SEO Advice and Guides

Google isn’t the only game in town. Diversify your traffic sources to be less vulnerable to problems

I’ve lost count of the number of times I’ve written the word “Google” while writing The Truth About SEO. Quite a few times, at least, I wrote “Google and other search engines” – the SEO author’s equivalent of saying “other carbonated beverages are available”.

Let’s face it – there’s no denying Google’s dominance in the search engine space. Only one search engine had its name entered into the Oxford English Dictionary as the verb for searching the internet, and it wasn’t Bing.

Having said that, it may be a small world but it is full to the brim with people… and a lot of them are searching the internet. Just because Bing, Yahoo!, DuckDuckGo, and others hold only a small percentage of the search engine market doesn’t mean that they have a small number of users of a small number of searches being run on them. Duck Duck Go, for example, recently broke the 60 million daily searches barrier. That’s a lot of searches to ignore if you’re not paying any attention.

YouTube player

DuckDuckGo is a great place to go to reach tech-savvy early adopters who value their privacy. Bing is the place to go to talk to people who don’t know how to change their default search engine.

There’s nothing wrong in my view with checking your search engine ranking on these search engines, and there’s definitely nothing wrong with running cost-per-click campaigns on these platforms.

Outside of Google, each other search engine has its own cost per click options available. Competition may be lower in your niche on one of the “other” search engines.

How to Apply the 80/20 Rule to SEO

The Pareto Principle states that 80% of your outcomes will come from 20% of your efforts. It’s a ratio that crops up in a wide range of industries; including economics, computing, sports performance, and health and safety.

In digital marketing the principle is applied as: Spend only 20% of your time on creating content and the remaining 80% promoting it.

The logic behind this is simple:

  1. You are unlikely to get any significant traffic to a website until you have built up a link profile, but link building is slow and resource-intensive.
  2. According to the Pareto Principle, 80% of your traffic will come from 20% of your content – so figure out what 20% that is and only produce that.
  3. Redirect the time you saved in #2 into bolstering your efforts in #1

Wait, what’s a Link Profile?

Practically all SEO experts agree that search engines, in particular Google, use links between sites as some part of their metric for measuring the quality of a site. That’s why there’s a whole section in The Truth About SEO dedicated to talking about backlinks.

The “going rate” seems to be around 30-40 backlinks will give you a website with sufficient “juice” to rank well for any content that you add.

Google like to occasionally deny, or least downplay, the importance of links in their algorithm, but also go to quite some lengths to document all the ways that you shouldn’t get backlinks.

Any links intended to manipulate PageRank or a site’s ranking in Google search results may be considered part of a link scheme and a violation of Google’s Webmaster Guidelines. This includes any behaviour that manipulates links to your site or outgoing links from your site.

Me thinks the lady doth protest to much?

Yes, Google list a huge number of “don’ts” for link builders and that leaves very little scope for automation or efficiency in the process. No, if you want links you really do have to do a lot of old fashioned networking, connecting, and sometimes out-right begging to get people to link to your content. One of the big problems is that a lot of people today know the value of linking. They know that by linking from their site to yours they are giving you something valuable and so, like any valuable commodity, links are something it is hard to get people to part with.

This, for me, is the essence of link building in 2020: It’s really hard.

And this is why applying the 80/20 principle makes sense. Or does it?

Caveat: Don’t forget your ABCs (or ACBs)

My three stage plan for improving the SEO of any website is this:

  1. Fix Architecture: Make sure the thing is built right.
  2. Create or Fix Content: Make sure the content is good and optimised (never just optimised!)
  3. Build Backlinks: Promote the content and build the link profile online

Why is Architecture first?

Fixing website architecture comes first because this is an area where 80% of your problems could come from 20% of your mistakes. A tiny error can lead to your whole site being invisible to search engines or, worse, ripping you out of the search index entirely.

100% of your effort should be on fixing architecture before anything else with your website.

Then, an only then, do you create your killer content and start promoting it.

How to make your life easier – with content.

I have a special “spin” on the 80/20 rule when it comes to content. That 20% of my time that I spend creating I spend doing only that – that means I’m writing for one of blogs, or recording a video, or recording a podcast. Crucially, I’m creating long-form content – long articles, long videos, long podcasts.

This stuff is my raw material when I start link building because the long-form content can then be cut down, re-packaged, re-worked, and turned into content for my various networks; Twitter, Facebook, LinkedIn, YouTube, Anchor, Medium…

All of that, including the editing, comes out of my promotional time budget (the other 80%) and that means my one hour video (or example) should generate a whole range of five-minute clips, sixty-second clips, thirty seconds clips, quote graphics, tweets, and social media posts. A transcript gives me another blog post for my blog and stripping the audio out gives me an instant podcast. It’s the Write Once, Publish Many system, and its the key to make the 80/20 principle work for you in your content marketing.

As you build up a library of content, you will also find opportunities to re-post old content again either because it becomes relevant to a current event or simply because its still “good” (or maybe even evergreen) as you’ve expanded your audience there are people who are yet to see it.

(And, as one last example – the graphics on this post were made months ago as promotional material and now I’m recycling them here!)

What if all you ever did was create content?

Google’s dream is of a web where all links are organic and spontaneous; kind of how the web used to be before people figured out that getting backlinks was a really good way to accelerate your Google ranking.

If you want to join the ranks of the moon children, you’re welcome to do no backlinking at all. Your site will (or should) still rank, but it won’t be able to compete against other sites in your niche unless your content really is utterly unique and truly fantastic. Of course, if you have such content (like me 😉 ) then you’ll find people link to it anyway… eventually. (Maybe)

The problem is with this approach is that it is slow and ultimately unreliable. You might be lucky but while you’re marching to the beat of your own drum and waiting for someone to sprinkle link-building fairy dust on your website, your competitors are hard at work grinding our their links.

Don’t miss the obvious opportunity – backlinking is essential.

How much is too much eCommerce category content?

When optimising an eCommerce website one of the first things I check is content on the category pages.

In the majority of cases I find that there’s no content on the category pages other than the products and filters. No description of the category, no information on the product range, no advice for new buyers… nothing.

It’s an easy error to make as many content management systems don’t make category level content easily available. Those that do often provide just a small plain text area. (I’m looking at you, Shopify and WooCommerce).

But, after your homepage, category pages should be a key landing page in your SEO strategy. Customers will often be searching for a type of product, not a specific product by name. The mistake many eCommerce site owners make is to ignore their categories and therefore weaken their position for these searches.

What should you do with eCommerce category content?

If your eCommerce site currently doesn’t have descriptions at the category page level, you should add them as soon as possible.

Category descriptions should contain information that will help the customer make an informed choice on which product they select, promote key features of a brand (for a brand category) and reinforce your own expertise in the category.

Category descriptions don’t need to be long. Remember, this function of this page is to move the customer on to the right product and convert them there – this is not the place for your full sales pitch.

eCommerce Category descriptions for boring products

eCommerce category descriptions can also be very useful if you’re dealing with a range of pretty dull products.

If you’ve got, for example, a category full of taps then there may not be much you can to describe them differently. They may have different specifications but, from a descriptive perspective, there may not be much to hang your hat on.

Your category description is your friend here – put everything you want to say about the entire range of products here rather than repeating it on every product. Hopefully, you’ll find there’s more to talk about across the range than on an individual tap.

What you shouldn’t do with eCommerce descriptions

Just to reiterate – category descriptions do not need to be long. They should provide enough info to guide the customer to the right product for their needs.

I’m not the only person who knows eCommerce category descriptions are important, but there are people out there dumping the equivalent of War and Peace into their category descriptions. That doesn’t help the user and doesn’t help your SEO either.

Be useful, be informative, be unique.

More info:

It’s not all about rank. Ranking 1st for a keyword that generates no sales is pointless.

I decided to leave this until the very end of the book as ranking websites for useless “junk terms” is literally the oldest trick in the SEO book.

I can definitely, 100%, without question get any website to the number one position on Google – if I get to choose the keyword.

Don’t fall for it.

The purpose of your website is not to be no.1 on a search engine. It’s to generate as much return on investment as possible.

Your SEO project will never, ever be finished. You can always improve. So can your competitors

SEO is an arms race. If you’re currently in #1 position for a lucrative search term, chances are the everyone from position #to infinity are gunning for you.

So, it’s not just SEO that is changing – it’s your competitors.

Maybe you’re not number #1 right now. Let’s say you’re number 4. The difference in click-through rate between the Top 3 organic results and the rest of page one is sharp – you want to be Top 3 if you can’t be Number 1. The only way to get one of those coveted top spots, of course, is to knock someone else out.

“This ain’t a scene, it’s a goddamn arms race”

Fall Out Boy

This is why your SEO project (encompassing all aspects of SEO, SEM, UX etc. as we’ve talked about) will never be finished. Even if the “rules” of SEO stood still, even if Google finally said “Yep, cracked it, that index is getting no better than it is today” you would still be up against the changes, improvements, and additions your competitors make.

That’s why they’re called “the competition” and not “people who do the same things as me”.

Google is not a black box. Read Webmaster Guidelines and watch Webmaster Help Videos.

SEO is full of contradictions – things that work for other people won’t work for you, things that didn’t work for other people will propel your sites to heights undreamt of. That’s the way the search engines need it to be – because if everyone knew exactly how to position a site at #1 in the search engine results, then everyone would do it, #1 position would become worthless and so would the search engines themselves.

Google, in particular, is not as much of a “black box” as it once was. They publish clear guidelines for search, they tweet regular pieces of advice about search engine algorithm changes, and they even share the guidelines that they give their human moderators on how to grade pages.

Wait a minute, did you say human moderators?

Indeed I did. It’s not the most widely known thing in the SEO world but Google employs thousands of people worldwide to grade websites and web pages manually. These gradings are fed back into the system and are, in many ways, the raw data from which improvements to the algorithm are formed.

Google wants to give you the “best” answer and has its algorithm tuned to do just that. But was does “best” mean? That’s why there are human moderators, testing the search index against the criteria that Google have written down for what they think makes a page good (or “best”).

And, yes, you can get a copy of this document. It’s a bit of a beast, 160 pages plus in the latest version I saw, but it is well worth reading – or giving to your web developer and asking them if they’ve built your site with these in mind.

(Hint: They will probably say “yes”, so check yourself. Be your own human moderator.)

The reality of moving website providers

In my experience, it’s rare for one website company to take on the work of another.

Developers hate working on what we call “legacy systems” – systems with a history we aren’t fully aware of and that we can’t be 100% sure is properly documented. 

Taking on a website from another developer/agency and providing support on it is like taking on the warranty for a second-hand car that you’ve never worked on before with no service history and the strong suspicion that that rattling noise under the bonnet is probably something serious.

The reality is that most migrations from one website agency/developer to another involve a rebuild of the website.

This doesn’t mean that it is impossible and this is a strength of working with one of the larger Open Source platforms such as WordPress. There are lots of WordPress developers around getting one to take on your website shouldn’t be that hard – just don’t be surprised when the suggestion of a rebuild comes along.

Be honest – do these guys just want my money?

In short – yes, of course they do.

The longer answer is that what they want, like any business, is profitable contracts.

Taking on a site built by someone else, even if you understand the technology in detail, always carries a risk. The value of the support contract, therefore, has to mitigate that risk, or the person taking on the website is potentially going to be losing money. Replacing the website is therefore more cost-effective; a higher project cost but lower ongoing running costs.

Don’t freak out when you don’t rank for something you didn’t optimise for

Sometimes, this happens…

“Hey, SEO guy!”


“We don’t come up on Google when I search for hats.”

“We don’t have any pages optimised for hats. We also don’t sell hats.”

Alright, I might be exaggerating for effect, but not by much.

Remember, SEO is weird. You will rank for things you don’t expect to. You may not rank as well for things you expect to as you want to. You definitely won’t rank for things that you don’t have on your site. 

Think about it – why the hell would you?

PPC is not SEO but good SEO can reduce the cost of your PPC and make your budget go further

Did you know that the cost of an advert in Google Ads is directly related to how well your web page matches the search terms you’re advertising for?

It’s true: It’s cheaper to advertise against terms your site is optimised for than terms it is not optimised for.

Why does Google do this? Well, my view is that it’s part of their approach to ensuring that their index, including their paid for listings, are as egalitarian as possible. By ensuring that websites that are highly relevant to a particular keyword or term can advertise against that keyword/term inexpensively, Google are trying to level the field between small or new businesses and the larger, more established players.

There’s also another way of looking at this – it helps Google too.

Google sell answers. Forget every other service they offer, every other project they are involved in, their number one cash-cow that underpins the whole thing is search.

If every time you went to Google to find something, to get an answer, you were inundated with a bunch of irrelevant adverts that took you to websites that weren’t relevant to your query, you wouldn’t be a happy customer. Maybe, if this happened enough, you wouldn’t be a customer at all.

So, by encouraging high relevant sites to advertise, Google is actually improving the quality of the adverts on its site. 

Either way, the effect for website owners is that the better your site is optimised for the keywords that you want to advertise for, the better price you will get per click from Google. 

That’s got to be worth investing some time in.

Have zero tolerance for duplicate content. Eradicate it at every turn

Search engines hate duplicated content. Google, reputedly, has a “good cop who just watched his partner get gunned down by the local mob boss at his own daughter’s wedding when he was just two days off retirement” level of hatred for duplicate content.

Eliminating duplicates applies to:

  1. Page titles
  2. Meta titles
  3. Meta descriptions
  4. Keywords
  5. Page content

Think about it this way: 

  1. Every page on your website must have a singular purpose/topic it addresses.
  2. Every page should be the best page on that topic.

If you’re able to tick checks 1 and 2 for each and every page on your website, duplication should be at an absolute minimum.

If you have two or more pages that address the same topic – combine them into one, really great, page. Users, and search engines, will love you for it.

Handling duplicate content with canonical tags

Sometimes, eliminating duplicate content is not as easy as we would like it to be.

This is often the case when trying to optimise eCommerce websites where there are multiple versions of the same product, perhaps with only small (but important) differences between them.

Thankfully, there is a technical solution to this problem – the canonical tag.

The canonical tag allows you to specify that a given page on a website is not the original (or “canon”) version of that page’s content and that another page should be indexed in its stead. The benefits of any links pointing to the page should be passed on, in whole or in part, to the page that is the true original (“canonical”) version.

Doing this can reduce duplication in the search index and promotes the importance of the original page. 

However, a word of caution… the pages that are canonicalised to another page disappear from the index.

Canonicalisation Example 1: Good canonicalisation

At we used to sell the Widget A1. Now, we’ve released the Widget B1. There’s not a huge difference between A1 and B1.

We want anyone searching for a widget with the properties of Widget B1 to find it, and not to find Widget A1. We suspect A1 is stealing some of B1’s thunder, but we don’t want to remove it from our website because existing users may want to access some of the documentation etc. for the A1.

By setting the canonical URL of widget A1 to point to widget B1, we make B1 the default page. Any links pointing to the A1 page now pass their link-benefits onto B1 and any search engine index results for A1 will be redirected to B1. However, customers can see still A1 on our website and find it by navigating to it from our homepage.

Canonicalisation Example 2: Bad canonicalisation

At we make the Widget C10. We also make the C11, C12, and C13. They are quite similar but there are key differences. Our widgets are well known in the industry (of widgets) and people will often search for them by name (e.g. “Widget C12”). 

The widgets sell for different prices and our eCommerce platform holds each as its own page. All of the pages are indexed at the moment, but our SEO consultant is telling us we have duplicate content and some of them need to go.

Is this a good case of canonicalisation? No.

If we canonicalise C11, C12, and C13 to C10 we would effectively remove these from the search index. Finding them independently would be impossible.