Most WordPress sites in the Philippines do not fail search because they forgot to install Yoast. They fail because the theme outputs thin templates, the plugin stack fights itself, staging leftovers stay noindexed, WooCommerce archives leak duplicates, and nobody checks what Google can actually fetch. The CMS is capable. The install is often treated as a brochure that happens to have a blog.
WordPress SEO is still technical SEO plus page quality, running on a platform that makes it easy to add weight. Every extra slider, page builder module, and “SEO pack” is another chance to hide the commercial copy, slow the first paint, or index a URL that should never have been public. The work is to make the pages that should rank crawlable, unique, fast enough, and worth the click.
Start With What Google Gets, Not What The Dashboard Shows
The WordPress admin is not the site Google sees. Page builders, cached HTML, CDN rules, and security plugins can all serve a different document to a crawler than they serve to you while logged in. If Search Console coverage looks wrong, do not start by rewriting meta titles. Fetch the live URL as Google, compare it to the public HTML, and look for noindex, canonicals that point at the wrong template, and content that never appears without JavaScript.
A surprising number of Philippine SMEs launch from a staging clone and leave the robots meta, a coming-soon plugin, or a password wall in place. The site looks live to the owner because they are logged in. Google sees a block. The inverse also happens: staging subdomains get indexed because someone turned indexing on to “test SEO”, then never 301ed the temporary URLs.
XML sitemaps from an SEO plugin are only useful if they list the URLs you want and exclude the rest. Default WordPress sitemaps will happily include author archives, date archives, tag pages, and attachment URLs. If those templates are thin, you are asking Google to spend crawl budget on pages that cannot convert. Turn off the archives you do not use. Keep the sitemap aligned with the commercial IA.
Search Console remains the source of truth. Rank trackers will show whatever keywords you typed in. Coverage, inspected URLs, and the pages that already earn impressions tell you whether WordPress is helping or leaking.
Themes And Page Builders Decide More Than Plugins
A premium theme with a homepage slider, five fonts, and a bundled plugin pack can bury the H1, duplicate H1s across modules, and ship image markup without dimensions. Page builders make that worse when every landing page is a unique snowflake with the same three sentences in different sections. Google does not need a unique animation. It needs a unique, readable offer.
Choose a theme that outputs clean heading hierarchy, uses semantic HTML, and does not inject sitewide keyword text in the footer. Then stop fighting it with six overlay plugins. If the builder cannot produce a service page that reads as one document, with one H1, supporting headings, and an obvious enquiry path, the SEO plugin cannot rescue it.
Philippine web vendors often hand over Elementor or similar because it is fast to demo. That is a delivery choice, not an SEO strategy. Builder HTML can rank. It also tends to bloat CSS, delay LCP, and make template-level changes painful. If you must use a builder, lock a small set of templates for services, locations, and blog posts. Do not let every staffer invent a new layout. Inconsistency is how internal links, schema, and conversion paths drift.
Child themes exist so you can change behaviour without losing it on the next parent update. Custom SEO logic in the parent theme, or in functions.php of a theme you cannot update, is a future outage. When the site breaks after an update, the first casualty is often the header, the canonical, or the robots output.
Plugins Help Until They Become The Problem
One well-configured SEO plugin is enough for titles, canonicals, sitemaps, and basic schema. Two SEO plugins is a conflict. Rank Math plus Yoast plus a “local SEO” add-on plus a separate schema plugin is how you get four JSON-LD graphs and a canonical that nobody trusts.
Security, caching, image, and optimisation plugins each touch HTML. A cache that serves logged-out users a noindex header copied from a preview session will quietly deindex the site. A minify plugin that breaks JSON-LD or strips heading tags will not show up in the visual editor. An image plugin that lazy-loads the LCP image without a noscript or proper priority hint will hurt Core Web Vitals on mobile, which is where most Philippine traffic still arrives.
Audit the plugin list by job. Caching: one system. SEO: one system. Images: one system. Forms: one system. Everything else needs a reason to exist. Unused page-builder add-ons and abandoned “click to tweet” plugins are attack surface and HTML noise.
WooCommerce adds its own SEO surface: product, category, tag, and filter URLs. Faceted navigation can create thousands of indexable combinations. If the shop is a catalogue, decide which category templates deserve unique copy and which filter states should stay noindexed or canonicalised back to the parent. Infinite filtered URLs are not a growth strategy. They are index bloat.
Indexation, Canonicals, And The WordPress Defaults That Leak
WordPress wants to publish. That is a feature for a newsroom. It is a hazard for a service business that never meant to index /author/admin, /category/uncategorized, or every image attachment as a standalone page. Disable unused archive types. Redirect attachment URLs to the parent. Noindex thank-you pages, account pages, cart, checkout, and internal search results.
Canonicals must match the URL you want in search. Trailing slashes, HTTP versus HTTPS, www versus non-www, and pagination parameters all create duplicates if the plugin, the server, and the CDN disagree. Pick one host, force HTTPS, and make the plugin canonical match the address bar. Mixed signals here waste months of content work.
Pagination on blogs and product listings should be crawlable without creating a second homepage. Rel next/prev is no longer a ranking factor, but unique titles, a clear canonical policy, and internal links from the hub still matter. Do not noindex page 2 if it is the only path to older posts you still want found. Do not let page 2 rank for the primary category term either. That is a template and title problem.
Hreflang only belongs on WordPress if you actually have language or market versions. A plugin that outputs hreflang for a single Filipino-English mashup page does not make you international. It makes the HTML noisier. If you serve PH and another country, the language versions need their own URLs, their own content, and a real return tag. Otherwise skip it.
Content Architecture On A WordPress Site That Has To Sell
The blog is optional. The service, location, and product templates are not. Philippine WordPress sites often have 80 posts about “tips” and one generic Services page. Buyers search for the job, the city, and the category. Those queries need pages that explain the offer, the proof, the area served, and the next step.
Use pages or custom post types for commercial work, not blog posts pretending to be services. A post dated 2023 that ranks for a core service term looks abandoned. A service page can still have a last-updated date and supporting articles linked from it. The commercial URL should be stable.
Internal links should follow the sale, not the menu. From a blog post, send people to the service they would actually buy. From a service page, send people to proof, FAQs, and related services that do not cannibalise. From location pages, send people to the unique proof for that area, not a cloned paragraph with the city name swapped. WordPress menus are navigation. They are not a substitute for in-body links.
Heading structure is still a quality signal and a usability one. One H1 that matches the page job. H2s that split real questions. Do not use headings as design styling for buttons. Screen readers and crawlers both suffer when the builder uses H2 for every callout.
Media needs filenames, compression, width and height, and alt text that describes the image for humans. Stock photos of handshakes do not need keyword-stuffed alts. Project photos, product shots, and location photos do, because they carry information.
Speed, Hosting, And The Philippine Connection
Core Web Vitals on WordPress in 2026 are mostly template problems: hero images that are too heavy, fonts loaded from too many origins, third-party scripts for chat and pixels, and builders that ship unused CSS. Hosting matters, but throwing a site on a cheap shared box and then stacking optimisation plugins is backwards. Start with hosting that can serve PHP and a cache close to the audience, then reduce what the theme requests.
Mobile is the default. If the tap targets sit under a sticky bar, if the form is below three screens of slider, or if the click-to-call number is only in an image, you will rank into a bounce. Test on a mid-range Android on a real connection, not only on a desktop emulator.
Chat widgets, Facebook pixels, tag managers, and heatmaps all compete with LCP and INP. Each one needs a commercial reason. If the site cannot convert without a chat bubble, fix the form and the phone number first. Third-party scripts are not a substitute for a visible enquiry path.
When developers break WordPress SEO, they often do it in functions.php, in a “defer all JavaScript” snippet, or in a theme update that drops image dimensions and changes permalinks. Performance and technical SEO share the HTML. Treat them as one change list, with staging, a crawl before launch, and a crawl after.
Launch, Migration, And The Handover Most Sites Skip
A new WordPress site should have a launch checklist that is not only “point the domain”. Permalinks set to a sensible post-name structure before content is written, not after. Redirects from the old URLs, including http, www, and common file extensions. Analytics and Search Console on the production host. Indexing allowed. Staging blocked. Forms tested on mobile. 404 template that points to useful pages. A crawl of the production domain before anyone runs ads against it.
Migrations are where WordPress SEO is most often destroyed. Changing permalinks, moving from a page builder to another, or cloning the site to a new host without a redirect map will drop the URLs that already had impressions. Export the old URL list from Search Console and a crawler. Map every URL that earned clicks. Keep the path when you can. When you cannot, 301 to the closest equivalent, not to the homepage.
Handover documents should name the SEO plugin, the indexation rules, the people who can publish, and the things that must not be “cleaned up”. We still see well-meaning staff delete “old” pages that were the only ranking URLs, or install a second cache plugin because a tutorial said so. WordPress is easy to edit. That is why it needs rules.
Maintenance is part of SEO. PHP, WordPress core, theme, and plugin updates are security work and HTML work. Unpatched sites get injected. Injected sites get hacked spam indexed. Then you are doing incident response instead of commercial pages. Schedule updates. Take backups you have restored at least once. Do not update production on a Friday before a campaign weekend without a rollback.
A Practical Way To Work The Platform
Do not buy another plugin until you have inspected coverage, the homepage and top five commercial URLs, the template that generates them, and the form. If those are wrong, a content calendar will add more pages to a leaky site.
Do not judge WordPress SEO by the traffic to the blog. Judge it by organic enquiries from service, location, and product URLs, and by whether Search Console shows those URLs gaining impressions for the terms that match the offer.
If you already publish on WordPress and growth has stalled, the stall is usually one of four things: Google cannot see the right pages, the right pages are thin or duplicated, the site is too slow or broken on mobile, or the enquiry path loses the visitor. Name which one it is. Then change the template, the IA, or the launch settings. The plugin you already have is probably enough.
When you want that diagnosis against the live install, send the production URL, CMS access rules, and the pages that should produce work. We will tell you whether the next step is indexation hygiene, template repair, or commercial rewriting. Get in touch with the domain and the outcome you need from organic search.
