It goes without saying that the March 2018 Roundup will be one to remember. It contains a lot of actionable information, alongside no less than 47 takeaways from an Ask Me Anything (AMA) session with John Mueller.

TLS v1.0 Depreciation

Most companies use some version of Transport Layer Security (“TLS”) for securing HTTP. As of June 2018 however, all transaction based websites must upgrade their security according to the PCI Security Standards Council (“PCI SSC”).

Google has not yet implemented any penalties for websites failing to transition to a newer TLS version, but we believe that web browsers may show a website as being “Not Secure”, much like the mixed content warning already in place.

The following cURL example returns the Google UK homepage using a TLS v1.0 connection.

curl https://www.google.co.uk --tlsv1.0 --verbose

See the Google UK cached cURL request here.

We have disabled TLS v1.0, so you can cURL the MERJ homepage to see what it looks like when the connection is refused.

curl https://merj.com --tlsv1.0 --verbose

See the MERJ cached cURL request here.

Action

Businesses must remove TLS v1.0 before June 2018 to remain PCI compliant.

NGINX SSL module documentation shows the protocols to use. Remove “TLSv1.0” from the list.

ssl_protocols TLSv1.0 TLSv1.1 TLSv1.2;

Let’s Encrypt Announces Wildcard SSL Certificate Support

On 13 March Let’s Encrypt announced that both ACMEv2 and wildcard certificate support had gone live.

The organisation stated that ACMEv2 (the updated version of its ACME protocol), had been designed with industry feedback, while also completing the IETF standards process.

Wildcard certificates allow webmasters to secure all subdomains within a domain using a single SSL certificate, making certificate management easier to handle for people with multiple domains.

In order to use ACMEv2 for certificates, webmasters will need a client that has been updated to ACMEv2, although it’s Let’s Encrypt’s intention to transition all clients and subscribers to ACMEv2 at some point in the future. No end-of-life has been set for ACMEv1 as of yet.

It’s also worth noting that wildcard domains must be validated using the DNS-01 challenge type, which means that DNS TXT records must be modified in order to demonstrate domain control.

Follow the ACME Github project for ACMEv2 updates.

Action Point

Check your DNS A and CNAME records that do not have SSL if intending to use ACMEv2.

Ensure www and non-www subdomains are secure and redirected correctly, as we have seen many websites forget about their alternative version.

If you have a staging/UAT website, add Let’s Encrypt to ensure it mimics the production website.

Lazy Loading Images

After admitting in 2015 that it can’t see lazy loaded content on mobile, Google has pledged to create advice and help documents so that webmasters can make lazy loading content and images more discoverable to GoogleBot.

Speaking on Twitter, John Mueller said that, “There are various ways to lazy-load images, it’s certainly worth thinking about how the markup could work for image search indexing (some work fine, others don’t). We’re looking into making some clearer recommendations too.”

The documents should also help webmasters understand how to reduce loading times when utilising lazy loaded content.

Action Point

We recommend using Lazyload by @verlok.

Use Fetch & Render testing to ensure images are visually available. Wait for further updates to Google’s documentation.

Google Image Captions Use Page Title Element

Writing in Google’s blog, The Keyword, Paul Banister announced on 13 March that captions will now be added to images in mobile search results.

Rolling out globally, they will be displayed for mobile browsers and the Google app (iOS and Android), and are taken from the title attribute on the page that a particular image is hosted on.

It is hoped that the captions will give searches more context so that they can easily discover what an image is about and whether the host site could contain more relevant content.

You can check the difference below:

Google pulls through page title in to image search results to aid accuracy of results

A spokesperson for Google stated that although the search engine will be using page <title> elements for the moment it could change this in the future “to improve the experience” for mobile users.

Speaking on Twitter, Danny Sullivan stated that HTML title elements are not used in any particular way for ranking an image, so webmasters should continue captioning images exactly as before.

Action Point

Consider revisiting title elements if you have not done so in the last six months. Ensure any time sensitive dates are updated, i.e.

"Best [thing] Reviews 2018"

Mobile-first Index Rolling Out & Best Practice Guide Update

Google announced on the 26th March that the new mobile-first index has started rolling out. This expands on the small sample set of websites during an 18 month experiment. This coincides with the mobile-first index developer guidelines being revamped. Webmasters will be updated via a Search Console alert once their website has switched over.

Minor Search Console Updates

On 19 March, Google Webmasters announced on Twitter that a series of visual changes to Search Console had been made to provide webmasters with more context to data included in reports.

Annotation cards have been added to bullet points within error reports so that webmasters can learn more about an issue on a particular date. By clicking the “view issue” link on the card, webmasters are also able to validate whether the issue has been fixed.

Furthermore, a historic line graph column has also been added so that it is possible to see how the data point has changed over a given period of time.

The filter/compare feature has been redesigned to offer a new look and pre-populated values have also been added.

More generally, improvements have also been made to the date picker and comparison view.

Action Point

Familiarise yourself with the new changes and make sure to utilise the new annotation cards within error reports for validating fixed issues.

Bing Confirms Support for JSON-LD formatted Schema.org Markup

Although not yet available in Bing Webmaster Tools, Bing has confirmed support for JSON-LD formatted Schema.org markup as it continues to actively bulk out its toolset.

This is particularly noteworthy as existing support includes validation for schema.org markup in Microdata or RDFa format only.

When pressed on the issue, Bing’s Head of Evangelism, Christi Olson, said that: “Bing has been using JSON-LD as a signal, but we are still building out the verification tools as part of the Webmaster Tools offering.”

As of yet, Bing hasn’t confirmed when support will become available to webmasters, but we’re confident that it be featured within a Roundup in the next few months.

Action Point

Like Google, as Bing will be using Schema.org, it is possible to use Google’s rich data testing tool. We shall wait to see if Bing offers any additional information.

Bing Releases Entity Search API

Writing in its blog on 1 March, Bing announced that its Bing Entity Search API became available to users in a variety of international markets including in the US, France, Germany, and the UK.

The API is designed to offer rich and contextual information about people, places, and businesses to any app, blog, or website — working in a similar fashion to Google’s Knowledge Graph Search API.

This means that developers and creators can now programmatically pull in Bing Entity Search structured data onto their platforms and tools.

Bing’s Entity Search is used by millions of users every day on platforms such as Xbox, Microsoft Office, and Skype.

Developers can get 100 transactions per second through the API, with the standard pricing being found at $3 per 1,000 transactions.

Action Point

You can get a 90 day free trial and preview and experiment the API’s output here.

Yandex 301 Replaces The Host

Writing in a webmaster update (non-English), Yandex announced that it had now replaced the Host directive for site migrations with HTTP-301 for permanent redirects, which offers better support.

New instructions for moving to a new address or protocol now look like this:

  • Add 301 redirects from specific pages on your old site on similar pages of the new one
  • Use the tool in Yandex.Webmaster “Site Relocation” to get the robot’s attention to the changes made as soon as possible.

Both migration schemes still work for the time being. If a site has already started moving according to the Host directive, developers may continue, however, they must also install corresponding redirects.

It is advised that those planning on a move should use the new scheme. More information can be sought in the Yandex Site Migration documentation (English).

Action Point

The Yandex website migration process is now similar to Google – there is a “Relocate Site” option in Yandex Webmaster tools for you to use similar to that of the “Change of Address” option within Google Search Console.

Yandex Introduces Regional Categories for Russia

Yandex.Catalogue regional submissions have now been moved to Yandex.Webmaster , while site owners can now submit their websites to specific regions within Russia.

All requests for websites that are to be added to regions must go through a moderation process, and businesses outside of Russia may request to be entered into a region such as Russia (which is generic), or a sublocality, if it is relevant. Sites may also submit to multiple regions within the country.

The change leads us to pose a number of questions, including which georeferencing would work best for regional ranking of a site in several cities. There could be three options for this:

  1. Regions indicated in the Yandex.Catalogue
  2. Regions indicated in Yandex.Directory
  3. Direction above the region in Yandex.Webmaster

Luckily, this is one question that has already been asked by a webmaster, and a Yandex employee responded (English translation): “There are three ways to submit regions where you operate, however, only the Directory submission would show a business on Yandex.Maps as well.”

Action Point

If you have cross-border sales, check your Yandex.Webmaster profile to ensure you have all your delivery regions are assigned. You will need to submit a URL that contains a specific affiliation in Russia, which can include information such as delivery details.

Assign the Regions on Yandex.Webmaster.

Subdomain vs. Subdirectory Debate Rages on (Again)

The fire that was lit in the January technical SEO newsletter still has a few embers lying around from when we discussed subdomains vs. subdirectories.

Discussing the issue on Twitter, Rand Fishkin found 14 case studies that showed successful results of moving from a subdomain to a subdirectory.

We would also like to know if any companies have had negative experiences of moving from a subdomain to a subdirectory when following migration best practices.

Action Point

Our recommendation is the same as in January. Webmasters should consider using a reverse proxy to migrate important parts of a website.

Google Tests Objective Answers Without Organic Results

During a short window, Google conducted an experiment where it showed only a single knowledge result for specific questions, such as:

  • What time is it in Los Angeles?
  • What is 4×6?
  • What is 10lbs in kgs?

This meant that when users searched for questions such as “what is the time in Los Angeles”, the search engine omitted to display organic results in their entirety.

Google stated that it carried out the experiment to speed up load time due to the fact that searchers “rarely use full search results,” for such questions.

It also mentioned that it provided a “show all results” button for those that did require further information.

The test no longer appears to be active, although some sites, such as time.is, may have experienced a considerable drop in click through rates.

Action Point

No action required.

John Mueller’s Ask Me Anything – The Golden Egg

Kudos to John Mueller for providing one of the most insightful few hours of SEO chat in the past five years for when he conducted an Ask Me Anything (AMA) interview on Reddit.

Some of the main topics he covered included:

  • Responsive design
  • Spambots & search volume
  • SVG Animation
  • HREFLANG
  • Search Console (and APIs)
  • A/B Testing
  • SERPs API (spoiler alert — there won’t be one)
  • Duplication
  • AJAX Crawling
  • Error Pages
  • Disavowing
  • Robots directives
Action Point

Take 20 minutes to read through the AMA Summary by Bill Hartzer or find the original AMA, here.

Additional Reading

Google Algorithm Data Injection Loophole Found by Tom Anthony

Tom Anthony found a really interesting way of hijacking page authority through HREFLANG XML sitemaps. He was awarded $1,337 (“LEET” status) as a bug bounty.

This exploit would be worth considerably more to businesses, as it seemed to guarantee a page one position for competitive terms.

Barry Adams Curates Considerable JavaScript Resource List

Barry Adams started curating an impressive list of JavaScript resources, which is a worthwhile read for any company considering JavaScript heavy frameworks.

JR Oakes Discusses SEO Audits Using Chrome Without Plugins

@JROakes runs through an SEO audit using Chrome in this article. It’s particularly impressive as plugins aren’t used, and features great use of the command line.

Autoplaying to be Muted by Default

Auto playing videos will now be muted by default on Chrome’s next update. It will also be possible to mute an entire website, should a user feel the need to do so.

Google Acquires Xively

Google Makes Intent to Acquire Xively, expanding its reach into Internet of Things. We may soon find more integrations with Google through their hardware and software.

Google to End URL Shortener Service

Google is going to be depreciating the Google URL shortener service (goo.gl). It has given a one year depreciation window.