Thứ Sáu, 18 tháng 1, 2019

Full Funnel Testing: SEO & CRO Together - Whiteboard Friday

Posted by willcritchlow

Testing for only SEO or only CRO isn't always ideal. Some changes result in higher conversions and reduced site traffic, for instance, while others may rank more highly but convert less well. In today's Whiteboard Friday, we welcome Will Critchlow as he demonstrates a method of testing for both your top-of-funnel SEO changes and your conversion-focused CRO changes at once.

Click on the whiteboard image above to open a high-resolution version in a new tab!

Video Transcription

Hi, everyone. Welcome to another Whiteboard Friday. My name is Will Critchlow, one of the founders at Distilled. If you've been following what I've been writing and talking about around the web recently, today's topic may not surprise you that much. I'm going to be talking about another kind of SEO testing.

Over at Distilled, we've been investing pretty heavily in building out our capability to do SEO tests and in particular built our optimization delivery network, which has let us do a new kind of SEO testing that hasn't been previously available to most of our clients. Recently we've been working on a new enhancement to this, which is full funnel testing, and that's what I want to talk about today.

So funnel testing is testing all the way through the funnel, from acquisition at the SEO end to conversion. So it's SEO testing plus CRO testing together. I'm going to write a little bit more about some of the motivation for this. But, in a nutshell, it essentially boils down to the fact that it is perfectly possible, in fact we've seen in the wild cases of tests that win in SEO terms and lose in CRO terms or vice versa.

In other words, tests that maybe you make a change and it converts better, but you lose organic search traffic. Or the other way around, it ranks better, but it converts less well. If you're only testing one, which is common — I mean most organizations are only testing the conversion rate side of things — it's perfectly possible to have a winning test, roll it out, and do worse.

CRO testing

So let's step back a little bit. A little bit of a primer. Conversion rate optimization testing works in an A/B split kind of way. You can test on a single page, if you want to, or a site section. The way it works is you split your audience. So your audience is split. Some of your audience gets one version of the page, and the rest of the audience gets a different version.

Then you can compare the conversion rate among the group who got the control and the group who got the variant. That's very straightforward. Like I say, it can happen on a single page or across an entire site. SEO testing, a little bit newer. The way this works is you can't split the audience, because we care very much about the search engine spiders in this case. For the purposes of this consideration, there's essentially only one Googlebot. So you couldn't put Google in Class A or Class B here and expect to get anything meaningful.

SEO testing

So the way that we do an SEO test is we actually split the pages. To do this, you need a substantial site section. So imagine, for example, an e-commerce website with thousands of products. You might have a hypothesis of something that will help those product pages perform better. You take your hypothesis and you only apply it to some of the pages, and you leave some of the pages unchanged as a control.

Then, crucially, search engines and users see the same experience. There's no cloaking going on. There's no duplication of content. You simply change some pages and not change others. Then you apply kind of advanced mathematical, statistical analysis trying to figure out do these pages get statistically more organic search traffic than we think they would have done if we hadn't made this change. So that's how an SEO test works.

Now, as I said, the problem that we are trying to tackle here is it's really plausible, despite Google's best intentions to do what's right for users, it's perfectly plausible that you can have a test that ranks better but converts less well or vice versa. We've seen this with, for example, removing content from a page. Sometimes having a cleaner, simpler page can convert better. But maybe that was where the keywords were and maybe that was helping the page rank. So we're trying to avoid those kinds of situations.

Full funnel testing

That's where full funnel testing comes in. So I want to just run through how you run a full funnel test. What you do is you first of all set it up in the same way as an SEO test, because we're essentially starting with SEO at the top of the funnel. So it's set up exactly the same way.

Some pages are unchanged. Some pages get the hypothesis applied to them. As far as Google is concerned, that's the end of the story, because on any individual request to these pages that's what we serve back. But the critically important thing here is I've got my little character. This is a human browser performs a search, "What do badgers eat?"

This was one of our silly examples that we came up with on one of our demo sites. The user lands on this page here. What we do is we then set a cookie. This is a cookie. This user then, as they navigate around the site, no matter where they go within this site section, they get the same treatment, either the control or the variant. They get the same treatment across the entire site section. This is more like the conversion rate test here.

Googlebot = stateless requests

So what I didn't show in this diagram is if you were running this test across a site section, you would cookie this user and make sure that they always saw the same treatment no matter where they navigated around the site. So because Googlebot is making stateless requests, in other words just independent, one-off requests for each of these of these pages with no cookie set, Google sees the split.

Evaluate SEO test on entrances

Users get whatever their first page impression looks like. They then get that treatment applied across the entire site section. So what we can do then is we can evaluate independently the performance in search, evaluate that on entrances. So do we get significantly more entrances to the variant pages than we would have expected if we hadn't applied a hypothesis to them?

That tells us the uplift from an SEO perspective. So maybe we say, "Okay, this is plus 11% in organic traffic." Well, great. So in a vacuum, all else being equal, we'd love to roll out this test.

Evaluate conversion rate on users

But before we do that, what we can do now is we can evaluate the conversion rate, and we do that based on user metrics. So these users are cookied.

We can also set an analytics tag on them and say, "Okay, wherever they navigate around, how many of them end up converting?" Then we can evaluate the conversion rate based on whether they saw treatment A or treatment B. Because we're looking at conversion rate, the audience size doesn't exactly have to be the same. So the statistical analysis can take care of that fact, and we can evaluate the conversion rate on a user-centric basis.

So then we maybe see that it's -5% in conversion rate. We then need to evaluate, "Is this something we should roll out?" So step 1 is: Do we just roll it out? If it's a win in both, then the answer is yes probably. If they're in different directions, then there are couple things we can do. Firstly, we can evaluate the relative performance in different directions, taking care that conversion rate applies generally across all channels, and so a relatively small drop in conversion rate can be a really big deal compared to even an uplift in organic traffic, because the conversion rate is applying to all channels, not just your organic traffic channel.

But suppose that it's a small net positive or a small net negative. What we can then do is we might get to the point that it's a net positive and roll it out. Either way, we might then say, "What can we take from this?What can we actually learn?" So back to our example of the content. We might say, "You know what? Users like this cleaner version of the page with apparently less content on it.The search engines are clearly relying on that content to understand what this page is about. How do we get the best of both worlds?"

Well, that might be a question of a redesign, moving the layout of the page around a little bit, keeping the content on there, but maybe not putting it front and center to the user as they land right at the beginning. We can test those different things, run sequential tests, try and take the best of the SEO tests and the best of the CRO tests and get it working together and crucially avoid those situations where you think you've got a win, because your conversion rate is up, but you actually are about to crater your organic search performance.

We think this is going to just be the more data-driven we get, the more accountable SEO testing makes us, the more important it's going to be to join these dots and make sure that we're getting true uplifts on a net basis when we combine them. So I hope that's been useful to some of you. Thank you for joining me on this week's Whiteboard Friday. I'm Will Critchlow from Distilled.

Take care.

Video transcription by Speechpad.com


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Full Funnel Testing: SEO & CRO Together - Whiteboard Friday

Posted by willcritchlow

Testing for only SEO or only CRO isn't always ideal. Some changes result in higher conversions and reduced site traffic, for instance, while others may rank more highly but convert less well. In today's Whiteboard Friday, we welcome Will Critchlow as he demonstrates a method of testing for both your top-of-funnel SEO changes and your conversion-focused CRO changes at once.

Click on the whiteboard image above to open a high-resolution version in a new tab!

Video Transcription

Hi, everyone. Welcome to another Whiteboard Friday. My name is Will Critchlow, one of the founders at Distilled. If you've been following what I've been writing and talking about around the web recently, today's topic may not surprise you that much. I'm going to be talking about another kind of SEO testing.

Over at Distilled, we've been investing pretty heavily in building out our capability to do SEO tests and in particular built our optimization delivery network, which has let us do a new kind of SEO testing that hasn't been previously available to most of our clients. Recently we've been working on a new enhancement to this, which is full funnel testing, and that's what I want to talk about today.

So funnel testing is testing all the way through the funnel, from acquisition at the SEO end to conversion. So it's SEO testing plus CRO testing together. I'm going to write a little bit more about some of the motivation for this. But, in a nutshell, it essentially boils down to the fact that it is perfectly possible, in fact we've seen in the wild cases of tests that win in SEO terms and lose in CRO terms or vice versa.

In other words, tests that maybe you make a change and it converts better, but you lose organic search traffic. Or the other way around, it ranks better, but it converts less well. If you're only testing one, which is common — I mean most organizations are only testing the conversion rate side of things — it's perfectly possible to have a winning test, roll it out, and do worse.

CRO testing

So let's step back a little bit. A little bit of a primer. Conversion rate optimization testing works in an A/B split kind of way. You can test on a single page, if you want to, or a site section. The way it works is you split your audience. So your audience is split. Some of your audience gets one version of the page, and the rest of the audience gets a different version.

Then you can compare the conversion rate among the group who got the control and the group who got the variant. That's very straightforward. Like I say, it can happen on a single page or across an entire site. SEO testing, a little bit newer. The way this works is you can't split the audience, because we care very much about the search engine spiders in this case. For the purposes of this consideration, there's essentially only one Googlebot. So you couldn't put Google in Class A or Class B here and expect to get anything meaningful.

SEO testing

So the way that we do an SEO test is we actually split the pages. To do this, you need a substantial site section. So imagine, for example, an e-commerce website with thousands of products. You might have a hypothesis of something that will help those product pages perform better. You take your hypothesis and you only apply it to some of the pages, and you leave some of the pages unchanged as a control.

Then, crucially, search engines and users see the same experience. There's no cloaking going on. There's no duplication of content. You simply change some pages and not change others. Then you apply kind of advanced mathematical, statistical analysis trying to figure out do these pages get statistically more organic search traffic than we think they would have done if we hadn't made this change. So that's how an SEO test works.

Now, as I said, the problem that we are trying to tackle here is it's really plausible, despite Google's best intentions to do what's right for users, it's perfectly plausible that you can have a test that ranks better but converts less well or vice versa. We've seen this with, for example, removing content from a page. Sometimes having a cleaner, simpler page can convert better. But maybe that was where the keywords were and maybe that was helping the page rank. So we're trying to avoid those kinds of situations.

Full funnel testing

That's where full funnel testing comes in. So I want to just run through how you run a full funnel test. What you do is you first of all set it up in the same way as an SEO test, because we're essentially starting with SEO at the top of the funnel. So it's set up exactly the same way.

Some pages are unchanged. Some pages get the hypothesis applied to them. As far as Google is concerned, that's the end of the story, because on any individual request to these pages that's what we serve back. But the critically important thing here is I've got my little character. This is a human browser performs a search, "What do badgers eat?"

This was one of our silly examples that we came up with on one of our demo sites. The user lands on this page here. What we do is we then set a cookie. This is a cookie. This user then, as they navigate around the site, no matter where they go within this site section, they get the same treatment, either the control or the variant. They get the same treatment across the entire site section. This is more like the conversion rate test here.

Googlebot = stateless requests

So what I didn't show in this diagram is if you were running this test across a site section, you would cookie this user and make sure that they always saw the same treatment no matter where they navigated around the site. So because Googlebot is making stateless requests, in other words just independent, one-off requests for each of these of these pages with no cookie set, Google sees the split.

Evaluate SEO test on entrances

Users get whatever their first page impression looks like. They then get that treatment applied across the entire site section. So what we can do then is we can evaluate independently the performance in search, evaluate that on entrances. So do we get significantly more entrances to the variant pages than we would have expected if we hadn't applied a hypothesis to them?

That tells us the uplift from an SEO perspective. So maybe we say, "Okay, this is plus 11% in organic traffic." Well, great. So in a vacuum, all else being equal, we'd love to roll out this test.

Evaluate conversion rate on users

But before we do that, what we can do now is we can evaluate the conversion rate, and we do that based on user metrics. So these users are cookied.

We can also set an analytics tag on them and say, "Okay, wherever they navigate around, how many of them end up converting?" Then we can evaluate the conversion rate based on whether they saw treatment A or treatment B. Because we're looking at conversion rate, the audience size doesn't exactly have to be the same. So the statistical analysis can take care of that fact, and we can evaluate the conversion rate on a user-centric basis.

So then we maybe see that it's -5% in conversion rate. We then need to evaluate, "Is this something we should roll out?" So step 1 is: Do we just roll it out? If it's a win in both, then the answer is yes probably. If they're in different directions, then there are couple things we can do. Firstly, we can evaluate the relative performance in different directions, taking care that conversion rate applies generally across all channels, and so a relatively small drop in conversion rate can be a really big deal compared to even an uplift in organic traffic, because the conversion rate is applying to all channels, not just your organic traffic channel.

But suppose that it's a small net positive or a small net negative. What we can then do is we might get to the point that it's a net positive and roll it out. Either way, we might then say, "What can we take from this?What can we actually learn?" So back to our example of the content. We might say, "You know what? Users like this cleaner version of the page with apparently less content on it.The search engines are clearly relying on that content to understand what this page is about. How do we get the best of both worlds?"

Well, that might be a question of a redesign, moving the layout of the page around a little bit, keeping the content on there, but maybe not putting it front and center to the user as they land right at the beginning. We can test those different things, run sequential tests, try and take the best of the SEO tests and the best of the CRO tests and get it working together and crucially avoid those situations where you think you've got a win, because your conversion rate is up, but you actually are about to crater your organic search performance.

We think this is going to just be the more data-driven we get, the more accountable SEO testing makes us, the more important it's going to be to join these dots and make sure that we're getting true uplifts on a net basis when we combine them. So I hope that's been useful to some of you. Thank you for joining me on this week's Whiteboard Friday. I'm Will Critchlow from Distilled.

Take care.

Video transcription by Speechpad.com


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Thứ Năm, 17 tháng 1, 2019

How to Implement a National Tracking Strategy

Posted by TheMozTeam

Google is all about serving up results based on your precise location, which means there’s no such thing as a “national” SERP anymore. So, if you wanted to get an accurate representation of how you’re performing nationally, you’d have to track every single street corner across the country.

Not only is this not feasible, it’s also a headache — and the kind of nightmare that keeps your accounting team up at night. Because we’re in the business of making things easy, we devised a happier (and cost-efficient) alternative.

Follow along and learn how to set up a statistically robust national tracking strategy in STAT, no matter your business or budget. And while we’re at it, we’ll also show you how to calculate your national ranking average.

Let’s pretend we’re a large athletic retailer. We have 30 stores across the US, a healthy online presence, and the powers-that-be have approved extra SEO spend — money for 20,000 additional keywords is burning a hole in our pocket. Ready to get started?

Step 1: Pick the cities that matter most to your business

Google cares a lot about location and so should you. Tracking a country-level SERP isn’t going to cut it anymore — you need to be hyper-local if you want to nab results.

The first step to getting more granular is deciding which cities you want to track in — and there are lots of ways to do this: The top performers? Ones that could use a boost? Best and worst of the cyber world as well as the physical world?

When it comes time for you to choose, nobody knows your business, your data, or your strategy better than you do — ain’t nothing to it but to do it.

A quick note for all our e-commerce peeps: we know it feels strange to pick a physical place when your business lives entirely online. For this, simply go with the locations that your goods and wares are distributed to most often.

Even though we’re a retail powerhouse, our SEO resources won’t allow us to manage all 30 physical locations — plus our online hotspots — across the US, so we'll cut that number in half. And because we’re not a real business and we aren’t privy to sales data, we'll pick at random.

From east to west, we now have a solid list of 15 US cities, primed, polished, and poised for our next step: surfacing the top performing keywords.

Step 2: Uncover your money-maker keywords

Because not all keywords are created equal, we need to determine which of the 4,465 keywords that we’re already tracking are going to be spread across the country and which are going to stay behind. In other words, we want the keywords that bring home the proverbial bacon.

Typically, we would use some combination of search volume, impressions, clicks, conversion rates, etc., from sources like STAT, Google Search Console, and Google Analytics to distinguish between the money-makers and the non-money-makers. But again, we’re a make-believe business and we don’t have access to this insight, so we’re going to stick with search volume.

A right-click anywhere in the site-level keywords table will let us export our current keyword set from STAT. We’ll then order everything from highest search volume to lowest search volume. If you have eyeballs on more of that sweet, sweet insight for your business, order your keywords from most to least money-maker.

Because we don’t want to get too crazy with our list, we’ll cap it at a nice and manageable 1,500 keywords.

Step 3: Determine the number of times each keyword should be tracked

We may have narrowed our cities down to 15, but our keywords need to be tracked plenty more times than that — and at a far more local level.

True facts: A “national” (or market-level) SERP isn’t a true SERP and neither is a city-wide SERP. The closer you can get to a searcher standing on a street corner, the better, and the more of those locations you can track, the more searchers’ SERPs you’ll sample.

We’re going to get real nitty-gritty and go as granular as ZIP code. Addresses and geo coordinates work just as well though, so if it’s a matter of one over the other, do what the Disney princesses do and follow your heart.

The ultimate goal here is to track our top performing keywords in more locations than our poor performing ones, so we need to know the number of ZIP codes each keyword will require. To figure this out, we gotta dust off the old desktop calculator and get our math on.

First, we’ll calculate the total amount of search volume that all of our keywords generate. Then, we’ll find the percentage of said total that each keyword is responsible for.

For example, our keyword [yeezy shoes] drew 165,000 searches out of a total 28.6 million, making up 0.62 percent of our traffic.

A quick reminder: Every time a query is tracked in a distinct location, it’s considered a unique keyword. This means that the above percentages also double as the amount of budgeted keywords (and therefore locations) that we’ll award to each of our queries. In (hopefully) less confusing terms, a keyword that drives 0.62 percent of our traffic gets to use 0.62 percent of our 20,000 budgeted keywords, which in turn equals the number of ZIP codes we can track in. Phew.

But! Because search volume is, to quote our resident data analyst, “an exponential distribution,” (which in everyone else-speak means “gets crazy large”) it’s likely going to produce some unreasonably big numbers. So, while [yeezy shoes] only requires 124 ZIP codes, a keyword with much higher search volume, like [real madrid], might need over 1,000, which is patently bonkers (and statistical overkill).

To temper this, we highly recommend that you take the log of the search volume — it’ll keep things relative and relational. If you’re working through all of this in Excel, simply type =log(A2) where A2 is the cell containing the search volume. Because we're extra fancy, we'll multiply that by four to linearly scale things, so =log(A2)*4.

So, still running with our Yeezy example, our keyword goes from driving 0.62 percent of our traffic to 0.13 percent. Which then becomes the percent of budgeted keywords: 0.0013 x 20,000 = tracking [yeezy shoes] in 26 zip codes across our 15 cities.

We then found a list of every ZIP code in each of our cities to dole them out to.

The end. Sort of. At this point, like us, you may be looking at keywords that need to be spread across 176 different ZIP codes and wondering how you're going to choose which ZIP codes — so let our magic spreadsheet take the wheel. Add all your locations to it and it'll pick at random.

Of course, because we want our keywords to get equal distribution, we attached a weighted metric to our ZIP codes. We took our most searched keyword, [adidas], found its Google Trends score in every city, and then divided it by the number of ZIP codes in those cities. For example, if [adidas] received a score of 71 in Yonkers and there are 10 ZIP codes in the city, Yonkers would get a weight of 7.1.

We'll then add everything we have so far — ZIP codes, ZIP code weights, keywords, keyword weights, plus a few extras — to our spreadsheet and watch it randomly assign the appropriate amount of keywords to the appropriate amount of locations.

And that’s it! If you’ve been following along, you’ve successfully divvied up 20,000 keywords in order to create a statistically robust national tracking strategy!

Curious how we’ll find our national ranking average? Read on, readers.

Step 4: Segment, segment, segment!

20,000 extra keywords makes for a whole lotta new data to keep track of, so being super smart with our segmentation is going to help us make sense of all our findings. We’ll do this by organizing our keywords into meaningful categories before we plug everything back into STAT.

Obviously, you are free to sort how you please, but we recommend at least tagging your keywords by their city and product category (so [yeezy shoes] might get tagged “Austin” and “shoes”). You can do all of this in our keyword upload template or while you're in our magic spreadsheet.

Once you’ve added a tag or two to each keyword, stuff those puppies into STAT. When everything’s snug as a bug, group all your city tags into one data view and all your product category tags into another.

Step 5: Calculate your national ranking average

Now that all of our keywords are loaded and tracking in STAT, it’s time to tackle those ranking averages. To do that, we’ll simply pop on over to the Dashboard tab from either of our two data views.

A quick glimpse of the Average Ranking module in the Daily Snapshot gives us, well, our average rank, and because these data views contain every keyword that we’re tracking across the country, we’re also looking at the national average for our keyword set. Easy-peasy.

To see how each tag is performing within those data views, a quick jump to the Tags tab breaks everything down and lets us compare the performance of a segment against the group as a whole.

So, if our national average rank is 29.7 but our Austin keywords have managed an average rank of 27.2, then we might look to them for inspiration as our other cities aren't doing quite as well — our keywords in Yonkers have an average rank of 35.2, much worse than the national average.

Similarly, if our clothes keywords are faring infinitely worse than our other product categories, we may want to revamp our content strategy to even things out.

Go get your national tracking on

Any business — yes, even an e-commerce business — can leverage a national tracking strategy. You just need to pick the right keywords and locations.

Once you have access to your sampled population, you’ll be able to hone in on opportunities, up your ROI, and bring more traffic across your welcome mat (physical or digital).

Got a question you’re dying to ask us about the STAT product? Reach out to clientsuccess@getSTAT.com. Want a detailed walkthrough of STAT? Say hello (don’t be shy) and request a demo.


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

How to Implement a National Tracking Strategy

Posted by TheMozTeam

Google is all about serving up results based on your precise location, which means there’s no such thing as a “national” SERP anymore. So, if you wanted to get an accurate representation of how you’re performing nationally, you’d have to track every single street corner across the country.

Not only is this not feasible, it’s also a headache — and the kind of nightmare that keeps your accounting team up at night. Because we’re in the business of making things easy, we devised a happier (and cost-efficient) alternative.

Follow along and learn how to set up a statistically robust national tracking strategy in STAT, no matter your business or budget. And while we’re at it, we’ll also show you how to calculate your national ranking average.

Let’s pretend we’re a large athletic retailer. We have 30 stores across the US, a healthy online presence, and the powers-that-be have approved extra SEO spend — money for 20,000 additional keywords is burning a hole in our pocket. Ready to get started?

Step 1: Pick the cities that matter most to your business

Google cares a lot about location and so should you. Tracking a country-level SERP isn’t going to cut it anymore — you need to be hyper-local if you want to nab results.

The first step to getting more granular is deciding which cities you want to track in — and there are lots of ways to do this: The top performers? Ones that could use a boost? Best and worst of the cyber world as well as the physical world?

When it comes time for you to choose, nobody knows your business, your data, or your strategy better than you do — ain’t nothing to it but to do it.

A quick note for all our e-commerce peeps: we know it feels strange to pick a physical place when your business lives entirely online. For this, simply go with the locations that your goods and wares are distributed to most often.

Even though we’re a retail powerhouse, our SEO resources won’t allow us to manage all 30 physical locations — plus our online hotspots — across the US, so we'll cut that number in half. And because we’re not a real business and we aren’t privy to sales data, we'll pick at random.

From east to west, we now have a solid list of 15 US cities, primed, polished, and poised for our next step: surfacing the top performing keywords.

Step 2: Uncover your money-maker keywords

Because not all keywords are created equal, we need to determine which of the 4,465 keywords that we’re already tracking are going to be spread across the country and which are going to stay behind. In other words, we want the keywords that bring home the proverbial bacon.

Typically, we would use some combination of search volume, impressions, clicks, conversion rates, etc., from sources like STAT, Google Search Console, and Google Analytics to distinguish between the money-makers and the non-money-makers. But again, we’re a make-believe business and we don’t have access to this insight, so we’re going to stick with search volume.

A right-click anywhere in the site-level keywords table will let us export our current keyword set from STAT. We’ll then order everything from highest search volume to lowest search volume. If you have eyeballs on more of that sweet, sweet insight for your business, order your keywords from most to least money-maker.

Because we don’t want to get too crazy with our list, we’ll cap it at a nice and manageable 1,500 keywords.

Step 3: Determine the number of times each keyword should be tracked

We may have narrowed our cities down to 15, but our keywords need to be tracked plenty more times than that — and at a far more local level.

True facts: A “national” (or market-level) SERP isn’t a true SERP and neither is a city-wide SERP. The closer you can get to a searcher standing on a street corner, the better, and the more of those locations you can track, the more searchers’ SERPs you’ll sample.

We’re going to get real nitty-gritty and go as granular as ZIP code. Addresses and geo coordinates work just as well though, so if it’s a matter of one over the other, do what the Disney princesses do and follow your heart.

The ultimate goal here is to track our top performing keywords in more locations than our poor performing ones, so we need to know the number of ZIP codes each keyword will require. To figure this out, we gotta dust off the old desktop calculator and get our math on.

First, we’ll calculate the total amount of search volume that all of our keywords generate. Then, we’ll find the percentage of said total that each keyword is responsible for.

For example, our keyword [yeezy shoes] drew 165,000 searches out of a total 28.6 million, making up 0.62 percent of our traffic.

A quick reminder: Every time a query is tracked in a distinct location, it’s considered a unique keyword. This means that the above percentages also double as the amount of budgeted keywords (and therefore locations) that we’ll award to each of our queries. In (hopefully) less confusing terms, a keyword that drives 0.62 percent of our traffic gets to use 0.62 percent of our 20,000 budgeted keywords, which in turn equals the number of ZIP codes we can track in. Phew.

But! Because search volume is, to quote our resident data analyst, “an exponential distribution,” (which in everyone else-speak means “gets crazy large”) it’s likely going to produce some unreasonably big numbers. So, while [yeezy shoes] only requires 124 ZIP codes, a keyword with much higher search volume, like [real madrid], might need over 1,000, which is patently bonkers (and statistical overkill).

To temper this, we highly recommend that you take the log of the search volume — it’ll keep things relative and relational. If you’re working through all of this in Excel, simply type =log(A2) where A2 is the cell containing the search volume. Because we're extra fancy, we'll multiply that by four to linearly scale things, so =log(A2)*4.

So, still running with our Yeezy example, our keyword goes from driving 0.62 percent of our traffic to 0.13 percent. Which then becomes the percent of budgeted keywords: 0.0013 x 20,000 = tracking [yeezy shoes] in 26 zip codes across our 15 cities.

We then found a list of every ZIP code in each of our cities to dole them out to.

The end. Sort of. At this point, like us, you may be looking at keywords that need to be spread across 176 different ZIP codes and wondering how you're going to choose which ZIP codes — so let our magic spreadsheet take the wheel. Add all your locations to it and it'll pick at random.

Of course, because we want our keywords to get equal distribution, we attached a weighted metric to our ZIP codes. We took our most searched keyword, [adidas], found its Google Trends score in every city, and then divided it by the number of ZIP codes in those cities. For example, if [adidas] received a score of 71 in Yonkers and there are 10 ZIP codes in the city, Yonkers would get a weight of 7.1.

We'll then add everything we have so far — ZIP codes, ZIP code weights, keywords, keyword weights, plus a few extras — to our spreadsheet and watch it randomly assign the appropriate amount of keywords to the appropriate amount of locations.

And that’s it! If you’ve been following along, you’ve successfully divvied up 20,000 keywords in order to create a statistically robust national tracking strategy!

Curious how we’ll find our national ranking average? Read on, readers.

Step 4: Segment, segment, segment!

20,000 extra keywords makes for a whole lotta new data to keep track of, so being super smart with our segmentation is going to help us make sense of all our findings. We’ll do this by organizing our keywords into meaningful categories before we plug everything back into STAT.

Obviously, you are free to sort how you please, but we recommend at least tagging your keywords by their city and product category (so [yeezy shoes] might get tagged “Austin” and “shoes”). You can do all of this in our keyword upload template or while you're in our magic spreadsheet.

Once you’ve added a tag or two to each keyword, stuff those puppies into STAT. When everything’s snug as a bug, group all your city tags into one data view and all your product category tags into another.

Step 5: Calculate your national ranking average

Now that all of our keywords are loaded and tracking in STAT, it’s time to tackle those ranking averages. To do that, we’ll simply pop on over to the Dashboard tab from either of our two data views.

A quick glimpse of the Average Ranking module in the Daily Snapshot gives us, well, our average rank, and because these data views contain every keyword that we’re tracking across the country, we’re also looking at the national average for our keyword set. Easy-peasy.

To see how each tag is performing within those data views, a quick jump to the Tags tab breaks everything down and lets us compare the performance of a segment against the group as a whole.

So, if our national average rank is 29.7 but our Austin keywords have managed an average rank of 27.2, then we might look to them for inspiration as our other cities aren't doing quite as well — our keywords in Yonkers have an average rank of 35.2, much worse than the national average.

Similarly, if our clothes keywords are faring infinitely worse than our other product categories, we may want to revamp our content strategy to even things out.

Go get your national tracking on

Any business — yes, even an e-commerce business — can leverage a national tracking strategy. You just need to pick the right keywords and locations.

Once you have access to your sampled population, you’ll be able to hone in on opportunities, up your ROI, and bring more traffic across your welcome mat (physical or digital).

Got a question you’re dying to ask us about the STAT product? Reach out to clientsuccess@getSTAT.com. Want a detailed walkthrough of STAT? Say hello (don’t be shy) and request a demo.


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!

Thứ Tư, 16 tháng 1, 2019

5 Real Examples of Advanced Content Promotion Strategies

Posted by bsmarketer

Content promotion isn’t tweeting or upvoting. Those tiny, one-off tactics are fine for beginners. They might make a dent, but they definitely won’t move the needle. Companies that want to grow big and grow fast need to grow differently.

Here’s how Kissmetrics, Sourcify, Sales Hacker, Kinsta, and BuildFire have used advanced content promotion tips like newsjacking and paid social to elevate their brands above the competition.

1. Use content to fuel social media distribution (and not the other way around)

Prior to selling the brand and blog to Neil Patel, Kissmetrics had no dedicated social media manager at the height of their success. The Kissmetrics blog received nearly 85% of its traffic from organic search. The second biggest traffic-driver was the newsletter.

Social media did drive traffic to their posts. However, former blog editor Zach Buylgo’s research showed that these traffic segments often had the lowest engagement (like time on site) and the least conversions (like trial or demo opt-ins) — so they didn’t prioritize it. The bulk of Zach’s day was instead focused on editing posts, making changes himself, adding comments and suggestions for the author to fix, and checking for regurgitated content. Stellar, long-form content was priority number one. And two. And three.

So Zach wasn’t just looking for technically-correct content. He was optimizing for uniqueness: the exact same area where most cheap content falls short. That’s an issue because many times, a simple SERP analysis would reveal that one submission:

benefits of content marketing (crowd content)

(image source)

...Looked exactly like the number-one result from Content Marketing Institute:

benefits of content marketing CMI

(image source)

Today’s plagiarism tools can catch the obvious stuff, but these derivatives often slip through the cracks. Recurring paid writers contributed the bulk of the TOFU content, which would free Zach up to focus more on MOFU use cases and case studies to help visitors understand how to get the most out of their product set (from the in-house person who knows it best).

They produced marketing guides and weekly webinars to transform initial attention into new leads:

image.png

They also created free marketing tools to give prospects an interactive way to continue engaging with their brand:

image.png

In other words, they focused on doing the things that matter most — the 20% that would generate the biggest bang for their buck. They won’t ignore social networks completely, though. They still had hundreds of thousands of followers across each network. Instead, their intern would take the frontlines. That person would watch out for anything critical, like a customer question, which will then be passed off to the Customer Success Manager that will get back to them within a few hours.

image.png

New blog posts would get the obligatory push to Twitter and LinkedIn. (Facebook is used primarily for their weekly webinar updates.) Zach used Pablo from Buffer to design and create featured images for the blog posts.

image.png

Then he’d use an Open Graph Protocol WordPress plugin to automatically add all appropriate tags for each network. That way, all he had to do was add the file and basic post meta data. The plugin would then customize how it shows up on each network afterward. Instead of using Buffer to promote new posts, though, Zach likes MeetEdgar.

Why? Doesn’t that seem like an extra step at first glance? Like Buffer, MeetEdgar allows you to select when you’d like to schedule content. You can just load up the queue with content, and the tool will manage the rest. The difference is that Buffer constantly requires new content — you need to keep topping it off, whereas MeetEdgar will automatically recycle the old stuff you’ve previously added. This saved a blog like Kissmetrics, with thousands of content pieces, TONS of time.

Header-screenshot-animated_02.gif

(image source)

He would then use Sleeknote to build forms tailored to each blog category to transform blog readers into top-of-the-funnel leads:

Screen Shot 2017-10-30 at 1.36.05 PM.png

But that’s about it. Zach didn’t do a ton of custom tweets. There weren’t a lot of personal replies. It’s not that they didn’t care. They just preferred to focus on what drives the most results for their particular business. They focused on building a brand that people recognize and trust. That means others would do the social sharing for them.

Respected industry vets like Avinash Kaushik, for example, would often share their blog posts. And Avinash was the perfect fit, because he already has a loyal, data-driven audience following him.

image.png

So that single tweet brings in a ton of highly-qualified traffic — traffic that turns into leads and customers, not just fans.

2. Combine original research and newsjacking to go viral

Sourcify has grown almost exclusively through content marketing. Founder Nathan Resnick speaks, attends, and hosts everything from webinars to live events and meetups. Most of their events are brand-building efforts to connect face-to-face with other entrepreneurs. But what’s put them on the map has been leveraging their own experience and platform to fuel viral stories.

Last summer, the record-breaking Mayweather vs. McGregor fight was gaining steam. McGregor was already infamous for his legendary trash-talking and shade-throwing abilities. He also liked to indulge in attention-grabbing sartorial splendor. But the suit he wore to the very first press conference somehow managed to combine the best of both personality quirks:

(image source)

This was no off-the-shelf suit. He had it custom made. Nathan recalls seeing this press conference suit fondly: “Literally, the team came in after the press conference, thinking, ‘Man, this is an epic suit.’” So they did what any other rational human being did after seeing it on TV: they tried to buy it online.

“Except, the dude was charging like $10,000 to cover it and taking six weeks to produce.” That gave Nathan an idea. “I think we can produce this way faster.”

They “used their own platform, had samples done in less than a week, and had a site up the same day.”

(image source)

“We took photos, sent them to different factories, and took guesstimates on letter sizing, colors, fonts, etc. You can often manufacture products based on images if it’s within certain product categories.” The goal all along was to use the suit as a case study. They partnered with a local marketing firm to help split the promotion, work, and costs.

“The next day we signed a contract with a few marketers based in San Francisco to split the profits 50–50 after we both covered our costs. They cover the ad spend and setup; we cover the inventory and logistics cost,” Nathan wrote in an article for The Hustle. When they were ready to go, the marketing company began running ad campaigns and pushing out stories. They went viral on BroBible quickly after launch and pulled in over $23,000 in sales within the first week.

The only problem is that they used some images of Conor in the process. And apparently, his attorney’s didn’t love the IP infringement. A cease and desist letter wasn’t far behind:

(image source)

This result wasn’t completely unexpected. Both Nathan and the marketing partner knew they were skirting a thin line. But either way, Nathan got what he wanted out of it.

3. Drive targeted, bottom-of-the-funnel leads with Quora

Quora packs another punch that often elevates it over the other social channels: higher-quality traffic. Site visitors are asking detailed questions, expecting to comb through in-depth answers to each query. In other words, they’re invested. They’re smart. And if they’re expressing interest in managed WordPress hosting, it means they’ve got dough, too.

Both Sales Hacker and Kinsta take full advantage. Today, Gaetano DiNardi is the Director of Demand Generation at Nextiva. But before that, he lead marketing at Sales Hacker before they were acquired. There, content was central to their stratospheric growth. With Quora, Gaetano would take his latest content pieces and use them to solve customer problems and address pain points in the general sales and marketing space:

By using Quora as a research tool, he would find new topics that he can create content around to drive new traffic and connect with their current audience:

He found questions that they already had content for and used it as a chance to engage users and provide value. He can drive tons of relevant traffic for free by linking back to the Sales Hacker blog:

Kinsta, a managed WordPress hosting company out of Europe, also uses uses relevant threads and Quora ads. CMO Brian Jackson jumps into conversations directly, lending his experience and expertise where appropriate. His technical background makes it easy to talk shop with others looking for a sophisticated conversation about performance (beyond the standard, PR-speak most marketers offer up):

Brian targets different WordPress-related categories, questions, or interests. Technically, the units are “display ads, but they look like text.” The ad copy is short and to the point. Usually something like, “Premium hosting plans starting at $XX/month” to fit within their length requirements.

4. Rank faster with paid (not organic) social promotion

Kinsta co-founder Tom Zsomborgi wrote about their journey in a bootstrapping blog post that went live last November. It instantly hit the top of Hacker News, resulting in their website getting a consistent 400+ concurrent visitors all day:

Within hours their post was also ranking on the first page for the term “bootstrapping,” which receives around 256,000 monthly searches.

How did that happen?

“There’s a direct correlation between social proof and increased search traffic. It’s more than people think,” said Brian. Essentially, you’re paying Facebook to increase organic rankings. You take good content, add paid syndication, and watch keyword rankings go up.

Kinsta’s big goal with content promotion is to build traffic and get as many eyeballs as possible. Then they’ll use AdRoll for display retargeting messages, targeting the people who just visited with lead gen offers to start a free trial. (“But I don’t use AdRoll for Facebook because it tags on their middleman fee.”)

Brian uses the “Click Campaigns” objective on Facebook Ads for both lead gen and content promotion. “It’s the best for getting traffic.”

Facebook's organic reach fell by 52% in 2016 alone. That means your ability to promote content to your own page fans is quickly approaching zero.

Screen Shot 2017 06 29 at 12.52.27 PM

(image source)

“It’s almost not even worth posting if you’re not paying,” confirms Brian. Kinsta will promote new posts to make sure it comes across their fans’ News Feed. Anecdotally, that reach number with a paid assist might jump up around 30%.

If they don’t see it, Brian will “turn it into an ad and run it separately.” It’s “re-written a second time to target a broader audience.”

In addition to new post promotion, Brian has an evergreen campaign that’s constantly delivering the “best posts ever written” on their site. It’s “never-ending” because it gives Brian a steady-stream of new site visitors — or new potential prospects to target with lead gen ads further down the funnel. That’s why Brian asserts that today’s social managers need to understand PPC and lead gen. “A lot of people hire social media managers and just do organic promotion. But Facebook organic just sucks anyway. It’s becoming “pay to play.’”

“Organic reach is just going to get worse and worse and worse. It’s never going to get better.” Also, advertising gets you “more data for targeting,” which then enables you to create more in-depth A/B tests.

We confirmed this through a series of promoted content tests, where different ad types (custom images vs. videos) would perform better based on the campaign objectives and placements.

(image source)

That’s why “best practices” are past practices — or BS practices. You don’t know what’s going to perform best until you actually do it for yourself. And advertising accelerates that feedback loop.

5. Constantly refresh your retargeting ad creative to keep engagement high

Almost every single stat shows that remarketing is one of the most efficient ways to close more customers. The more ad remarketing impressions someone sees, the higher the conversion rate. Remarketing ads are also incredibly cheap compared to your standard AdWords search ad when trying to reach new cold traffic.

(image source)

There’s only one problem to watch out for: ad fatigue. The image creative plays a massive role in Facebook ad success. But over time (a few days to a few weeks), the performance of that ad will decline. The image becomes stale. The audience has seen it too many times. The trick is to continually cycle through similar, but different, ad examples.

Here’s how David Zheng does it for BuildFire:

His team will either (a) create the ad creative image directly inside Canva, or (b) have their designers create a background ‘template’ that they can use to manipulate quickly. That way, they can make fast adjustments on the fly, A/B testing small elements like background color to keep ads fresh and conversions as high as possible.

(image source)

All retargeting or remarketing campaigns will be sent to a tightly controlled audience. For example, let’s say you have leads who’ve downloaded an eBook and ones who’ve participated in a consultation call. You can just lump those two types into the same campaign, right? I mean, they’re both technically ‘leads.’

But that’s a mistake. Sure, they’re both leads. However, they’re at different levels of interest. Your goal with the first group is to get them on a free consultation call, while your goal with the second is to get them to sign up for a free trial. That means two campaigns, which means two audiences.

Facebook’s custom audiences makes this easy, as does LinkedIn’s new-ish Matched Audiences feature. Like with Facebook, you can pick people who’ve visited certain pages on your site, belong to specific lists in your CRM, or whose email address is on a custom .CSV file:

If both of these leads fall off after a few weeks and fail to follow up, you can go back to the beginning to re-engage them. You can use content-based ads all over again to hit back at the primary pain points behind the product or service that you sell.

This seems like a lot of detailed work — largely because it is. But it’s worth it because of scale. You can set these campaigns up, once, and then simply monitor or tweak performance as you go. That means technology is largely running each individual campaign. You don’t need as many people internally to manage each hands-on.

And best of all, it forces you to create a logical system. You’re taking people through a step-by-step process, one tiny commitment at a time, until they seamlessly move from stranger into customer.

Conclusion

Sending out a few tweets won’t make an impact at the end of the day. There’s more competition (read: noise) than ever before, while organic reach has never been lower. The trick isn’t to follow some faux influencer who talks the loudest, but rather the practitioners who are doing it day-in, day-out, with the KPIs to prove it.


Sign up for The Moz Top 10, a semimonthly mailer updating you on the top ten hottest pieces of SEO news, tips, and rad links uncovered by the Moz team. Think of it as your exclusive digest of stuff you don't have time to hunt down but want to read!