Google Webmaster Tools, as well as marketing automation and CRM software.
Presenter: Janet Driscoll Miller, President and CEO, Search Mojo
Presented on: May 30, 2013
Hi everyone and welcome. Thank you for coming to today’s webinar, Search Marketing Metrics that Matter. I’m Kari Rippetoe, content marketing manager at Search Mojo, and I’ll be serving as your moderator for today’s webinar. Before we get started I just have a few reminders for you. There will be a Q&A at the end of today’s webinar, so if you have any questions for our presenter today, please enter them in the GoToWebinar questions box at the right of your screen. Also, as always, we are recording this webinar. Once the full recording is available you’ll receive a follow-up email which is usually by Monday at the latest. Finally we encourage you to tweet about today’s presentation; any interesting insights or takeaways you’d like to share. If you do please use the hashtag #mojowebinar. Plus you can also follow us on Twitter at SearchMojo.
Now I’d like to introduce to you today’s presenter. Janet Driscoll Miller is President and CEO of Search Mojo. She has nearly 20 years of marketing experience and in addition to her work in search engine marketing, Janet has a background in marketing communications. She holds a degree in public relations and communications from James Madison University and she’s a frequent speaker at marketing conferences and writes for several blogs and print publications.
Search Mojo was founded in 2005 and specializes in all things search marketing, including; SEO, pay-per-click, social media advertising, online reputation management and content marketing. Search Mojo is headquartered in Charlottesville, Virginia and we also have an office in Charleston, South Carolina. We’ve been featured in several marketing publications and blogs and we also speak at several conferences, including SMX, MarketingProfs B2B, and PubCon. Our clients include a variety of B2B and consumer brands, nonprofit organizations and educational institutions.
Before I hand it over to Janet I’d like to ask you our first poll question of the day. That question is, “Are you still measuring SEO success by using rankings?” If you could select one of those three answers, yes, no or perhaps you may not know at this point. Then we will display the answers very shortly in just a few seconds.
It looks like it’s a little half-and-half. Some of you are still measuring SEO with rankings, some of you aren’t. Janet has some information to share with you about how you might go about measuring this, so I’ll turn it over to Janet.
Thanks Kari. The first thing we’re going to talk about today is the reason that we might not want to always consider using rankings as our measurement for SEO. Really, search has evolved. That’s really one of the primary reasons that rankings may not be the best way for you to measure your organic search efforts going forward. So let’s talk a little bit about what I would consider some of the reasons why.
First of all, this is an eye tracking study that was done on Google results. What you can see here is the red area is where people look and click the most and the darker blue areas are where people look and click the least. In the past we always knew that top rankings were getting the most clicks because that’s where people looked the most. Ergo, it would make sense for us to measure and understand the potential for visibility and traffic, and thereby SEO success, based on rankings. Because, hey, if you’re at the top you’re clearly going to get more clicks and more visibility.
However there are really, in my mind, five major disruptions that occurred, especially within Google, that really changed the ability for rankings to really be the most accurate measurement. Whether you’re in-house or you’re an agency, it really probably isn’t the best measurement of your SEO success. You may be selling yourself short in a lot of ways too, so let’s talk about what those five disruptions are.
The first one is Universal Search. If you’re not familiar with what Universal Search is, let me take you over to my browser and I’m going to show you an example of that now.
Here’s a search that I did for Siberian tiger as just a generic search. What you’ll notice here on these search results is of course, that we have images of the Siberian tiger and if I scroll down, you’ll see there’s video of the Siberian tiger and you’ll see news for the Siberian tiger. This is an example of how Google evolved its search rankings and started to show more results other than just generic webpages. It shows lots of different types of content.
Google’s effort, of course, was to try and do this so that searchers could get the type of content they were looking for. When you’re looking for Siberian tiger they’re trying to anticipate what exactly might you be looking for. ‘Are you looking for images? Are you looking for news? What exactly is that searcher looking for?’ They’re trying to predict to some level what you’re searching for.
As you can see here, one of the challenges with looking at rankings is the number three ranking which is the [World Wide Wildlife Foundation], wwf.panda.org, which would technically be our number three ranking, really isn’t the number three ranking because the images show up as sort of the number three ranking. It really almost makes that one like the number four ranking. But when you pull rankings, guess what, panda.org is going to look like the number three ranking in your ranking reports. So it’s a little but challenging to really understand truly where you’re placement is for that webpage on the page to begin with.
The other thing is, what if you’ve got an image ranked in this particular universal search, or a video, or a news article that showed up as the number three ranking? You wouldn’t necessarily see that through traditional ranking reports. So Universal Search definitely presents some types of challenges when you’re just trying to pull traditional rankings as we’ve known them in the past.
The second disruption is Personalized Search and that’s become a lot more prevalent in the past few years. Google’s been doing a lot more, even Bing, to really personalize your search results based on lots of things like your geolocation and so forth, so let’s take a look what that looks like.
If I do a search for, as an example, pizza, and again, Google’s trying to predict what I’m looking for and why I might be searching for pizza, you’ll see here that because I’m based in Charlottesville, Virginia, today when I did this search, you’ll see that the second result here says, Pizza, “dominos.com/virginia/charlottesville.” I’m getting that solely based on the fact that my computer is located right now where I’m doing this webinar from, in Charlottesville, Virginia. So Google is geolocating me and saying, “Well, when you’re searching for pizza you’re probably looking for something close to home, so let me show you a result that’s close to where you are.” If I were in Washington D.C., I would see something different, so Google is definitely trying to predict things and personalize my results better.
You’ll also notice that part of this personalization is map results. Here you’ve got Belmont Pizza and Pub, which literally, I could walk there from where we are now, or Mellow Mushroom which is a couple blocks away really. Those locations closest to me, it’s going to show closest for me in the search results as well. It’s really personalized that result. Again, especially if you’re a local-based business, if you think of somebody with franchises or you’re a car manufacturer and you have dealerships, or you have franchises throughout the country, the search results that I get in Charlottesville may be different than you get in L.A. or D.C. because it’s based on where I’m currently located.
So if you have geographic locations, if you’re a local-based business with local-based settings in offices, then you’re going to have this challenge. In a way, it’s real beneficial for those who have local-based businesses, but know that if you’re looking for let’s say, “How does the Charlottesville URL rank in L.A.?” you’re not going to be able to see the same results. It’s going to rank in that particular geolocation.
One of the other challenges around ranking software is some of them don’t have the ability to look and do a search based on where you’re located. Depending on the software that you might use, they’re not always able to say, “Here’s how results look if you were in Charlottesville versus if you were in Washington D.C. or L.A.” So that’s another [inaudible 00:08:46] challenge. We’re trying to almost to make these rankings generic essentially and show what it would be like across the country, which isn’t really the case if you’re in a local-based type business.
The third disruption is Google Search + Your World. This of course came out when Google+ was launched. We have Google Search + Your World which is a very long way of saying… another form of personalized search in a way. Basically what that’s doing is when you log into Google, which you could be logged in through your Gmail account or however you might be logged into Google, what you’re going to see is this entry here that has the little man beside it. Now, if you look up here to the upper right you’ll see there’s two options here. There’s a little man option and then there’s a little world option.
The world option will hide those results that I only am seeing because someone I know shared it. Right now I’ve got my little personalized options right there and if I look down here you can see I happen to follow allrecipes.com on Google+. And because they shared it socially and I followed them, guess what, I’m seeing that result. I might not see that result if I turn off my personalized settings. This is showing me two different versions. That’s what Google Search + Your World is. It’s showing a lot more social results based on who I’m friends with.
That too has really thrown things out of whack for doing ranking searches. Let’s say your brand has lots of followers on Google+, you may not know if they’re seeing things you share because those things may be showing in the rankings but you don’t hardly have a good way to measure that through just traditional ranking reports. So your brand could be ranking very highly through social shares on Google+ for people who follow you, but you may have no idea from a traditional rankings report where that shows up in the rankings.
The fourth disruption is Rich Snippets and rich snippets have been around for a while. There’s lots of different types of rich snippets out there. This is where you have a thumbnail typically, or some other information. For instance, if we go back to these results here, a rich snippet would be these stars as a rating and then this image as well for a recipe. There’s lots of different types of rich snippets, but I’m going to talk to you a little bit about how that’s changed rankings results. We used to be able to believe like I said on the previous slide, that high rankings equaled the most clicks, but that’s not always the case and I want to show you this next slide.
Rich snippets has really changed the game in where people are clicking. If you’re trying to measure how effective your SEO is, rankings may not be the best way to look at that because if you have a rich snippet showing up, you may be actually getting more clicks than you would without a rich snippet. You can’t automatically assume the number one ranking gets the most clicks anymore, or the number two, or the number three, or the number four, etc., in a certain order. You used to be able to somewhat surmise that based on your ranking, but you may not be able to do that now.
Here’s an example of when you have no rich snippets. Look at how many clicks as a percentage of the top four results clicks that the number one result gets. It gets 68.9 percent of the clicks and the second result gets 23.3 percent, third result, etc. So you see a little bit less each time the further you get down the page. However, if you use the authorship snippet as just an example, notice how the third and fourth result, yellow and green, and even the red which is the second result, garnish a lot more of the clicks than they used to. So you can’t just make an assumption that the number one is going to be necessarily the most clicks because this example shows if the authorship snippet were to appear on the number one result, and the number two, and the number three and the number four. But what if there is no snippet on the first one, but there is on the second one? Does that one get more clicks?
What we know is snippets help improve click-thru rate. It could be, technically speaking, that your second result if it has a rich snippet, could very possibly get more clicks than even the first result. That’s a possibility. Knowing that this view really starts to change how people click, we can’t really say for sure with certainty that people are going to click on the first result always the most. That’s something I want you to take with you. In this particular case when snippets are showing on all four results, they still click on the first one the most, but look at how much that shifted. It may not always be the case that it always means the first one gets the most clicks.
The final disruption I want to talk about is Mobile. This is huge this year. Mobile is really changing how we view results and how Google is showing results and so forth because again with that geolocation, it’s even to a greater extent on mobile that you’re seeing different results because Google is trying to define what could you possibly be looking for where you have a mobile search from a particular device and from a particular location for a particular word. They’re trying to anticipate better what you might need.
Let’s take a look at a search I did earlier. On the left is a search from my mobile phone for the word pizza and on the right is a search from my desktop for the word pizza. You’ll notice that on the right-hand side I get Charlottesville – Pizza Delivery from Dominos, I get Wikipedia, etc. and the first entry that I get is Belmont Pizza, right there, from my desktop. But when I do it from my phone I get a different result. Notice that the Wikipedia is gone. It’s further down the page on my mobile result. Google actually prioritizes map results, although it didn’t do a great job from a location standpoint on this one because Belmont Pizza is actually closer than Papa John’s.
What I’ll say here is they move those results further up the page because they’re anticipating that if I’m on my mobile phone, I’m looking for pizza right now. I’m looking to go somewhere and eat pizza. I’m out and about on my phone versus if I’m on my desktop where maybe I’m just looking to order pizza or I’m looking to do pizza research, but I’m probably not doing pizza research on my phone. So that’s their anticipation and that’s why they rank things differently. Again, map results and local results tend to rank higher directly on mobile than they do on the desktop.
Why is this important? Again, the rankings are going to be different for a mobile site versus a desktop site. How things rank and where they rank may be different. Google may serve up results again differently based on the fact that you’re searching from a particular device or from a particular location than they would if you were in a static area.
If you don’t think mobile’s important, let me just say that even on our own website at Search Mojo, so far this year 13percent of our website traffic comes from mobile. Now you might say, yeah, but as a search engine marketing company, the rest of the visitors probably come from tablets. Not the case. In fact, a large majority of visitors to our site come through the iPhone. It could be they’re looking for directions or what-have-you, but again, those people visiting our site are looking for us through mobile, so we need to start really paying more attention as businesses to mobile. We actually have a webinar about that coming up in two weeks and that’s something you may want to take a look at because we’re seeing a lot of mobile traffic increase. But again, from a rankings perspective it’s very difficult to say specifically, “Rankings are always ‘X’.” That’s not always the case, they’re always changing.
If I could leave you with nothing else about this, I want to just leave you with the idea that these five disruptions have really changed the fact of how things are actually ranked in Google and how you see them. I want you to understand that trying to genericize rankings doesn’t always do you any good because you might actually have better rankings than you think. If you’re goal is to drive people to your pizza parlor, and you’re getting a lot of searches from mobile through that, and you’re ranking well there, but you don’t know that because you’re doing a rankings report traditionally, you may not realize exactly how well you’re doing with SEO. You might actually be doing a much better job than you realize and you’re giving yourself credit for. So you want to make sure that you are measuring from different angles and different ways and not just using rankings. It’s not that rankings are all bad, but again, they’re really not telling the true, full story that they could be telling.
The first question you have to ask yourself is what are you trying to achieve and the answers are likely leads or sales. I’m going to go ahead and Kari now who’s going to put out a poll and we’ll find out what you all are trying to rank for.
Yes, indeed. Our next poll is exactly that. What ultimate goal are you ultimately trying to achieve with your SEO efforts? Just think of the one thing that you ultimately want from everything that you’re doing with search, with SEO specifically, and go ahead and input your answer. I’ll just give you a few seconds and then I’ll share those answers with you.
Okay, great. Thank you so much for your answers on that. I’m going to go ahead and close that out and share with you. This is no big surprise, a lot of you ultimately want to generate more leads from your SEO efforts and who doesn’t, but obviously sales is a big one too. A lot of you might be ecommerce sites so you do want to generate those direct sales which Janet will be talking about a little later on as well, but I’m going to turn it back over to her and she’s going to talk about goals.
Thanks Kari. If our goal is leads and sales, which I fully expected you all to say because that’s really what drives us all, then really, are rankings the best measure of your success? If the end goal is sales and leads, then are rankings really telling you that full story? The answer is no, not at all. It’s not telling the full story. Especially if you’re an agency, or even if you’re in-house, and you’re trying to report to your boss about your success of your SEO efforts, the reality is that just taking your boss rankings isn’t going to be enough. You really need to be thinking outside of the rankings box and be thinking about how you’re contributing to the bottom line and how you’re initiatives with SEO are generating basically a return on an investment, and rankings don’t tell you that story. So let’s talk about some ways that you can measure.
How should you measure? One of the things I simply tell people is kind of like what we were talking about a minute ago, what do you really want to measure and how do you define that? The first thing I like to tell people is I would start at the end. What is your ultimate goal and then work backwards from there and figure out, “How do I measure that?” If your end goal is sales and leads, then what are some effective ways to measure the path to get from a search engine or your SEO efforts all the way to that end goal?
The first thing I would say that you want to measure that’s really fairly simple and straightforward to measure is your organic site traffic. What it does is it identifies if you’re bringing in traffic via organic search. It’s a very simple measurement. Anyone can do this really from just about any analytics tool. This pie chart here is an example of showing organic search and SEO efforts as a percentage of all referring methods to your site. In other words, of all the types of ways that people come to your site, how much does organic search account for that? In our case, the blue area is organic search and I think the red area might be email or…I’m not sure which one. I’d have to go back and look and see what our different pie chart pieces were, but you can compare that to your efforts in SEO and notice organic search makes up the majority of the site traffic to our site. This is a great way if you’re trying to justify a return on investment that you’re actually generating some value for the company through your SEO efforts. This is a great way to show how your value even compares with other efforts that are going on in the company and how deeply valuable that organic search is.
But I do want to caution you. Caution. Caution. Caution. I see a lot of folks say, I’ve got to know what keyword generated this and yada, yada, yada. Okay that’s all lovely, but we have some challenges. I think most people probably who use Google Analytics are aware of the situation with “not provided”. Especially if you’re using Google Analytics, but really any analytics package, no analytics package can tell you what those keywords are anymore with any type of certainty because what’s happened is Google has started to encrypt organic search. Most people are mostly concerned about Google search results because it is the gorilla of the search engine industry and so you’re going to see a lot of “not provided” and it may be very difficult to track success back to an individual keyword.
The other thing is you may be optimizing for keyword A, but you may actually get residual effect on keywords B, C, and D. While you may be optimizing for one word, you may get a secondary benefit on other words that you weren’t intending to optimize for but brought traffic to you. So it can be very difficult to say with any type of certainty the really specific keywords that you are optimizing getting traffic on.
You can use Google Webmaster Tools for a lot of this data though. You can also import your Google Webmaster Tools data into the queries area in Google Analytics. It’s a two-step process. It’s pretty straightforward. You can look at it, but I caution you about getting too granular here. Keywords are on a level that is super granular and I’ll tell you, trying to map that back, especially if you have a lot of data, can be very difficult. Try not to get too granular even if you want to know the success on particular keywords. Remember, our end goal isn’t keywords, our end goal is sales and leads.
Now, I’m going to hand it back to Kari for another quick poll.
Yes, we have lots of polls today in the webinar. Our next one has to do with Goals and ecommerce within Google Analytics. If you use Google Analytics you might know what Goals are and you might have those set up. You might have ecommerce set up in Google Analytics, or if you don’t or you’re not sure, you can go ahead and choose one of those as the answer. Just let us know which one of these you’re using, and you might be using all of the above. We’ll just give you a couple more seconds. In fact, I’m going to go ahead and close that poll out now. A lot of you actually aren’t using those quite yet. There are a few of you that do have goals in ecommerce setup in Google Analytics, but Janet is going to share with you some more information about that. Janet?
Thanks, Kari. Wow, this is going to be a wealth of data for you then. For those of you who do not have these setup yet, that’s the first thing I want you to do when you leave this webinar today is go set these up because it’s a wealth of information for you.
We’ll start with Google Analytics Goals. If you have Google Analytics installed, you definitely want to have Google Analytics Goals. The reason is that it can show you not just leads. For instance, we use ours to look at lots of different types of goals we might have on our site, like e-newsletter signup which aren’t necessarily leads for us, but this does give an indicator of for instance, how successful are we at getting signups for our newsletter. We have different types of goals and in an organization you have different ones. You can have up to 20 goals per profile in Google Analytics. You can create multiple profiles within an account. As an example, here’s a view of ours that shows Goal one which is inbound leads, and it can show you from organic search. This particular chart is from organic search, how many leads are we getting over a time period from organic search, so it shows us as organic being the lead source, how many leads is that driving in. That can show us with some level of certainty the fact that we are actually driving leads in through SEO efforts. That’s a really important thing because we want to again justify our return on investment to the company.
This is very similar to AdWords conversions. They can be literally about anything, just like conversions could be a newsletter signup as well. This is very similar to the AdWords side of conversions if you know what that entails and it doesn’t actually involve putting any extra code on your site. On my conversion tracking in AdWords you can actually go into Google Analytics, set these up and it can even be something like event tracking we can use as a goal to see, did people play my video? Really cool stuff, did they push play on the video? Did they press stop? All sorts of neat stuff, so definitely start looking at Google Analytics. Then you can take a look at your goals and see how many people are completing them from organic search and it really can show some value there.
The other thing you need to think about is measuring leads, opportunities and sales. At Search Mojo we have really two levels of leads and this is something you can think about when we talk about goals versus leads. One of our goals is a lead, but it may not always be so we have things called marketing qualified leads and sales qualified leads, or MQLs and SQLs.
A marketing qualified lead is really just somebody who’s in our marketing database and we market to, but a sales qualified lead is somebody who is qualified to go to our sales department and actually is pursued as a sales lead. There’s two different definitions for that. You may have to define that in your own organization as well. You may or may not have that set up, but if you’re a B2B, you often will have that type of setup because you want to market and nurture to certain people, but certain people are mature and ready to be passed to sales. You can pull that kind of data from your marketing automation or you CRM tool like Salesforce.com.
Here’s an example. If you look at the bar graph there, you’ll see that you can’t really read all of the different lead sources, but we look at the lead sources that are driving in the leads for our marketing qualified leads. This is a graph from [inaudible 00:27:44]. I just built a report and, boom, I can pull that information out pretty easily for a particular timeframe. I’ve [inaudible 00:27:51] the funnel there because I want you to think about the funnel and think about measuring your organics touch on leads, opportunities and all the way down to sales. You can do that through things like Salesforce.com, through lead source and through some other tools, but it’s a good start to help you understand really, how is organic touching these folks that come in and become a lead, opportunity or a sale and really again, cost justify the SEO effort and show your return on investment.
Now, if you’re in ecommerce, your goal is typically are not leads, but it’s just typically direct sales right through the website. Another piece of Google Analytics you want to consider setting up is the ecommerce module. Do it, do it, do it, do it. If you’re an ecommerce vendor, you should be doing this because it is such a wealth of information. I cannot tell you. I love, love, love the ecommerce module and it’s so helpful for our ecommerce clients.
You can see here as an example. This is a one month’s log of one of client’s and you can see from organic search again, the revenue, the actual revenue that organic search brought in. That’s huge. Because in many cases in B2B we may not be doing ecommerce, we may not be taking transactions online, but in ecommerce and in online retail, you definitely have to be doing this because you have a benefit that a lot of folks don’t have in trying to measure offline conversion that you can measure online. You can actually measure the sales and direct ROI online, so this is a hugely important tool. I cannot stress it enough. If you’re ecommerce, definitely go and set this up. Then you can actually say, “Here’s the contribution the SEO made this month to the bottom line,” and there’s no better conversation to have with your boss than to be able to say here’s exactly what we contributed. That’s a huge, huge benefit.
One of the biggest challenges though is attribution. Attribution is a big conversation going on in marketing all over the place, not just in SEO. Many CRM tools, including Google Analytics, really only currently allow single attribution, meaning first touch or last touch. What’s the first thing that brought someone into my website or what’s the last thing? For instance, if you’re pulling data from Salesforce.com, there is one lead source field, that’s it. You can either have the first thing they came in on to populate that field, or you can constantly update it and then the last thing that touched them before they became a sale will be entered into that field. That’s not really ideal because when you think about it, it’s not really showing you all the things on the path necessarily. You can also consider setting up a custom field for tracking. We’ve done that at Search Mojo inside of our Salesforce.com to really understand better all the touches there, but there’re also some tools at Google Analytics you can use and some new things that are out there that can really help you understand this better.
The first thing I want to show you real quick, and then we’re going to talk into another poll, is Assisted Conversions. No matter what type of profile you have, likely if you have goals set up in Google Analytics, you should be able to see a assisted conversions and you should be able to see either from AdWords or from all types of conversions that you’re getting, what were the other types of things that assisted along with that; did email, did someone come directly to your site, etc.? You can see this right there in Google Analytics today, it’s under your conversions area under Google Analytics on the menu and you can see directly there again, what’s affecting what. You can see email’s a really good assisted conversion method for us in emailing and nurturing people, it’s doing very well. What you can do is even compare them through a Venn diagram. It’s very nice on the overview page.
There’s lots of things you can mix and match to see how, for instance, email and direct work together or email and referral work together and so forth. You can actually, those categories you see there, channel groupings as they call them, you can adjust those. You can create custom ones. There’s a lot of power to that, so I encourage you to take a look at that as well.
Now I’m going to hand it back to Kari for a quick poll about Google Analytics Universal Analytics which is something different than what we’ve been talking about which is the traditional analytics. I want to see how many of you are using that so I’m going to hand it back to Kari before we get into Universal Analytics.
Yes, and this will be our last poll for today. Basically, have you set up Universal Analytics? It is I suppose, relatively new, so some of you may already be on the ball and have it set up, but you may also not know the difference. Feel free to choose that if you don’t because Janet will certainly be explaining that as well.
I’m going to go ahead and close that poll out. As we expected because it is pretty new, a lot of you don’t have it set up yet, so that’s what Janet is going to be talking about next to give you a little bit more information about Universal Analytics. Janet?
Thanks Kari. What is Universal Analytics? As Kari mentioned, it’s a new form of analytics. It’s a little bit more robust than traditional analytics has been and it has a piece in it that I think is really valuable for this conversation. That is Attribution Modeling and it allows you to set up various types of attribution for your lead sources and so forth, even custom ones based on how you want to track how things contribute to a conversion, or a sale or a lead generation. It does however require a new property or account to be set up.
This is what I would suggest. You have to add tracking code again to your site, so if you’re using Google Analytics and you have a profile code up there now, you’ll have to add a second one. I would add a second one so you don’t lose the historical stuff when you create the new one, but I would create a new one with Universal Analytics in it because it allows you to have this extra reporting and attribution modeling. You cannot apparently change your existing account to Universal. It kind of stinks. I don’t like that. I don’t know why that is, but you have to create a new one to do it.
This is the link here that you can go to learn more about Universal Analytics and how you can implement it on your site.
Now let’s talk a little bit about the different attribution models that are available. You can also do a custom one, but let’s talk about the ones that you can choose through attribution modeling and we’ll talk about how they work.
The first one is “last interaction”. This is not uncommon from the way many people are already tracking how they attribute different lead sources and tactics to a particular sale or goal conversion and that is, “What is the last thing the person touched?” Basically what that means is the last thing they touched is going to get 100 percent of the credit for sending that person as a lead or a sale-100 percent-last interaction.
By contrast, some people prefer “first interaction”. What is the first thing that brought someone to my website, and if they go off and do other things and come back and convert later, what is the first thing they did? This is giving 100 percent of the credit to the first thing they did, what brought them in.
When you look at last and first, the way I think about them is, the last one is, “What is the thing that put them over the edge, what kind of put them over the hump to become a lead or a sale?” The first one is, “What’s the first thing that attracted them?” That’s kind of how I look at that. Again, these are giving 100 percent of the credit to either the last or the first. These two are probably the most commonly used attribution models right now because of the lack of tools.
I think that a lot of people have a lack of tools in their organization, but with this new Google Analytics, you’ll be able to do a little bit more than first and last. But like I said, with Salesforce.com, you’ve kind of got to be first or last. It kind of stinks that way. The nice thing is this new analytics attribution modeling should help you get more of a clear picture of what’s going on between first and last. As we all know, there could be a million things that go between first and last touch so we want to know what those things are and how they are helping in transaction of a lead generation or a sale happen.
The next one we want to talk about is linear. Linear basically gives equal amount of credit once the goal is reached. If the goal is a lead or it’s a sale, it gives equal amount of credit to everything along that path-equal amount. You can have as many things in the path as you need, but it gives equal amount.
Time decay is where it says I’m going to give the most credit to the most recent thing they did, but still give a little bit of credit over time, but dwindling over time back to when they started. What was the first thing they did and that gets the least amount of credit, but it still gets some level of credit, unlike last interaction which would only give all the credit to the last one and the first thing would get nothing. So this gives a little bit to the first, but it decreases over time backwards until you get to the first one. The amount of credit is decreased backwards from the time that they initially started with you.
Position-Based is one that I like also. It basically gives you 40 percent credit to the first, 40 percent to the last and then 20 percent for the things in-between, so those things will share credit in-between. This is another option you can look at if you say, well first is important to me because I really want to know how they came in, and last is really super important to me, but I still want to know all those things in-between and I’m just going to give them equal footing, you can use position-based.
You can come up with your own custom model as well and define that how you want to split up that 100 percent of credit and split it out between the different areas. This is really going to be powerful. But again, in order to do attribution modeling, you need the new Universal Analytics which will require you to set up a new property to your account to do that and add the code again.
Here is an example what attribution modeling looks like in action. This is a graphic that I took off of…how do you say that, Occam’s Razor? I don’t know how to say that, never have known how to say that. It’s basically Avinash’s blog. He is the Google Analytics guru and he has so many great pieces on that blog. If you’ve never checked it out, please do. It’s very good. Attribution modeling, this shows a comparison of last interaction versus time decay, versus position-based and you can see the conversions and how it’s actually changing the quantity of conversions based on the type of modeling, the attribution modeling that we’re using there. You can see last, time decay and position-based each has a different value for organic search based on how we have attributed it.
Notice here, last interactions, that first column, that last interaction for organic search, actually has the least amount of conversions associated with it. This is why that last interaction may not be the best thing if you’re doing SEO. We want to give SEO some credit, so time decay, and even position- based actually allow you to see a little bit more giving you more credit to organic search for having driven that person in at some point and then eventually converting them. Organic search may not always be the last interaction, but it may have been a touch point that aided in that conversion at some point. That’s why it’s so important to take a look at this attribution modeling piece.
Key takeaways for today-
Try to start moving away from rankings. It’s really not your best measurement of your success. I don’t want you to get bogged down by this and always be looking at something that may not be the most exact thing or may not be giving you all the credit you deserve. It also is really not telling you the full story about how you want to reach your goals. Then I want you to make sure you determine the attribution model you want to use.
Before you go in and install Google Analytics Universal Analytics, before you start thinking down that path, decide how you want to model your measurements and then think about that attribution model, what looks good to you, what works for your organization. Then you want to definitely install the Google Analytics Universal Analytics module as a new property and begin setting up attribution modeling. As you remember in Google Analytics a particular property only starts once that property is started in Google Analytics. So, if I start one today and I set this up today, I can’t get information from yesterday. It’s search from this point forward, so know that. And you want to install this as soon as you can so you can so you can start gathering data in that particular property and being able to look at it.
Lastly, you want to rework those measurements. Definitely start reworking them and how you attribute your SEO success to the people that you’re accountable to in your organization or your clients so that you can make sure that you’re really measuring the right thing and really truly measuring the right success.
With that, I’m going to hand it back to Kari who’s going to talk about our upcoming webinar.
Yes, indeed. Janet did mention that we have a mobile webinar coming up. She talked a lot about how important mobile search is and that’s exactly what is going to be discussed during our next webinar which will be in two weeks’ time. It’s called Mobile SEO: Designing and Optimizing for the Third Screen. That will be on June 13th at 2 p.m. Eastern and you can register today at search-mojo.com/mobile.
Also, if you’d like to get in touch with Janet, there is her contact information there, firstname.lastname@example.org and her phone number. Then also, she is on social media so there’s her email again, her phone number and all of her ways to connect with her through social media. With that, we will be taking some questions so make sure to enter your questions in the box at the right. We do have a few to start with.
First of all, Janet, you were just talking about attribution models, which one is the best to use?
There is no best one to use because it depends on your organization. For instance, in our organization, the sales cycle is longer and they may take more touches, and our personas, our buyer personas may need more types of content to make an informed buying decision versus something that’s a quick decision. If I’m buying a pair of sneakers online, running shoes or something, I might not need as much information per se, or research, and my time to purchase it probably pretty quick versus maybe a longer term decision that’s a larger decision for my company.
We have a client as an example, who has a two year sales cycle, so it really depends a lot on your sales and how much it takes to get that person over the hump to become a sale or become a lead. However, that link that I showed you. I’ll go back to that really quickly here. This one here, if you go to that particular link, it will tell you all about attribution modeling and there’s some really good information there about why you might consider one version versus another and gives some examples. That link right there would be a good one for you to visit and start to think about which one works best for you in your organization.
And just so you know, if you go on Twitter and find the hashtag, #mojowebinar, that link has been shared through the Search Mojo Twitter account on that hashtag, so you can grab it from there as well.
Our next question is having to do with going back to the whole rankings debate. This person, their boss still wants to see rankings, but is there a way for them to get accurate rankings even with all of these changes that you talked about?
The short answer is not really. You do your best to get the best rankings and information you can. There are a couple of different tools you can use to really understand how some of your keywords are ranking. Google Webmaster Tools is a great resource. If you’re not using that, you should be. From an SEO perspective, it gives you a lot of great data about positions of your keywords and your sites, and so forth. You can still get some accuracy from that.
However, third-party tools are having a much more difficult time. Again, the real challenge here, especially with Universal Search and personalized search, it’s still not going to tell you all the time. It’s going to give you sort of a rounded figure. It’s not always going to give you the most exact information, so you’re not always going to know that someone came through a shared status that showed up on Google through Google+ or that I have a ranking that’s higher for a particular item because of the fact that someone shared it who’s in my circles, or I shared it and someone else saw it, so that’s really very challenging. That’s again why I say it’s really not the best measurement.
I will tell you though, I would be remised in not saying it’s really hard to get away from rankings because clients I know still expect them. That’s the way they think about measurement. We’re trying to move away from that as an organization ourselves, and we’re showing a lot more in our reports than rankings. There is a lot more data. I think that clients and people I have been talking to in marketing have started to really embrace the idea somewhat of moving away from rankings, but it’s not 100 percent there yet so it is very challenging.
I think the best argument you can make is again what we talked about. All the rankings are genericized if you try and put them into rankings report versus really trying to understand the traffic you’re generating through different methods of SEO. I would say, try and make that argument to whoever you’re giving reports to and show them this data. Show them this great data you can get from analytics that really tells the story of what their goal really is. I have a lot of people still who come to me and say I want a number one ranking on X. You’re like, “No you don’t. What you really want is sales and leads. You don’t care how you get it.” If that’s the number one position or number two, it doesn’t really matter. What you’re caring about is increasing sales and leads. That’s what organic should be focused on, not necessarily the rankings themselves and what that actual number might be. So try and make that argument, show some data and see if that argument also works for you. Google Webmaster Tools is probably one of your better sources to see some of that data, but again, it’s still going to be more genericized in average. It’s not going to necessarily be exact rankings ever because really, no two searches are the same, even on the same keyword by two different people. It’s never going to be the same.
Great, and just as a reminder, we do have a few minutes left so if you do have any questions, make sure to pop those into the questions box. This one has to do with CRM tools. You were talking earlier about tracking leads in Salesforce and marketing automation tools, etc., but specifically, is there a way to pass data having to do with specific organic keywords through to opportunities in Salesforce and track back?
There are, but again, you’ve got to get away from the keyword thing. Just get away from that, because it’s going to be very difficult for you to track keywords. There’s actually a countdown calculator somewhere. I think if you probably searched it on Google you could probably find it. Countdown to “not provided” I think is what it’s called and it’ll tell you about how long it’ll be before every keyword shows up as not provided, 100 percent.
The reason it’s so hard to track keywords in general is because Google not only is saying we’re going to encrypt things from organic search when people are logged in to Gmail and their Google account,… lot of people like Facebook, just leave their Gmail open and they don’t think about logging out before they move on to the next thing and do a search… but also everything, I believe it’s from Chrome and everything from Firefox is now encrypted. So if you’re using one of those browsers even, every search from Google is encrypted. That’s a problem because more people use that. Unless you have 100 percent of people using Internet Explorer and not being logged into Gmail or to Google when they do a search, you’re never going to get 100 percent representation of those words and you’re not going to get accurate representation of those keywords.
First and foremost, I’d say you should drop the keyword thing. You can see again, through Google Webmaster Tools, some of that keyword information. If you really feel the need to pass that keyword along, what you can do is search for the keyword in the string, if you can find it in the string and if it’s not encrypted, and you can then pass that into a custom field you put into Salesforce through a hidden field in the form when a conversion comes through. But again, I think you’re going to get a lot of encrypted stuff. I think you’re not going to get as many keywords as you think.
I can tell you that on our own site, at least 50 percent of the traffic right now is encrypted from organic search. There’s no way to see those keywords from over 50 percent of our traffic. So if you’re only taking a small percentage of that and looking at that information, it’s not going to be an accurate measurement. I wouldn’t look at that necessarily as a measurement if you can avoid it.
Okay, great. Thanks. We have one last question. Obviously, we are talking a lot about Google Analytics during this webinar, but are there any other analytics platforms out there apart from Google Analytics?
There are. There are lots of different analytics platforms. I happen to like Google Analytics. I think it’s very flexible and I think the reason I like it so much too is it’s free, that’s not a bad thing, but also the fact that it integrates so easily with so many packages. That’s why I really like it. For instance, I have a client who uses Magento as a content management system and very simply you just put in the ID for your Google Analytics profile and, bam, it’s on every page and it does it for you automatically. It does have a lot of great stuff associated with it; great usability and so forth as well, but there are other packages out there.
However, I caution you that no two packages really are going to give you the same data, which is interesting about how they work. I highly recommend if you’re not familiar with how analytics works, the different types of packages, to do some reading up on it if you were thinking about changing packages and deciding which package is best for you. But no matter what you use, the keywords as an example, will not be able to be pulled through a different analytics package other than Google Analytics. Google, from the search engine, is passing this information in encrypted fashion, so it doesn’t mean for instance, if you used Omniture’s, or I guess it’s Adobe now, SiteCatalyst, if you use that, they still won’t be able to pull the words either. Don’t think that it’s a problem unique to Google Analytics.
There are lots of very good packages, you have to just decide what you want to measure and which one works best with the types of tools you like. I have some clients who use two tools. They use Google Analytics in combination with something like an Adobe SiteCatalyst to just compare information. You don’t have to be restricted to one analytics tool. Just know that every time that you stick the code on the site for another thing to load, it may reduce your load time on your site so you [inaudible 00:52:36] conservative with your load time.
Great. Thank you all so much for coming to today’s webinar, and thank you to our presenter, Janet Driscoll Miller. We will see you in two weeks at our next webinar, Mobile SEO: Designing and Optimizing for the Third Screen. We hope to see you soon and thank you again for coming.