Blog SEO

I start with a keyword and a SERP check. For example, for 'how to clean suede shoes', the top results are step-by-step guides with photos, not product reviews. That tells me the intent is instructional, not commercial. I pick a primary keyword and two supporting ones: 'clean suede shoes at home' and 'remove water stains from suede'.

I outline with an H1 as the title, then H2s for materials, step-by-step instructions, and tips. H3s cover specific stain types. I write naturally, covering those subtopics without forcing the keyword. Next I craft a meta description that includes the primary keyword and a click incentive — something like 'Learn the two-ingredient method that works every time.'

I add at least one internal link to a related post. For the suede guide, that might be a piece on general shoe care. I also include an image with descriptive alt text. Finally, I set a calendar reminder to review the post in six months.

A counter-argument: some SEOs say write first, optimise later. I used to do that, but rewriting took longer. I prefer to build SEO into the outline from the start.

Here is a worked example with numbers. A client's blog post on 'best running shoes for flat feet' sat on page three for six months. After we restructured it to match the intent of the top results — a comparison table rather than a listicle — it moved to position four in three weeks. Organic traffic to that post increased by 42%. The change? Not the content, but the format and intent alignment.

I use a decision rule: if the top three results are all long guides, I write a long guide. If they are short tips, I keep mine short. Matching the format is often more important than hitting a word count.

A quick aside on linking strategy: don't limit your internal links to other blog posts. Link to relevant product pages or service pages when it makes sense. I once linked from an informational post about 'how to fix a leaky tap' to a tools product page, and that page saw a 20% increase in referral traffic from the blog.

I also check for featured snippet opportunities. If the SERP has a featured snippet for a related query, I structure a clear H2 with a concise answer in the first paragraph. That can capture the snippet and earn extra visibility.

Biggest mistake I see: ignoring internal links entirely. I once worked on a blog that had 150 posts, none with any internal links. Each post was an island. The entire blog had only 200 visits a month from search. After a link audit and adding relevant internal links, the traffic doubled in two months.

Internal links distribute authority and help Google understand your site's structure. A post with zero internal links is nearly invisible. If you have a [SEO content strategy](/seo-content-strategy/) in place, you should plan which posts link to which.

Second watch-out: over-optimising headings. I used to stuff my H2s with exact-match keywords until a client pointed out how unnatural it read. Now I write headings for humans first. They should make sense out of context. I check them against the [on page SEO](/on-page-seo/) basics, but I never force the keyword.

Third: neglecting the [meta description](/meta-description/). Many people leave it blank and let Google auto-generate one. That is a missed opportunity. A good meta description can improve click-through rate by 10-20%. I write one every time, around 155 characters, that summarises the post and includes a reason to click.

Edge case: what if your blog is brand new and has no existing content to link to? Then you write the post anyway and link to it later when you have more content. Do not skip the post just because you have no internal links to add. That is the wrong decision.

Another watch-out: ignoring external links. Some SEOs avoid them because they 'leak authority'. But Google uses external links as a quality signal. If you reference a statistic or claim, link to the source. I once updated a blog post to include a link to a government study, and that post moved from position 6 to position 3 in four weeks. The external link added credibility.

Finally, watch out for content decay. A post that ranked well a year ago may have fallen because competitors updated theirs or because your information is outdated. I schedule a review every six months. I check for broken links, outdated stats, and new subtopics to cover. This is part of a regular [content audit](/content-audit/) process.

A decision rule I use: if a post has not been updated in over a year and traffic has dropped by more than 30%, it needs a refresh. Otherwise I leave it alone.

I used to believe blog post length was a ranking factor. I wrote 3000-word posts thinking they would outrank shorter ones. Then I saw an 800-word post outrank mine because it matched search intent better. The short post answered the question directly; mine buried the answer in fluff. I changed my mind: word count matters only as much as the topic demands.

I also got keyword density wrong. For years I tracked the exact percentage of keyword occurrence in the text. I tweaked paragraphs to hit 1-2% density. That led to unnatural writing. I stopped tracking it altogether, and my posts read better. Google is smarter than a density check.

Another thing: I neglected image optimisation and [alt text](/alt-text/). I thought it was a nice-to-have. Then a client asked us to add alt text to all product images on their blog. Image search traffic increased by 15% within two months. Now I treat alt text as essential, not optional.

What I still am not sure about: optimal blog post frequency. I tried publishing daily for three months. Traffic barely moved. I tried weekly — the same. I suspect consistency matters more than frequency, but I do not have a definitive number. If you run a [SEO blog](/seo-blog/) and publish once a month, that might be enough if each post is high quality.

I also got wrong the value of refreshing old posts. I used to write and forget. One of my own blog posts on 'how to tie a tie' dropped from 5000 visits to 2000 over two years. After I updated it with new images, clearer steps, and added a section for the half-Windsor knot, it returned to 4000 visits within a month. That taught me that updating is often cheaper than writing new.

Finally, I changed my mind about [seo writing](/seo-writing/) for commercial intent. I used to think you could not rank a blog post for a commercial keyword like 'buy running shoes' without being a product page. But a well-written informational post that compares brands can rank for transactional queries if you include a clear call to action and internal links to the purchase page. I still get this wrong sometimes — I over-optimise the blog post for the keyword instead of the user's next step.