We actually wrote a guide on this very topic I think could be very useful for any start ups looking to invest in SEO.
Technical SEO covers the technical health of your site, site architecture, indexation, site speed, mobile friendliness, structured data markup, robot files and sitemaps, canonical tag reviews, and internal link structure.
That’s a lot.
Additionally, because sites change and grow over time, technical SEO is also an ongoing process and an integral part of upkeep.
When we run a technical check on a client’s site, we check hundreds of individual factors.
Unless you’ve got a dedicated SEO team, that may not always be feasible. So, for the sake of space and practicality, let’s just quickly cover the very most essential and basic best practices — things you should implement when you launch a site and should keep up on as you build out a library of content:
- Indexation
- Site architecture
- Site speed
- Titles and H1 tags
- Content quality (duplicate and thin content)
Indexation. Indexation has two parts: (1) making sure content that should be indexed is indexed and (2) making sure content you don’t want indexed is not indexed. You can request indexation with Google Search Console, and you can block indexation with either a noindex tag or through the robots.txt file. Here’s Google’s documentation on the Index Status Report found in Search Console, which looks something like this:
Site architecture. Site architecture refers to the structure of your site, which is determined by both your internal linking scheme and the structure of your folders (usually determined by your categories if you’re using WordPress). A clear and logical structure that promotes relevance can help boost overall rankings.
A few basic rules of thumb include:
- Minimize page depth (number of clicks a page is from the homepage)
- Group relevant content together
- Link from content to other relevant content within the same category
If you really want to dig into site architecture, this is one of the most definitive articles ever written on it (although it’s a bit outdated). To visualize your own site architecture, we recommend Screaming Frog, which generates stuff like this:
Site speed. Google rolled out a site speed update this year. While it really only directly affected the slowest sites on the internet, it’s still a good practice to make your site as fast as possible. Why? Because fast sites typically yield good user engagement metrics, which do affect how you rank. There are lots of things you can do to speed up your site, but the basics are:
- Deliver your site through a content delivery network (CDN)
- Set up and enable caching
- Aggregate and minify CSS, HTML and Javascript
- Reduce image sizes
- Reduce plugin bloat
- Splurge on good hosting
To check and optimize your site speed, you use tools like GT Metrix or Pingdom. Google also has one of their own. Reports usually look like this:
Titles and H1s. Optimizing title tags and H1 tags are — and have always been — the most important on-page ranking factors. Luckily, they’re fairly straightforward. All you have to do is plop your main keyword in there somewhere, make sure the length is good (so it shows up correctly in Google), and you’re good to go; however, as sites grow, it becomes easier and easier to lose track of even basic tactical SEO like this. Screaming Frog gives you lots of good insight into this stuff as well:
Content quality. In this context “content quality” doesn’t mean how subjectively “good” a piece of content is; we’re mostly just trying to avoid too much duplicate or “thin” content. In general, every piece of indexed content on a site should be unique, substantial and meaningful. There are lots of ways to really dig into content quality, but one of the easiest ways to do it is to use a free tool called Siteliner. Plug in your site, and you’ll get a simple report like this one:
So those are the basics. And they’re really not rocket science. We recommend getting at least one full and comprehensive technical audit toward the beginning of your business’s life; it’ll save you a lot of time and headache in the long run.
Now, onto the other two more subjective, more nebulous, and more wily pillars of SEO…
The Engine: Content Strategy & Keyword Research for Startups
An SEO-driven content strategy has two basic parts:
- Traffic-generating content
- Linkable assets
Traffic-generating content is designed to rank well in Google and (surprise) attract traffic; linkable assets are marketing tools that help build links, which are still a core part of Google’s algorithm and are necessary for a site to rank on their platform.
You generally need both. The one exception would be if you already have a site with very high authority (i.e. lots of links from other sites already pointing to it), which sometimes happens with startups that generate lots of press. But most of the time, especially for newer startups, a steady stream of links is required to maximize other SEO efforts (more on link building below).
Traffic-generating content should comprise the bulk of the content on a site and typically targets lower-competition keywords (keywords that are easier to compete for in Google).
The trick to creating high-yield traffic-generating content is to find topics that have the best combination of traffic potential and low competition (when talking about keyword competitiveness, we in the industry typically refer to it as “keyword difficulty”).
So, in other words, you want a good traffic potential : difficulty ratio.
Let’s unpack each of those separately.
Traffic potential is a measure of the upper limit of the traffic we might expect if we rank well for most of the available keywords for a given topic.
We use this to supplement — and sometimes in place of — search volume (how many times a given keyword is searched per month) because often, traffic is much higher than the search volume of a specific keyword since successful content usually ranks for many hundreds of keywords.
To see traffic potential, we have to use a third party tool, and currently, the one and only tool that adequately does the job is Ahrefs.
Ahrefs is so good for this particular task because it allows us to see a decent (albeit not perfect) estimate of traffic for a given set of search results. Let me show you what I mean:
Here’s the Ahrefs data for the keyword “furniture design software.”
In the SERPs (search engine results page) we can see all the pages’ rankings along with Ahrefs’ estimate of each page’s monthly organic traffic.
These are the numbers we’re interested in when looking at traffic potential.
The most successful pages generate over 1,000 visits per month (quick note: Ahrefs tends to underestimate traffic, so the real traffic could likely be higher). Other ranking pages seem to attract several hundred visitors per month.
For one piece of content, this is fairly good, especially if you sell furniture design software. It’s also just one keyword in one topic. By and large, this is a higher traffic potential than the average for most content.
If we wrote a blog post on this article and it ranked in the top 10 results in Google, we might expect to attract somewhere between 500 and 2,000 visits per month.
Let’s plug this into a quick hypothetical scenario: if you’re publishing weekly, and about 3/4ths of your content (perhaps 40 posts per year) is traffic-generating content optimized to rank for topics similar this one, you might expect to increase your monthly traffic by 20,000-100,000 visits over the course of a year.
Compare that to a keyword like this one: “best cart abandonment software.”
Here, the SERPs look quite a bit different. There are a few pages with decent traffic, but quite a few pages here attract less than 100 visits per month.
It’s not nothing, but it’s also not ideal.
A second hypothetical scenario: If we added 40 pages optimized for topics that had traffic profiles like this one, over the course of a year, we might hit a few home runs, but it’s more likely we’d bat the average and add somewhere between 50-100 visits per month, or 2,000 – 4,000 monthly visits.
That’s a lot less than the numbers we were just talking about with our first keyword. In fact, it’s about 10x less.
We didn’t change much here. We published 40 traffic-generating articles in each of our hypothetical scenarios. The only real difference was that in the first, we tried to pick a topic with a health traffic potential.
In my view, traffic potential is possibly the most important part of a good SEO-driven content strategy. Understanding this metric alone — even if you’re just using the best estimates of a third party tool — can easily 10x your content ROI.
It’s not the only part though.
A good content strategy also requires you to understand keyword difficulty.
Keyword difficulty is the measure of how easy it would be to rank for a given keyword.
If a keyword is too difficult, you could end up spending time and money on content that will never rank, and if it never ranks, both your traffic and your ROI will be zero.
Keyword difficulty is measured in lots of different ways (e.g. content quality, domain authority of the sites in the SERPs), but the best way to measure it is by measuring the number and power of the backlinks pointing to the pages that already rank for the keyword.
Almost all keyword research tools on the market include some way to measure keyword difficulty. Because I’m a bit of a nerdball who wants the best possible SEO tools, I did an extensive analysis on which tool had the most accurate keyword difficulty scores (measured against my own expertise, so take it with a grain of salt).
In my opinion, Ahrefs (yep, the same tool; notice a trend?) has the most consistently accurate keyword difficulty scores, so it’s what we usually use when we’re trying to find that Goldilocks traffic potential : keyword difficulty ratio.
Let’s return to one of the keywords we were looking at just a moment ago: “furniture design software.”
This is Ahrefs’ KD score:
Ahrefs gives this keyword a KD score of 3.
That’s very low.
Ahrefs has a logarithmic keyword difficulty scale, which means as the numbers get higher each point represents more and more real-world difficulty.
At the lower end of the scale, though, each KD point represents roughly one backlink needed to realistically compete with the pages that already rank.
However, I’ve found that we can realistically compete for keywords with KD scores less than 5 with no link building directly to those pages whatsoever.
So, unless we’re building out a blog on a website that already has lots of authority, we like to target keywords with KD scores below 10 (and below 5 if we can find them) almost exclusively, making exceptions for keywords that are necessary for some other non-SEO reason.
Compare this keyword to our second keyword: “cart abandonment software.”
Here, the KD is 29.
In my experience, the only way we’d rank for a keyword with this kind of KD score is if we paired it with a targeted, sustained outreach campaign that could eventually build 20-30 links to our page.
In other words, we haven’t got a prayer of ranking unless we devote loads of additional resources.
When building out a content strategy, we want to vigorously hunt for keywords that have:
- A traffic potential of at least a few hundred monthly visitors, and
- A keyword difficulty of less than 10
Of course, it’s not that easy. There’s one more component.
The last and (arguably) most crucial part of a good content strategy is understanding and properly targeting search intent.
We touched on this a bit earlier: SEO allows us to target people at specific points in the buying cycle.
We do this by understanding search intent.
In other words, we need to ask and be able to answer, “What did the searcher want when they typed in [keyboard]?” — and we need to understand where that intent falls in the buying cycle.
Unfortunately, understanding search intent is sometimes more of an art than a science, but (at risk of oversimplifying), here’s a quick breakdown.
Keywords that represent non-buying intent (i.e. things being searched for by people who will never buy, and are thus keywords we want to avoid) include modifiers like:
- “free…”
- “DIY…”
- “…torrent…”
- “streaming…”
- “cheap…”
- “…discount”
…And anything else the signifies “I don’t like spending money.”
Keywords that indicate people who are potentially at front end of the buying cycle — people who might buy if we can sufficiently help them — commonly include problem-solution-type modifiers like:
- “how to…”
- “…tips”
- “faster…”
- “get rid of…”
- “ways to…”
- “…strategies”
- “…solutions”
- “…service”
And finally, the juiciest keywords, buying-intent keywords — the ones that almost always make the most money — usually include modifiers that indicate a user is looking at products and making comparisons:
- “…reviews’
- “best…”
- “…vs…”
- “top…”
So, we obviously want to avoid non-buying-intent keywords. And buying-intent keywords almost always produce the highest direct ROI. However, that does not mean we should only target buying-intent keywords. If we neglect other types of keywords — those problem-solution keywords in the middle — we’re needlessly taking on massive opportunity cost since they’re often the easiest to find.
Instead, we need a mix; we just have to be extra sure we understand what people are looking for and that we produce content that meets their needs in a way likely to convert them into a user or customer down the line.
The goal of a great SEO-driven content strategy is to combine these things and find keywords that:
- Have high traffic potential,
- Are easy to rank for, and
- Appropriately match the most valuable kinds of search intent for our market
Good content strategy and keyword research is not easy, especially for startups who are often establishing new web presences. But the dividends can be huge.
We just need the last piece of the puzzle…
The Gasoline: Tactical Outreach for Startups
Outreach is perhaps the most difficult part of SEO because it has by far the most variables.
It’s analogous to (and possibly even the same as) a sales process: you’re reaching out to real human beings and pitching them. These sorts of processes are by nature less data-driven and more about sweat equity and relationship building.
There are hundreds of outreach tactics out there. Some of them work; some of them don’t. Over the years, we’ve filtered out the ineffective stuff and have zeroed in on tactics that work consistently (for the most part) across niches. A few of our tried-and-true favorites include:
- Guest posting
- Links and resources pages
- Skyscraper
- Infographic promotion
It’d take a novel to write about all of them (or even to cover an handful of them in detail).
So, I’m just going to cover one that we like to use here: it’s called mention link building.
The basic steps behind mention link building are:
- Build the best possible guide on one of the biggest topics in your industry — something people are talking about daily
- Track those conversations and pitch your guide to people who have actively blogged about that topic in the last couple days
It’s a super powerful tactic and can get you really good links. Even more importantly, because you’re reaching out to people who have published articles on this topic in the last couple of days, the links you earn can be timely, which is relatively rare in most types of outreach.
Here’s how we do it.
First, we need to find topics for which blog posts are published on a daily basis. To do this, we use a third party tool called Ahrefs. Ahrefs has lots of functionalities, and we use it for a lot of stuff. Here, we’ll be using their Content Explorer tool, which tracks content, social signals and links. It also allows us to sort by date, which is crucial to these kinds of outreach campaigns.
In Ahrefs, we’d navigate to the content explorer. In the dropdown box, we want to make sure we have “In title” selected, which will give us much more relevant results (it’ll only search for articles with search terms in the title).
Now, we need to search for a few topics. We want to go big. These are not articles we want to rank for; we’re using them expressly to supplement our outreach campaigns. So generally, we can just start with the biggest topics in our industry, and more often than not, if you’ve got a startup, you know exactly what those are.
As a demo, let’s imagine we’re a fintech company selling billing and coding software to hospitals.
If we just type “billing and coding” into Ahrefs, we’ll get some results…
… over 1,000 results, in fact. Looks good right? Almost, but if we narrow this down to blog posts and articles published in the last 24 hours, we get bupkis.
The topic isn’t big enough.
We need something bigger. Even if it’s not explicitly related to our products, we need something huge that people are writing about everyday.
Let’s see what comes up if we type in “medicare.”
That’s a ton of posts, but are people writing about it on a daily basis? Will there be people to pitch everyday? Let’s see what’s been published in the last 24 hours.
Bingo.
Yes, this number (90 articles) is a lot smaller than the previous number, but if even half of those people turn out to be good prospects (and if this is the average conversational output for this topic), we could feasibly send out 40-50 good, timely emails every day.
Of course, we’d need to build a killer piece of content — something big or fresh or interesting or, ideally, all three — we’d have a solid number of people to pitch it to everyday.
What’s that content look like?
The frustratingly simple answer is: to earn links, content has to be really good. Of course, that’s vague, but it’s vague by nature. Content can be as “good” as your resources and imagination allow.
In general, though, “good” = (1) highly useful, (2) extremely timely, (3) totally original, or (4) exhaustive/comprehensive.
A few examples.
Check out this page built by Nerdwallet.
It doesn’t have many words (in fact, it has under 1,000 words). However, it has a a custom, well-designed retirement calculator. It’s both totally original and extremely useful.
How many links did it earn? Try 472.
Here’s another. It’s an article on creatine, of all things, published by Examine.com (don’t read it, or you’ll be MIA for several hours).
Just in case you missed that line at the bottom, it says “Our evidence-based analysis features 735 unique references to scientific papers.”
And it’s true. Here’s the very bottom of their citations.
This article is also (and this is a nice way of putting it) fairly long. It clocks in at over 46,000 words, and it was written, researched and edited by people who have master’s degrees and Ph.D.s in scientific fields.
In other words, it’s probably as exhaustive and comprehensive as a piece of content could possibly be.
How many links did it earn? As of the time of writing: 585.
Obviously, it’s not feasible for most of us to put together content like Nerdwallet or Examine.com. The good news is we don’t have to. This content is on the extreme end of “good.”
A typical article written for a solid, ongoing mention-driven outreach campaign might be 2,000 – 3,000 words. It would need to be exceptionally written, of course, and it would likely need some other special X factor: a fresh angle, an infographic, original data, compiled data, the advice of an expert… just… something.
After writing and publishing our amazing content, whatever it is, we can start tapping into the ongoing conversation we already know exists.
Returning to our hypothetical fintech startup, after we’ve written our piece on Medicare, we’d simply log into Ahrefs every day, look at articles published in the last 24 hour hours (these):
We’d go down the list, and if we saw an article that could benefit from including a link to our asset, we’d give them a shout.
The email might look like this:
Nothing fancy.
Short, sweet, simple, and communicates a very explicit proposition: we’ve got amazing content that your article could benefit from.
If anyone is interested I'd be happy to share the full article.
Best of luck!
[link] [comments]
from Search Engine Optimization: The Latest SEO News https://www.reddit.com/r/SEO/comments/mh4ed1/seo_for_startups_in_2021_and_beyond/>
No comments:
Post a Comment