Welcome to the Q&A Forum

Hi Christy,

Site launched! The e-commerce part is still under development but the basic site has been up a couple months. Masked text doing great! No issues whatsoever on the SEO side. Ranking super high still and load speeds are good. Service workers will be activated in the coming weeks as we build out our food delivery platform. So, I'll mark my question as answered. https://www.88k.com.tw

Site not launched yet but no warnings on any SEO tools. You can run this site through any tests you want and see. https://website-1b14f.firebaseapp.com/

Schema all good and AMP valid. Content coming up next... FYI this is not a public site and content will change as we test new designs and functionality.

I agree with Nigel and Google. You will be spliting link juice between two domains rather than building a single domain using langauge subfolders. The language of the page should be documented in the html tag. https://www.w3.org/International/questions/qa-html-language-declarations This is the way to go.

My first thought would be to have an IVR on your phone system with a "Press 1 if you're located in the XXX metro area, press 2 if you're located outside this area." Then if callers press 2, you have have an apologetic message that states your business qualifications.

If that's not possible for some reason, you could use redirects based on geolocation with services from Maxmind. Alternatively, paid accounts on Cloudflare allow geoip rules that could redirect visitors from outside your IP range to a "sorry" page.

My opinion on this is to modify your business processes to accommodate out-of-area customers.

Thank for your thoughts. You're right that I can't find a single article on this anywhere, but I've never been conservative when I comes to SEO. I'm always looking to see what's possible. I concluded that since unsupported browsers will simply display the original text without the mask-image (Firefox/Opera), I'm going to assume google search bots won't care about the image mask either.

On the SEO side, this method shaves 3 to 5 seconds off load times, so that can't be bad. The effects are amazing, even on Chinese fonts. I'll report back after launch and post here.

In my attempt to decrease page load times, I ditched my custom fonts for google fonts. I figured out how to apply CSS mask-image to make this blazing fast loading google font have a chalk texture, which was an awesome improvement over the 3-5 second load times for those locally hosted web fonts.

I've read that Google penalize a site for poor contrast ratios between the background and text, but do search engines go by CSS or do they somehow compare the actual rendered site as an image? Using CSS mask-image to give my text that chalk appearance does produce minor transparent patches in text.

So have I saved 3 seconds on page load just to have search engines knock points off for funky text issues? All input welcome. The temporary site is here. https://website-1b14f.firebaseapp.com/

Kevin

I think the community needs to look at the AMP project roadmap and recognize that AMP HTML isn't only for blogs and news sites. Google is very active in promoting AMP standards, and now it's possible to create a full e-commerce website using 100% AMP HTML. And because analytics tells us 86% of visitors find us on a mobile device, we've committed to building out a 100% AMP site. I'll post back after the launch and report on our results.

Redirects usually don't pass all link juice to the target URL but may in time. You won't get accurate reports on the www version of your site because it's redirecting to the non-www URL, so use this primary URL when running reports with SEO tools. Regarding the number of links, you may find that people generally link to the www version just by habit, and that's why I personally still use this as my primary for most projects.

Another thing to do is check Google WMT to make sure both www and non-www properties are listed. You'll then be able to select the preferred domain in site settings.

You didn't ask but I'd strongly suggest you go over some PageSpeed and YSlow recommendation. It's very slow and JS heavy, and this is simply bad for all the reasons you already know.

We are moving to a custom AMP HTML site now and I'm doing most of the work. The whole site doesn't have to be valid AMP HTML, just offer an AMP version of your pages with a rel attribute. For content heavy pages, we think his is a good thing. Over 90% of our traffic is viewed on a mobile device, so we agree it won't hurt to go this way. Benefits include near instant page rendering on mobile devices, top spot on mobile SERP carousel, and free caching on Google CDN. Nothing wrong with all that.

AMP HTML currently doesn't suppoort input fields of any kind. No search, no forms, no nothing. No external CSS or JS, and fonts are very limited, especially for Asian fonts due to the JS restrictions. So you can't create an e-commerce site with valid AMP HTML. But your e-coomerce pages don't have to be, but you'll need to get creative. There is iframe support, so carts like Ecwid might work. Will update as we get closer to that part.

We've started with Skeleton framework and cleaned it up a bit, but we're finding the CSS rules a bit restrictive, like !impotant is not allowed. So even just the Bootstrap grid will need a lot of scubbing to validate. And you must validate or there's no point.

So far, we're ok. Love to hear what others have learned working with AMP.

You probably have other redirects in your .htaccess and possibly in your website code. The order of your rewrites is also important. Publish your Apache config and I'll take a look.

FYI, there are better resources for technical issue than MOZ. Most here are not developers/IT specialists; we're more like SEO strategists and business managers.

Just food for thought, another option is to host a Wordpress multi-site or even two separate versions of Wordpress, one for each language. I find this less complicated when it comes to plugin and template compatibility, plus you can control access a bit better.

Avoid using Javascript to translate text.

Avoid putting content in multiple languages on a single page.

Do link each page in one language to the translated page to avoid 404 errors. If your language selector automatically directs users from an www.site.com/en to www.site.com/es domain, make sure your URLs for translated pages match or you'll get a lot of 404 errors. This will hurt you a great deal.

Hi Mark,

I do suggest moving to a provider like DigitalOcean that offers great hardware that costs less than some shared hosting solutions. With DO, you have complete control. I wouldn't suggest PHP redirects with Wordpress extensions unless you absolutely must. I moved to Nginx years ago, but with Apache I know you can still have an .htaccess file in your website home directory, and rewrites should work these just fine.

Are you using an particular flavor of shopping cart or a custom e-commerce solution? You might want to try searching for plugins/extensions that can do this for you. Google Tag Manager is another way to do this.

Yes, server redirects are necessary. Try these solutions to see which one works for you:

http://ralphvanderpauw.com/seo/how-to-301-redirect-a-trailing-slash-in-htaccess/

http://enarion.net/web/htaccess/trailing-slash/

You might want to consider moving to Nginx. You'll notice amazing speed and stability improvement with Nginx, Redis Session Cache, Memcached, OpCache, Ngx_pagespeed, and Magento Cache Storage Management. I can help much more with Nginx redirects and conf files--I gave up Apache years ago. Sorry I couldn't be of more help.

Patrick, everybody should definitely check external links. We had hundreds of linking domains to our site, but after the switch to HTTPS we lost them all. Google now shows only one single linking domain in GA. Our rankings for performing key words have dropped six to ten spots from our page one 1-4 rankings in less than a month. Indexed pages went from nearly 300 to only six in the same time. Structured data records show a drop from nearly 220 to only 17.

Since Google recommends this, we'll give this another month or two, but if this Google recommendation switch to HTTPS doesn't start showing positive results, we'll surely switch back. My recommendation for now is to only use HTTPS for pages that transmit sensitive data.

Marcel, wait until Google gives the countdown notice to HTTPS like they did with mobile before you migrate your whole site.

Background: I run a Magento 1.9.2/WordPress 4.2.3 multilingual site on LEMP with fine-tuned Google PageSpeed module, Magento FPC on dedicated Memcached instance, and Redis session caching. I know what I'm doing here, but if anyone has suggestions please do share. https://www.88k.com.tw

Check out SPDY and see if your web server supports it.

Old post but wanted to add a little. Been with Cloudflare for years but want to switch, so I searched Moz to learn what might happen to site ranking. I'm now neutral about Cloudflare and actually turned off extra services just to use their DNS for a few sites. While Cloudflare offers many optimization services and DDoS protection, many of them break Magento and Wordpress functionality. Implementing caching and Google PageSpeed with Nginx or Apache shows much better performance, especially in Asia. Worse though, Cloudflare doesn't allow you to use your own SSL on any plan under $200/moth. So if you have your own EV SSL, plan on dishing out for this. If you're a Pagespeed/YSlow nut, they you'll really be upset to learn Cloudflare adds a security cookie to everything, so there is no way to have serve content on a cookie-free domain when using Cloudflare DNS. Customer service is also slow and mostly automated, so don't expect great service on unpaid plans.

Hi Thomas,

While just a bit off-topic, I was curious about your Hubspot experience. While I'm seeing great performance from my current application set up, I've been eyeing Hubspot for some time now, but I just can't justify the cost. Does Hubspot allow you to do anything to increase conversions that couldn't be done with Wordpress? Any thoughts, please do share.

Thanks!

Hi Joshua,

Glad I could be of some help. Sorry to hear your dev team dismissed the idea of Fishpig's integration. Using this integration method, you wouldnt have to pay for a custom theme development to match your Magento site--in fact your Wordpress content would appear inside Magento resulting in a big performance boost. Imagine visitors navigating from your store to your blog and back loading all those resources. That's a few second delay at least depending on your set up. If your dev isn't in-house, consider the motivation not to use Fishpig integration, like profits on a custom Wordpress theme and maintenance of two sites Instead of just one.

If you have any othe questions about Magento performance, feel free to ping me. FYI, just finished New Relic integration and seeing sub-second page loads on my Magento/WP BBQ site. If you like how this site performs on both mobile and desktop, I can point you in this direction.

Cheers!

Magento Wordpress installs are my specialty. From a marketing perspective, a blog will definitely be helpful for your visitors seeking relevant info. Most clients I set up tend to neglect their blog, so make sure to approach this with a plan. From the feedback I've collected from customers and clients, a consistent brand image tends to build more trust.

From a technical perspective, there are several ways to integrate Wordpress and Magetno. Fishpig http://fishpig.co.uk/magento/wordpress-integration/ will take a lot of work out of the project if you select to run Wordpress "inside" Magento. This extension basically uses your Magento template and puts Wordpress content into Magento, but you administer Wordpress through the traditional back end. If you have multiple languages/blogs, this gets a bit tricky. You can visit www.88k.com.tw for my site example running a Wordpress multi-site inside a multi-store Magento instance on a custom LEMP stack.

Anyway, to answer your question, a blog would be good if done right. Be careful not to cannibalize other site pages by adding the same keywords, and also make sure to link related blog posts to products. I personally agree with you--the purpose of your blog should be to inform visitors, not to sell. However, consistency is good, too. If you want to sell on your blog, there's nothing wrong with that, as long as you have a good plan.

After you complete your profile, you will need to verify your page and claim your vanity URL. Then you should be good.

It would surely be helpful to post the HTML so we can see what's going on. W3Schools has a great tutorial here http://www.w3schools.com/svg/default.asp. If your title tag is properly embedded with the svg element, everything will be fine. From what you posted here, it seems you're afraid search engines will confuse the svg title with the page title. That won't happen if your HTML is correct. Post it here and I'll take a look.

Hi Justin. Most search engines take into account your physical location when displaying results. If you are physically in NZ and search for your high ranking keywords, you will more than likely get the NZ site. If you have access to a server in the USA and run the same search, chances are you'll get the .com site. Likewise, if you're in AU then the .au site "should" show higher in the search results. Google has made it very clear here https://www.youtube.com/watch?v=8ce9jv91beQ that just because a site ranks well in one language doesn't mean it'll rank well for a translated site or language-specific subdomain.

Other than that, your biggest issue is that you have very little content for search engines to consider. Google used to recommend about 250 words of relevant content, but these days I'm seeing longer articles rank much better. Your site performance could also be better, so I recommend a read on YSlow scores and how certain front-end optimizations might speed up your site.

But you should have already know what we were going to tell you. Your psychics should have told you! LOL

Cheers!

To add to Tafser's response, you're YSlow score could be better, currently overall performance score of 67. There is a whole lot you can do to increase performance for both visitors and search engines. Another mention is that your static content is rather thin. I compared the meta keywords you added to your page content, and I imagine if you used Moz On-page grader you'll find that you could do much better.

It's understandable to have many products, but you may want to try moving those product links off your homepage and onto a product category page, and then create content to help your homepage rank better for your selected keywords.

Besides MOZ local, you'll need to submit sitemaps and feeds to GWT and BWT. Good luck!

If I understand correctly, you're ranking well for a page and related keywords but you'll be adding new "news" content related to the title page topic. Sounds good to me. I'd make sure to note my changefreq in the sitemap to hourly or daily. While I'd expect you to continue to rank well for your static content, there are other factors to consider in a "redesigned" page. For example, how your new HTML code is structured; the order of content displayed to search engines (source code); how much of the original content you keep above the fold; relevance of your news to your current keyword(s); the number and quality of links added to this page; change in the amount of traffic to this page; and of course all the other technical and editorial SEO factors. Only thought about this for a few minutes but I'll post back if I think of anything else.

I'm a little confused about the redirect you want to implement. Are you unable to just replace the original page or use the original page URL? Note that permanent redirects tend not to pass 100% of your link juice http://moz.com/learn/seo/redirection. Try to do that, but if you can't then the redirect is likely your best option.

Nice link Mark. News to me, really. But the fact that Schema.org and HTML5 both have author identification methods shows that it may be used by other search engines and/or services. And the followup article to your link there is "Google Authorship May Be Dead, But Author Rank Is Not." http://searchengineland.com/google-authorship-dead-author-rank-202254

But darn, man! All that time wasted getting authorship to work back then. Google's authorship verification process was indeed grueling.

Two schools on that one. They may not be hurting your business now, so you can forget about them. That's only until you can't. If they continue rip off your work, they may take from you in the future--ad revenue, traffic stats, e-commerce, news reports, whatever you're doing--that's all money. If I had time to fill out the form, I'd do it.

First thing to do is insert authorship markup and check that google recognizes you as an author of the site you're posting to. There is something to say for original content, and Google knows. If your content goes up first and is indexed first by Google, chance are you're going to rank better than the scrapper sites.

If these sites really bother you, you can submit a Copyright Removal form here https://www.google.com/webmasters/tools/dmca-notice, but a legal order to remove the content would be better (acted upon faster). Filing copyright infringement reports for eBay listings was very effective for me, but my experience with Google is limited. Let us know if you do file and how the process goes.

Generally speaking, it's actually pretty good that site are linking to your posts. If you are extremely uncomfortable with any particular site's backlink(s), you can use the GWT Disavow tool https://support.google.com/webmasters/answer/2648487/?hl=en&authuser=1

Good luck, and let us know what you do.

You'll want the shortest, most technically feasible URL structure based on whatever website platform you're using. Subfolders like http://example.com/es/ and http://example.com/en/ are ideal, and there is a lot of talk about subdomain vs. subfolder on MOZ. Keep in mind you also have a http://example.com landing page, so you'll either need to redirect users or have visitors select a language on this home page. There is pretty thorough documentation here https://support.google.com/webmasters/answer/189077?hl=en on how to use the hreflang attribute in each case.

I leave my default language as the base URL and put additional languages in subfolders. Statistically, I tend to rank higher for keywords in my default language than for the translated keywords in additional languages. You might want to target the market with the most traffic or conversions (whatever metric you prioritize) with the default URL and then add additional languages as subfolders, preferably without hyphens or underscores in the locale code, i.e. en-us, en-uk, etc. This is more for your visitors and not a particular ranking factor, but shorter domains are preferable.

Thanks for the input, Gianluca. I'm actually a dev myself and am about to abandon all these open-source platforms for a static site, but I do need a blog. It's a terrible case really, but there are absolutely no platforms built first on top of a multi-language framework. You either need a plugin, a module, a different store view, etc. Unbelievably short sided in my opinion, but that's where we're at. New open-source packages like Grav and PageKit don't support multiple languages, either. Just baffling! And SEO standards are flat out being ignored even most of the hosted platforms. WTF?

FYI, Odoo is an amazing piece of software, but their new CMS is far from ready. There is no URL redirect feature; no structured data; unfriendly URLs with post IDs and underscores; dev unfriendly--you'll really need to dive into the code to make changes, override, or customizations. On performance, Odoo runs on it's own openerp server through a Nginx/Apache proxy. All-in-all, the language support is ok and the design features are pretty cool, but these very capable developers almost completely ignored performance and SEO best practices. In my opinion, it will take years for this platform to compete with Magento or Wordpress as an viable SMB solution. There just isn't an easy migration path to Odoo, and it's too much work for no real advantage over the other packages.

Seems like developing a multi-lingual, multi-currency, e-commerce platform that follows current SEO recommendations would be a huge step for mankind. Still looking...

I'm pretty well rounded when it comes to SEO, but I'm just frazzled when it comes to YSlow and PageSpeed Insights. Of course, individual factors are important to site performance, but it has become increasingly difficult to recommend open-source and hosted platforms that don't pass muster on many of the performance standards being tested. For example, entity tags, expires headers, and cookie-free domains are nearly impossible to set with hosted platforms, and none of the major open-source CMS like WordPress, Joomla, Magento, Odoo, etc., consistently put javascript at the bottom or make "fewer" HTTP requests.

Mobile is now king, so quite a few people (including myself) need to "mobilize" their website by late April or risk dropping in mobile search rankings. Nearly all my clients run multi-lingual e-commerce websites, so that really limits options but makes it that much more important to keep current with Google's SEO recommendations.

What platforms perform best taking into account any correlation with YSlow scores/PageSpeed Insights to high floating sites on SERPs? Would one spend the money to "fix" their current platform that has worked very well to date or switch to a mobile-ready platform?

To prevent the URL rewrites by PageSpeed, follow the documentation here https://developers.google.com/speed/pagespeed/module/restricting_urls. You can add the following to your config file:

ModPagespeedDisallow "http://example.com/.jpg"

or

ModPagespeedDisallow ".jpg"

It would sure help to get more specifics on the geographical area you intend to serve and the amount of traffic you expect. Where is your app/db server(s) and what kind of hardware will you be running? What kind of data will you be serving?

Generally speaking, CDNs are built to scale to high traffic loads with low latency and high availability. A VPS would need to be configured for the type of data being served and would need to be constantly and consistently monitored for performance, whereas a CDN could be considered more of a set-it-and-forget-it service, granted you've built your application to utilize it properly.

My best answer for you is to go with a CDN. The only case I could see you wanting to use a VPS is if it's cheaper and you server load is known and constant.

I would strongly recommend Cloudflare to address this type of problem. They have massive data on malicious sources and offer tools to mitigate attacks like you're facing.

The fist thing to look at is to see if the template/plugin/page is published. If so, I'd check if your CloudFlare cache. Read up on how to purge the cache here https://support.cloudflare.com/hc/en-us/articles/200169246. This is most likely your problem. If this doesn't help, I'd use dev mode on Cloudflare and start checking caching plugins and memcache.

It's also helpful to rename your new images, i.e., Web-Header-New.jpg to Web-Header-New-2.jpg, so you're not overwriting the previous image that was already cached. You'd have to wait for the cache to expire for a new image to show if the names are the same.

It's hard to pinpoint the problem without seeing the site but there are a few things you can do. Keep in mind that pages get ranked by popularity and relevance, so use MOZ to see what keywords this page is ranking highly for, if any.

Some templates for major CMS load content blocks in a particular order. Check the raw source for you product page in question and examine the code to see what information comes at the top. You could also fetch your page as Google bot and see what Google sees https://support.google.com/webmasters/topic/6065797?hl=en&ref_topic=4617736.

About metadata, check out the kind of metadata Google likes here  https://support.google.com/webmasters/answer/79812?hl=en.

Now you can learn more about structured data on Schema.org. Rich snippets are a great way to let search engines know what you want potential visitors to see in SERPs.

How do you know the main issue is over-optimized anchor text? Did you get a manual action notification in GWT or is this effect your assumption? This would be an important distinction because you'd get some good info from the GWT team about what happened to your site. Some sites take a temporary hit after new algorithm updates, so if this just happened I'd give it a little time before making drastic changes to your site.

It's unclear whether it's your site that's hosting the links or if other site(s) are linking to yours. Follow Googles guidelines about linking https://support.google.com/webmasters/answer/2648487?hl=en and read up on how and why to disavow backlinks here https://support.google.com/webmasters/answer/2648487?hl=en. Being honest to yourself is the best policy here, so if you're taking or paying money for these links, you or your associates should use the nofollow attribute according to Google's guidelines. You can learn more about anchor text here on MOZ http://moz.com/learn/seo/anchor-text. It's believed that links with the nofollow attribute won't pass PageRank, so the anchor text of such links won't hurt or help with search engines but may be valuable to your readers. More about how to use the nofollow attribute here https://support.google.com/webmasters/answer/96569?hl=en.

Now if you actually got notified of a manual action, then you'll need to submit a reconsideration request in GWT. We don't have enough information here on how Google will deal with your site on the next crawl, but I would recommend a consistent, site-wide approach to SEO rather than just changing a few links you think might be hurting you.

Hello Minted representative. I'm in the SF bay area for another week then headed back out to Asia. Before you go hire a $100K salaried employee, check out my Linkedin profile and call me in for a chat. You'll be happy you did. https://www.linkedin.com/in/kwoolf

I was just going to comment on Don's post about your issue but it seems you've cleared it up. As I understand here, you will be keeping your domain name, just building a new site on a new host. Good, good. You're link juice will keep flowing if you keep your current domain, so relax.

Since you're starting from scratch, you should consider using a platform or static HTML site that you will have complete control over. Those hosted SaaS website builders are not what you want if you're here on MOZ looking to rank well organically. I've had really great results with Magento (not Magento Go), so I can safely recommend it. Odoo, WordPress, Joomla, and other popular open-source platforms all can get you want you need in terms of technical SEO features, but you'll need plugins and modules that may affect performance. Due diligence will pay off here.

Your next step is to do your own site index, noting all the page URLs now and either replicate this structure or set up redirects to avoid a hit in rankings. Since your site seems small and your referring links all point to your homepage, I'd say you might be able to skip this step without a major hit. If you can scrape your current site to save for reference, now is a good time for that.

For DNS, I suggest CloudFlare. Read up and you'll see why. Good luck and post back if you have any more questions about this migration. Cheers!

Without being able to view stats on the domains in question, it is impossible to give you the best advice. Generally speaking, subfolders get better results than subdomains, but there are many other factors to consider.

If you can't move to a subfolder structure, then skip this issue and start down the list of other technical optimization best practices. With time and traffic, I'm sure you'll gain in your organic rankings for selected keywords. In my experience, a domain name is but a small factor of the overall picture. The issues you'll have with these subdomains are more of a marketing cost than an SEO cost.

In my experience, A/B testing for content placement yields different results for different products/languages/seasons. So many variables, I'd just use those resources for more valuable content testing.

Are your buyer guides downloadable or HTML? For downloadable guides, I'd have a thin content block above the fold on category pages right above the products. Another effective spot would be a taller block above the fold in the sidebar on the same category pages. If your guides are a really big deal, have a content block on your homepage introducing these new guides and then link internally on where to get/view them.

Whether you're hosting downloadable or HTML guides, I'd still have links to them on your blog. But without more details on your site, I can only offer this general advice but feel free to update us.

Haven't read anything or experimented with using canonical link elements to indicate your preferred domain. You should have server redirects in your Apache/Nginx config file for this and also set the preferred domain in GWT to really take care of that issue.

Yes, Google will certainly decide for you if you don't have other indicators on which site you want to rank for select keywords. I suppose you'll need to decide which site to keep, but if you want both then you've got a lot of work ahead of you. Ideally, you should be ranking each product page on each site for unique keywords rather than duplicating efforts and wasting resources on trying to rank the same product page on two different sites for the same keywords. Not sure why anyone would want to do that for any other reason than testing.

Anyway, my strong advice is pick a site and then merge the other site into it. If you're using Magento or WooCommerce or some other mainstream e-com platform, this should be rather easy, if not just time consuming. If they are different platforms, then I recommend Magento or a custom solution for best technical SEO features. If you can share more details about your job, I'd be glad to help.

My question to you is why shut down the niche site if it's ranking better than the main site? There is surely some risk of taking a short-term hit when merging these two sites, but I'm confident in suggesting that combining the traffic from the two sites will eventually earn similar organic ranking of your niche site pages for the same keywords.

Yes, 301 redirect your pages to the new domain if possible. If you run into technical challenges with this, use MOZ to see which pages rank the highest and prioritize those redirects to the main site domain.

You stated "The niche site has about 70K product pages that have the same content (except for navigation links which are similar, but not dupliated). " I understand this to mean you found 70k products with the same/similar content on both the main site and niche site. If you simply ran a report that told you you had 70k product pages with duplicate content, I'd make sure to check your URL structure for www subdomain, trailing slashes, and the variation of both, i.e., www.domain.com/product, www.domain.com/product/, domain.com/product, domain.com/product/. These are technically all different URLs, so double check that and reply.

If you really have the same pages, obviously you'll need to fix that if combining the two sites. If you don't combine them, cross-domain canonicalization is an option (http://googlewebmastercentral.blogspot.com/2009/12/handling-legitimate-cross-domain.html and https://support.google.com/webmasters/answer/1716747?hl=en).

You really can't/shouldn't control who links to you and how they link, but if Google's fancy algorithm finds something in common with the way people link to your site, it might do more harm than good. You said "a few people," so this really won't be an issue for you. If you're speaking at an industry conference and suggest to 1000s of folks how to link to your site, then I would carefully follow Richard's advice and note Oliver's tips.

The only addition to Richard's comments would be to keep domains short and add keywords as appropriate. A single "About Us" page, XYZ.com/about-us, would be preferable to four different pages, i.e., XYZ.com/auto-detailing/about-us, XYZ.com/auto-sales/about-us, etc.

You might want to check Google page policy about having multiple pages for the same business. That might have consequences.

The biggest concern I would have as a business here is site visitors going to the wrong physical location because the website wasn't clear. That would really upset visitors, so this company should have SOPs in place in case this happens to minimize complaints. You'd be right to emphasize this in your site design suggestions.

Google has stated they are better now at relating subdomain content to the TLD domain, but you're probably still better off using a subfolder. If you do go with a subdomain, make sure to link them in your GA code for better reporting.

There are so many languages spoken in Malaysia that this domain issue is really not going to help visitors on your site. If you're going for consistency, you'd need subfolders or subdomains for all supported languages. If you're only using English, then I wouldn't even consider this change. Simply use www.domainname.com.au/malaysia/ to host content relevant to this market until you can get your TLD. This would send the strongest signal to both search engines and site visitors that your content is targeted for Malaysia.

FYI, "Google prefers microdata for web content." This was taken directly from their site at https://support.google.com/webmasters/answer/3069489?hl=en

"Google is in the process of adding JSON-LD support to more markup-powered features. So far, JSON-LD is supported for all Knowledge Graph features, sitelink search boxes, and Event Rich Snippets; Google recommends the use of JSON-LD for those features. For the remaining Rich Snippets types and breadcrumbs, Google recommends the use of microdata or RDFa." From https://developers.google.com/structured-data/schema-org

I was curious about this, too. Spent quite a bit of time looking through schema.org and other examples, and it seems the price property must be included. We can't just use maxPrice and minPrice to specify a range.

I replaced the minPrice property with the price property and my testing worked just fine. Depending on your business model, you could consider using price for either maxPrice or minPrice, depending on which price you'd like search engines to show your visitors. Since my prices only increase when adding variations, i.e., extra sauce, pre-cut, etc., I want visitors to see my lowest base price on search results. If for some reason you want visitors to see your high price on search results and then awe them with a lower price when they hit your site, maybe substitute the price property for your maxPrice.

Hope that helps!