Twelve SEO tips that survived the last five algorithm updates.

The list: The rest of this post expands the five most important ones, because those are the ones that get botched most often. The other seven are straightforward enough that the one-line version is enough.

• Write for the person who clicked, not for the algorithm.
• Update your best existing pages before you publish new ones.
• Use Search Console as your primary keyword tool.
• Match the dominant content format already ranking for the query.
• Add internal links every time you publish.
• Fix your Core Web Vitals before you do anything fancy.
• Earn one good backlink rather than a hundred mediocre ones.
• Use real photography and real screenshots, not stock.
• Build a single deeply-linked topic cluster rather than scattered posts.
• Write a clear author byline and link to a real bio page.
• Delete or merge pages nobody reads.
• Measure organic conversions, not organic traffic.

The single shift that has correlated most strongly with surviving Google updates: writing pages a real human would finish reading. Google trains its quality raters to judge this when it tests its results. Google's own write-up on the Helpful Content update is the cleanest statement of intent.

The practical test: read your draft aloud. If you would not say it to a customer over coffee, do not publish it. The pages that pass that test tend to keep their rankings through updates that punish everything else.

A two-year-old page that ranks at position 6 has more accumulated SEO equity than any new page you could publish next week. A focused refresh, expanded content, updated examples, refreshed publish date, and internal links repointed at it can move the page up within a month or two. The same effort spent on a brand-new page rarely produces anything for a quarter.

The math is simple: it costs less to harvest an existing page's near-miss than to grow a new page from zero.

The fuel for this work is Search Console. It is free, it is built by Google, and it shows you the exact queries where you rank between positions 5 and 15. Those near-miss queries are the highest-leverage opportunities on your entire site. A quarterly hour pulling that report, mapping the queries to existing pages, and writing a fix list will produce more traffic than any third-party tool.

Internal linking is fully under your control and few sites do it well. When you publish a new page, link to it from three to five relevant existing pages. When you update an existing page, audit it for opportunities to link out to your other strong pages.

Moz's primer on internal links covers the underlying mechanics. The short version: internal links pass authority between your pages and tell Google which ones you consider important. Most sites tell Google their most important pages are the ones in the navigation. Internal links from body copy do far more.

A slow site cannot be saved by clever content. Google's Core Web Vitals are part of the page experience ranking signal, and more importantly they are part of every other signal. Bounce rates rise on slow pages. Conversion rates fall. Users leave. The compound effect on rankings is much larger than the direct signal.

If your largest contentful paint is over 2.5 seconds, fixing that one metric will produce more SEO gain than three months of content work. Run PageSpeed Insights on your top ten pages and start there. Then move down the list.

A single link from a major newspaper, an industry-leading publication, or a credible university page is worth more than a hundred links from forum profiles, low-tier directories, or guest posts on sites nobody reads. This has been true since 2012 and has only become more true with each penalty Google has rolled out against link spam.

The operators who have built durable backlink profiles spend their time on a handful of relationships, not on outreach campaigns. They publish original research that journalists actually cite. They contribute to publications their customers read. They earn links by producing things worth linking to. It is slower, and it works.

The remaining tips do not need long sections. They need to be done.

• Match the SERP format (tip 4). Before you write anything, search the query and look at what is ranking. If the top ten are listicles, your essay will not rank. If the top ten are calculators or short tools, your 3,000-word guide will not rank. Two minutes here saves a week of wasted writing.
• Use real photography (tip 8). Stock photos are a quiet quality signal in the wrong direction. Real photos, real screenshots, and original diagrams are not. Google's quality raters notice.
• Build clusters, not scattered posts (tip 9). One landing page surrounded by ten supporting articles all linking back to it will outrank fifty unrelated posts. Pick three topics where you actually have expertise and build a cluster around each. Ignore everything else.
• Real bylines, real bio pages (tip 10). Author authority is part of how Google scores quality, especially in your-money-or-your-life topics. A named author with credentials and a real bio page outranks anonymous content covering the same material.
• Delete or merge dead pages (tip 11). Pages getting under 10 monthly visitors over six months are diluting the site. Delete the irrelevant ones, merge the related ones, redirect everything.
• Measure conversions, not traffic (tip 12). A page ranking second for a query bringing 10,000 visits and zero leads is worth less than a page ranking fifth bringing 200 visits and twelve qualified calls. Most SEO reports lead with traffic because traffic moves and looks like progress. The reports worth reading lead with conversions, by page, tied to revenue. If your analytics is not set up to track this yet, fix that conversation before you fix anything else.

None of these twelve tips are clever. There is no secret here. The reason they work is precisely because they are unglamorous, hard to outsource cheaply, and require sustained attention over years. The operators who win at SEO are the ones who do these twelve things, more or less, for long enough that the compounding shows up. The ones who chase the next trick keep starting over.