Ongoing SEO Processes

After you have optimized your platform to consist only of

• targeted pages
• distribution
• a startpage
• correct deadweight handling

you must focus on the ongoing SEO processes.

The ongoing SEO processes are the things you will need to do continuously to see SEO success, such as:

• creating new content
• updating old content
• marketing your webproperty in new and different ways

Maintaining your ongoing SEO processes is just as important as everything else we have discussed in this book because without SEO processes all your platform work will lead to nothing.

All SEO processes, as you will see below, are formulated as questions. Why is this? Because the answer to the processes is different for every company.

Companies have radically different activities, both online and offline, and, even in the case that the technical setup is the same, the people, the vertical, the market and the overall context are always different. The answers to the processes must come from within the company, and cannot be blindly copied straight out of a book, even one as awesome as this.

An SEO process that works for one company can not be 1:1 copied to another company. That’s why all I have are questions and experience. The answers must come from you!

Once your platform is optimized for SEO, the following SEO business cases will be possible:

• if we add x% more targeted pages to our site, we gain y% more traffic
• if we update x% of our targeted pages, we gain y% more traffic
• if other sites point x% more sensemaking links to our targeted pages, we gain y% more traffic
• if we make the user experience of our site better, our traffic gets more sustainable
• if we gain more brand recognition, our traffic gets more sustainable

Therefore, these business cases lead to the following questions:

• How do we create x% more targeted pages?
• How do we update x% of our targeted pages?
• How do we get other sites to point x% more links to our targeted pages?
• How do we ensure that the user experience of our site gets better?
• How do we ensure that our site gets more brand recognition?

Let’s go through them in detail.

Business Case #1: How do we create x% more targeted pages?

Once we have a set of targeted pages that performs this questions arises: “How do we ongoingly create more targeted pages?” The logic behind this is quite simple: If the CTR between impressions and SERP clicks stays more or less the same, and then we create more targeted pages, we get more traffic.

There are a few things to consider. An optimal case might be 100% more pages equals 100% more traffic. But that said, different targeted phrases have different search demand and varying competition in the SERPs. So, 100% more pages can lead to 150% more traffic, but can also lead to only 20% more traffic.

It’s not possible to calculate how much traffic x% new pages will bring you, and all the paid SEO tools which promise you this are just guessing. The one thing you can rely on though if you optimize your SEO platform correctly is: more targeted pages equals more traffic.

You must be very mindful of internal duplicate targeting: It’s not about more pages! It’s about more targeted pages. If you create more and more pages that want to get found for the same targeted phrase, you will actually lower your overall search traffic performance due to internal competition.

You must also be aware of internal duplicate content: If you are automatically creating a unique combination of content (such as lists), you must make sure that every combination is unique. It’s not enough that the sorting is different. As a minimum, the utility, targeting phrase and at least one list item (but really the more, the better) of the visible list must be different.

For better visualization: A, B, C, D, … are list items.

• A page with A, B, C and another page with A, B, C are internal duplicate lists.
• A, B, C and A, C, B (different sorting) are also internal duplicate lists.
• A, B, C and A, B, D (one item difference) are unique lists.
• A, B, C and A, B (one missing item difference) are unique lists.

Also, with more targeted pages and therefore more traffic business cases you must be careful of internal interlinking. Internal interlinking is easy with a small, more or less static number of pages. It gets hard with a big and growing number of pages. Make sure that all pages are sensemakingly interlinked, with a touch of serendipity.

SEO does not care about the (main-)navigation. The argument that you cannot create more and more targeted pages because “where will we put it in the main navigation?” is not valid. Nobody cares about your ultra-complex, hierarchical main navigation - not the user, not Google.

One easy way to create more pages is to lower the quality criteria (QC) of the targeted pages. You could:

• adopt shorter lists (for example, job pages with just two vacancies instead of three)
• add less content on the pages or
• look for “cheaper” created content.

This is a valid strategy, but be sure to realize: the lower the QC, the higher the volatility of search traffic. The pages may perform for a while, and then not. You may decide to update them, and they perform again, but when the competition moves in with higher QC pages, you will lose traffic. It becomes a difficult cycle to keep up with to maintain steady traffic.

In general, you should start with a very high QC, and only lower it slowly. As soon as you see any traffic volatility, work hard to improve the quality of your targeted pages once again.

Important: if you create more targeted pages and you don’t get more traffic then something is wrong. In this situation, stop creating more pages and look for the reason (most likely a QC, internal duplicate content or internal duplicate targeting, internal interlinking or distribution issue, web performance issues, or a combination of them).

Sliding Content

One common issue with the case of creating more pages is that there is quite often content that only has limited validity. In most cases, that content is only valid over a period of time, for example, job offers, used car sales, special offers, and events.

For targeted detail pages (pages that list only one item) this leads to a loss of pages. An item existed, and then it doesn’t. For targeted list pages this might also lead to a loss of pages, as the QC might not be fulfilled any longer (so for instance, a list of four listings suddenly has three entries, and the QC is > 3).

But if we want sustainable, growing traffic we cannot delete and recreate pages all the time. So what to do? The answer differs for targeted list pages and targeted detail pages.

For targeted detail pages (TDP), I recommend reusing the pages. For example, perhaps an event targeted at [happy puppy pet show vienna] happened sometime in the past. You can keep the page around, provided there is a big visible disclaimer that this is an old entry (and include some further links to new events that might also be interesting). If the event happens again (maybe next year for example), you simply update the page.

If you want to reuse the same URL, you must make sure there are no dates in the URL (dates in URLs are a dependency and violate URL rule number two: permanent).

If reusing pages does not make sense or is not possible in your vertical, I recommend to just delete the old entries (ideally with a HTTP 410 status code). Do not keep old pages around “just because” - they will lead to internal duplicate targeting and an unmanageable site quite soon. Also if you can’t offer a good user experience on the old targeted detail page, just delete it.

For targeted list pages, I recommend using old listings to fill up the pages. So for example, in a case where you have a QC of at least three entries on a page that shows up to 15 list items, but you only have one active listing and 10,000 old ones, you would just list the one active and 14 old entries (no pagination). You used the old ones to fill up the page with up to 15 entries.

This keeps your list page in existence, with the logical consequence being that once a valid targeted list page existed, it keeps existing, and attracting traffic.

This can be seen as quite an aggressive SEO strategy. Showing old content can be valuable to the user (at least they know that there was an offer, event, job, … in the past) but it’s not easy. You must make sure that the usage metrics of these pages and the user experience is still up to your standards, and the utility of these pages to the users is still given.

In any case, do not mislead the user! Make it clear via a strong visible indicator that the old entries are outdated. You might not even link the old entries to the targeted detail pages (in the instance that they no longer exist).

You can offer the user other options, link to other useful pages or even link to your competition if it makes sense. But the user must not bounce back (much) more to Google than on your other pages.

Overall, the “keep old stuff around” strategy is easy to do in a bad way, and hard to do well.

Business Case #2: How do we update x% of our targeted pages?

Another underlying business case of SEO is that if “we update x% of our targeted pages we will gain y% more traffic.” Well maintained web pages deliver a much better user experience, with the bonus that Google is looking to deliver diversity in its SERPs by showing old established web pages, as well as pages with “fresh” up-to-date content.

Updated (up-to-date) pages sometimes perform astonishingly better. We did an experiment with one client and updated 50 targeted pages (added information, structured the page content better, deleted some outdated content) and the pages attracted 650% more traffic.

So what does “update” mean? It means adding content, but not “let’s add another paragraph of text.” The goal should be to add to the diversity of the content. If you have paragraphs of text, add a bullet point list, later add a table, then a video or infographics and so on.

Updating your content also means to delete outdated content. It means restructuring the content to be more effective (for example, if information is woven into a paragraph of text that nobody reads anyway, restructure it as a bullet point list or table). In the case of list pages, updating most of the time means simply bringing the list up-to-date. Of course, adding diverse supplemental content to these pages helps, too.

Even though it sounds simple, updating content is often a major challenge for organizations. This is especially true for companies where everything needs to be checked with legal, or where the CEO wants to have the last word about everything that appears on the website. These are operational challenges which need to be solved, as there is no way around updating targeted pages.

So, how often must you update your content? All the time - it’s an ongoing process. It obviously depends on your vertical: if you offer health information, your content must be up-to-date as anything else is a horrible and potentially dangerous user experience. Whereas, if you offer information that isn’t as critical, it can go longer without an update.

Targeted list pages should also be updated in or near “real time,” so if you have a new list item that belongs on the list, update the list.

As a general rule of thumb, I recommend updating 2% of your editorial targeted pages every week, which means roughly 100% of your editorial targeted pages in one year. But please, just use this as a rule of thumb as it may be different for your vertical or business.

However, there is one thing that is true for all verticals: if you want your targeted pages to perform better over time, you need to update them. After you created a page, the work is only just beginning. You then must establish a content update process within your company. Updating pages is as important as creating pages.

Business Case #3: How do we get other sites to point x% more links to our targeted pages?

With more inlinks from other valuable webproperties, we can expect our existing targeted pages to perform better overall for their targeted phrases, therefore getting more clicks from the existing impressions.

And although this is not a book about link building strategies (and I will never write one), there is one thing I have learned over time, and that is that every company must find their own strategy to attract links.

Google has set some quality standards when it comes to the topic of links. You can read these in the [Google Webmaster Guidelines]. If these guidelines are violated, Google will automatically or manually penalize the site that set the link, the site that was linked or both.

Whenever you do something majorly wrong with your links, Google will send you a message via Google Search Console. This message includes samples of what you did wrong from their point of view and what you should do now to fix it. Simply do what they tell you in the message. If you do not, the penalty might result in a lack of growth, diminished search traffic performance or even complete removal from Google.

Remember: whatever you do for inlinks, you must make sure to get additional value out of these activities, as 75% of all resources in plain link building are just wasted.

However, the web is not a one way street, and good sites link to other good sites. It’s good for user experience, it’s good for your usage values, and it’s good for Google.

Often people are afraid that when they link to other websites, that they will lose the user. This idea is somewhat foolish - if the user can’t go further, you have lost them already - they will bounce back to Google and remember that your website isn’t a very good resource.

Therefore, link as much as you want to other web pages that you wish to recommend to your users. Do not think about SEO while doing so. That said: if you have a direct monetary or material benefit (see Google’s spec [rel nofollow google] ) you must set these links to rel="nofollow":

<a href="https://www.example.org/external-page" rel="nofollow">Somewhere Else</a>

It’s important you don’t obsess over nofollow, as in 99.9% of all cases your external links will benefit you and your users. And in the cases where Google thinks that you did something shady (either receiving or giving “unnatural” links), it will tell you so in the Google Search Console.

Any SEO which tells you to “nofollow” all your links or “nofollow” links to “low Whatever-Rank™ websites” is an idiot and should be treated as such. “Nofollow” is just a peripheric topic for SEO. Just follow the Google spec.

Now for some advantages and disadvantages of common link building strategies:

• Natural Links “Creating awesome content and people will link to you.”

  • Advantage: All resources end up where they belong on your webproperty.
  • Disadvantage: People are lazy and hardly ever link to anything if not actively encouraged.

• Paid Links “Paying sites (directly or via a link network) and get more links over time.”

  • Advantage: It’s easy.
  • Disadvantage: It’s a clear violation of the Google Webmaster Guidelines. May result in an underperforming site or complete removal from Google Search. Links are mostly poor quality and clearly identifiable as spammy, resulting in a bad brand experience. I, and Google recommend not to do it.†

• Link Bait “Creating an awesome targeted page that is so good that it goes “viral” and gets you lots and lots of links.”

  • Advantage: If it works, it gets you lots of links and referral traffic (social media), as well as being a great marketing opportunity.
  • Disadvantage: Very hard to do. 99% of all approaches you try will not work and only have a minor response. If something worked for another company does not mean it will work for you.

• Outreach “Pitching your website or targeted pages to other websites, mostly blogs.”

  • Advantage: This is permitted within Google Webmaster Guidelines as long as no exchange of money or goods is involved. Integration in the micro-communities of the blog.
  • Disadvantage: Time intensive. And annoying.

• Guest Posts “You posting a guest article on another website, and including a link from the article to a targeted page on your site.”

  • Advantage: Easy, fast, very good manageability of what gets linked and how (since you write the anchor text yourself).
  • Disadvantage: If done badly, it can be a violation of Google’s Webmaster Guidelines. If the anchor text is overdone or there is a clear link building purpose, it may have major negative impact on both parties involved.‡

• PR “Classically, telling lots of other sites how awesome you are or how awesome what you are currently doing is, and they might link to you in their resulting coverage. This is a specialist task but can be incredibly successful.”

  • Advantage: Within the Google Webmaster Guidelines. Lots of additional value from positive PR.
  • Disadvantage: Time intensive. Most PR is boring and sucks.

• Widgets & Badges “Creating some manner of embeddable content, such as image-badges, iframes or javascript-embeddables that link back to you.”

  • Advantage: Good manageability of what gets linked and how (anchor text).
  • Disadvantage: If it’s spammy, it’s spammy. If the links are not sensemaking or not on topic with the widget, then it’s a violation of Google’s Webmaster Guidelines. It’s best just to link your brand to the startpage. Most sites will not integrate your widgets/badges.

• Deep Integrations “Part of your content or service gets deeply integrated into another webproperty, whereby the other webproperty links to you. For example, a travel website fetches (via scraping or an API) weather data from a weather website, which is displayed alongside the main content. The weather website is cited as the source, with relevant links.

  • Advantage: Both websites win, with the one gaining useful structured content that can be used for the diversity of content, and the other website gaining links to their targeted pages, increasing their visibility and having a return link.
  • Disadvantage: Complicated, and lots of effort for both parties involved (for example, one site must offer an API, the other must integrate the data). You also need links from many different web properties, not lots of links from a single one. Might be considered spammy or even spam by Google if the integrations or links are nonsensemaking.

• Your Strategy There are other strategies and combinations of different strategies, depending on your business.

  My recommendation is to go with the strengths of your business - if you have a strong sales team then don’t be shy to ask them to ask (ask, not force) clients for links (“as we are partners anyway”), and even make it part of the sales team’s KPIs.

  • Advantage: If you have creative developers, you can use them for awesome link bait single page apps, embeddable widgets or more advanced integrations. And if you can do great marketing, use this, as your marketing team will have millions of ideas.

    If you are changing the world for the better, or have an interesting angle for press attention, go with PR and don’t forget links.

  • Disadvantage: The wrong link strategy can break a company because the resources are not focused on your product or website, but on an external factor. In many companies, there is quite often a misallocation of resources towards “link building” using the worst aspects of the techniques mentioned in this table. In the long term, this will kill a company.

†Also on this topic, do not participate in the link provider side of this deal. It is common for sites to be approached and asked to accept links in exchange for payment (or another reciprocal arrangement). What happens is if you sell links, Google will remove your links from the equation, and they will no longer count as a vote. In the long term, your site will perform more and more poorly, and you will deliver a bad brand experience for your users.

‡Only do this if it really makes sense to get covered on that particular website (one that’s within your vertical and has the relevant opportunity for articles). The value of this method must be in the publication of the article and reaching the audience of the publishing site; any link back should only be an afterthought. Don’t overdo the links - it’s best just to link your brand to your startpage and maybe one specialized on topic targeted page.

§ Note: For iframes and javascript-embeddables the link must be outside the actual widget. For badges, the alt-text of the image substitutes the anchor text of the link.

Basically, whenever you do widget or badge-bait, you need to treat it as a product with a product manager and a “sales” team that asks, begs, encourages, and helps sites to integrate the widget/badge.

There is no such thing as a “viral” widget or badge - you must plan your distribution strategy together with the widget. Most websites will not integrate your widget no matter what you do, so you can use the widget as part of your outreach strategy. Ask them if they want that awesome widget, and if not, ask them if they will at least link to your awesome website.

To end this “link growth over time” topic: Getting links is hard and mostly wasted work. It’s easy to do link building badly and spammy and therefore with a negative value. None the less, you need valuable and good references (links) from other websites. Find a way that works for you (together with your other outreach and brand building strategies) but don’t over prioritize the whole “link” part of “link building.” You need good references, not “links.”

Business Case #4: How do we ensure that the user experience of our site gets better?

It is true that “if we make the user experience of your site better, your traffic will become more sustainable.” If you have bad UX, some pages might sometimes perform for a short period of time, most of them won’t. With awesome UX most of your pages will perform most of the time for a long time.

Every company must have at least one person responsible for UX. This person should not be the person responsible for SEO. UX is a big part of SEO, but UX is not SEO. UX is bigger!

The most crucial factor when it comes to user experience is speed. A slow site greatly affects the user experience, and most of the time it’s a gradual decline from a fast site, to okay, to “little bit slow”, to “completely and utterly broken slow.” As I have repeatedly said, slow sites do not perform because they deliver a terrible user experience and a bad brand experience.

Of course, user experience is not only about speed. Things such as feature bloat and ads can also have a major negative impact. If you continuously add more features to your site, then your user experience will decline.

If you earn money from advertising with ads, then you can always earn more money with more ads. You might only make gradual changes at first (one more ad placement here, let’s put the ads a little further up …), and then one month later, when you see the additional money these small changes have earned you, you will do more … and more … and more of those.

Over time you will effectively destroy the UX of your site, and you can’t go back because any change (fewer ads or, less aggressive positioning) will have a negative impact on your bank account.

Therefore, if you want to maintain the existing traffic you have and continue to grow it, then user experience is a must. Set some usage metrics as UX KPIs, with the most simplest one being:

• bounce rate (down is good)
• time on site (up is good)
• average pageviews (up is good)

Measured via Google Analytics, make sure that they only get better, and never worse. If they do get worse (and occasionally they will), then actively counter it. Don’t be afraid to drag the ads into the discussion. Ads are not out of the scope of user experience only because they earn money.

Business Case #5: How do we ensure that our site gets more brand recognition?

In the past, I have always summarized brand recognition and experience within the user experience business case. Yes, the most important branding tool you have is your website once the user has reached it. So user experience is a big part of brand experience, but it’s also more.

So how can you make sure that your site continuously gains more brand recognition? You can simply go into Google Search Console > Search Traffic > Search Analytics and investigate both:

• branded searches (queries which only contain your brand or an alternate spelling of your brand (so for example, if you are AirBnB you would look for [airbnb], [air bnb], [airbb], and so on)
• generic + branded search combinations (queries which contain your brand and some organic other part for example: [airbnb berlin])

And make sure that this traffic segment is growing over time. All sites that have issues growing their traffic, even though “they have tried everything,” have a diminishing amount of branded traffic. Even “simple” things like a logo change might have a major positive impact on such underperforming sites.

In addition, think about how to reach users before they reach your site. Any kind of “nonannoying” advertisement is ok, for example:

• Offline: TV, print, billboards, radio, word of mouth, and flyers
• Online: PPC, AdWords, AdSense, retargeting, social media ads

I said “not annoying” because you must be careful to not “negatively brand” your users. Remember if they don’t actively click on you when they see you in the SERPs anymore, you have lost and failed in the business of SEO.

Ongoing Feedback

Okay, so let’s say you’ve done everything written in this book: you’ve created the perfect optimized web platform, you’ve set the right KPIs, you’ve introduced SEO processes such as ways to create new content and update old content throughout the entire company, plus everything else. And then everything turns to shit.

Mistakes happen, wrong decisions happen, shit happens. Always, all the time. And there is no way - no test, no setup, no experienced manager, no megastar 10x developer, no overpaid consultant - that can prevent this. And, as you cannot prevent these things, you must react to them, and to react to them, you need to know when something happens. That’s why it’s so important you constantly get feedback.

Many of the things that SEO involves are obvious, especially the content parts. You browse your web page; you see that there is no content or the content is rubbish, and you react accordingly. But, as you most likely have realized, many aspects of SEO are not so easily visible, for example:

• page speed
• distribution
• user experience
• brand experience
• canonicals, and the list goes on.

That’s why there are two tools which you must use daily. Otherwise, everything will go wrong. They are:

• Google Search Console
• Google Analytics or another accessible analytics tool

Small websites should look at Google Search Console at least every two days for a minimum of half an hour, and they should look daily for ten minutes at their usage values via Google Analytics. If you see something that you do not understand or you think looks worrisome, investigate deeper.

Bigger websites should have at least a part-time position for someone who looks at these tools. Really big websites should probably have their own team designated to them. These positions must be defined and their responsibilities (GSC, GA, report findings, initiate action) clear. These are the minimal viable SEO feedback setups.

An even better SEO feedback setup: once the platform is optimized, it must be easy for everyone to access the feedback of how their actions are affecting the website. Whenever somebody does something, they must receive direct feedback and know what it means, including the consequences.

For example, if you have a marketing team, the marketing team must know what works and what doesn’t. When did the brand searches rise? When did it drop? What gets inlinks, what doesn’t?

The development team must know the impact of their changes in the real world. Speed is not just the responsibility of DevOps (the server guys), it’s the responsibility of everyone who touches the site. Distribution and traffic development must be set into the context of application releases.

The content team must experience direct feedback how “their” content is performing, how much traffic it gets and why (search, social, internal). Feedback must be delivered where it belongs, to the people who do the actual work.

If the SEO feedback processes are implemented than a virtuous cycle can start. The content will get better (from both a SEO and user point of view), marketing will get more creative (as they will see that creativity works) and development will be more accountable and responsive. You will start to see results and be able improve on them.