How to write a blog

  A few months ago, I ran an experiment to seeif we could get backlinks using ads. And it worked decently, but there weresome obvious flaws in my process. So I talked with my buddy Josh who runs Ahrefs'blog, and we strategized a way to do this right. Meaning we'd create a new link-worthy page,get links to it, and have a cool case study to share with you. So today we're kicking off a three-part serieswhere I'll show you how to strategically write a blog post that generates links. Then in part 2, I'll show you how we built linksto that page, and we'll finish off the series with where it all began. An ad campaign specifically designedto generate links passively. I'm pumped and if you are too then stay tuned.

 [music] So this is part one of our three-part seriesand I'll be dropping the other videos over the next two weeks. So if you're not subscribed to our channel,then make sure to do that now because the videos will go hand-in-hand and I don'twant you to miss out on the rest of this series. And if you're watching this at a future time, thenfeel free to binge watch these and implement as you go. Alright, so I've created a lot of videos oncontent creation and link building. But what I haven't talked about much is howthey're connected. And there are multiple facets to this. The first and most obvious is content quality. It's easier to get links to an awesome pieceof content than to something mediocre. And I'm sure you've heard this advice a milliontimes but that's because it's true. The second factor that often gets overlookedis whether the topic is actually "linkable." And this comes down to searcher intent.

You need to ask yourself, why are these peoplesearching for the topic first place? For example, let's say you've created a posttargeting the query "best decaf coffee." Who do you think would be interested in searchingfor this on Google, or what type of people might click it if they saw it on social media? Probably people who want to drink decaf coffee. The vast majority of people probably aren'tbloggers and journalists or anyone that has the power to link to your page. And as a result, you'll see that none of the top 10pages have many backlinks pointing at them. Now, what about a topic like "coffee stats?" Who do you think would be searchingfor something like that? Probably bloggers, journalists, and peoplewho are looking for supporting facts to include in their articles. In fact, words like "stats," "statistics," and "facts"are very common modifiers people search for when writing content. And this applies to tons of industries. For example, this page on coffee stats has584 referring domains pointing at the page. And referring domains are unique websitesthat link to the page. Nerdwallet's statistics page on average householdcredit card debt has nearly 2,500 referring domains. And multiple pages for the query, "marriagestatistics" have hundreds of referring domains. Bottomline: people link to stats pages whenciting facts and figures in their content. 

And the third facet is timing. Even if your content is great, it won't translateinto links if it reaches the right people at the wrong time—at least not inthe immediate future. But nail the timing, and your chances of gettinga link increases exponentially. And we'll get into timing later on in the series. So the first thing we need to do is find aproven topic that's "link-worthy." So to find a topic for our site, I went toKeywords Explorer and searched for a bunch of broad keywords related to our business, like "seo,""search engine optimization," "content marketing," "keyword research," "link building," and "blogging." Next, I went to the Phrase match report, clickedon the Include filter, and added a list of keyword modifiers like "stat," "stats," "statistics," "fact," and "facts." Finally, I clicked on the "Any" tab so the listwould show us keyword ideas that include any of these modifiers combined withour original list of seeds. Now, a lot of these look like great topicsfor our site, but since our primary focus is on SEO, "seo statistics" is the mostrelevant query to our business. So that's the one we chose. Now, it's important to note that creatinga "stats" page isn't necessarily about generating a ton of search traffic. So if you see lower search volumes like this,then there's no need to worry. These pages are about generating a ton of links. 

Then you can use that page to "power" otherposts or even your money pages. For example, our study on featured snippetsis our most linked-to post on Ahrefs' blog. And if you visit that page, you'll see thatwe link to a couple of our tool landing pages, helping them rank higher in Googlefor queries we care about. Now, if you're having trouble understandinghow this SEO strategy works, we have a full video on it called "The Middleman Method,"so I'll link that up in the description. Alright, so if we go back to Keywords Explorerand look at the SERP for "seo statistics," you'll see that these three pages have over4,200 referring domains pointing at them. Meaning, there are more than enough link prospectswe can reach out to as soon as we create our SEO stats post. Alright, so the next thing we needed to dowas to create the stats page. Now, let's have a moment of honesty here. It's tough to make a page of stats truly standoutamongst other pages. They're all going to be curated lists andthere's definitely going to be overlap between competing pages. So to ensure we had the best page we couldmake, we thought about what a user would want to see when they land on the page. Basically, people are searching for curatedinformation so they can add bits and bobs to their posts to support their claims. 

So it was vital that we made our post organizedand easy to digest. So we asked SQ, one of our most efficientresearchers to compile a nicely organized list of SEO stats by category. And the entire first draft of the post wasdone in a couple of hours. Now at this point, we had a decent list ofstats and it was ready to be published. But these are all based on what SQ thoughtwere interesting. So I had an interesting thought. What if we included stats from these popularpages that are responsible for attracting links. Then would that not make our page a curationof the quote unquote "best" SEO statistics? Afterall, if people are linking based ona specific stat, then that's telling of what people want to find. So the next thing we did was analyze the backlinkprofiles of competing pages to see which stats attracted the most links. So as I was researching in Ahrefs' Site Explorer,I noticed that the majority of links that stats pages get can attributed to just a handfulof points, despite having tons of different stats on the page. Translation: around 10% of the stats are responsiblefor 90% of the links. So for the most part, they were just creatingnoise to the stats that people want to link to.

  So we took a few of the top ranking pagesthat have tons of links and put each of them into Site Explorer. Then we went to the Anchors report. As you can see, people are linking to thesepages based on specific stats. So looking at the Anchors reports for SearchEngine Journal's page, 93% is the most used anchor by far. Then there's 51%, 32.5%, and so on. Next, we went to the Backlinks report to getmore context on these links and we also wanted to see the number of unique websites linkingto the stats page, which would help us gauge popularity and ultimately tell us whetherwe should include a similar stat in our post. To do this, I set the view to "One link perdomain," since we don't need to contact the same website multiple times. And then searched for the number "93"in the Include box. And I set the filter parameters to only searchwithin the anchor or surrounding text of the page, since that would tell us if we can attributethat stat to the link. Finally, I'll set this filter to only showfollowed links, which are value-passing links. And as you can see, there are 652 referringdomains pointing at this page that can be attributed to this one stat alone, so it was definitelyworth including something similar in our post. Now, there's an interesting thing we noticed here. The original backlink for most of these linksare pointing at an old URL, which has been redirected to the new page. 

And when there are redirects, there are oftenproblems that come with them. After visiting SEJ's stats page we learnedthat there isn't a single mention of 93, which means that 652 links from unique websitesno longer make sense, so this would become a part of our outreach angle, and we'llget into this in part 2. Now, in order to find where that 93% statcame from, I clicked on the caret beside the URL and then clicked on "View on archive.org." And if I search for that stat, you'll see that it comes from a 2006 study done by Forrester which is 14 years ago. So after choosing around 5 to 10 stats basedon our competitors' anchors, we did an audit of these stats. Basically, with any data point that was super oldor outdated, we did our best to find a recent stat. Then we added our list of popular and up-to-datestats near the top of the post with the heading "Top SEO Statstics," so they'd get more exposureand hopefully generate more links. And there we have it! A stats page that we think will generate links. 

And the only way to find out is if we actuallytry and build some links to it. And that's what we're doing in the next video,which will drop in a week from this video's publishing date. So make sure to subscribe to our channel soyou don't miss out on that because I'll share everything about our outreach campaigns: allthe stats, and some interesting things that I learned about link building in 2020. And it wasn't exactly pretty.