SEO Toolset

I reach for an SEO toolset when I need to find problems I cannot see by browsing a site. A recent crawl of a 12,000-page ecommerce site revealed 340 pages with duplicate title tags and 180 orphaned product pages. Without the tool I would have missed the orphans entirely — they had no internal links and were invisible to Google. The tool also flagged 23 broken canonicals that were pointing to the wrong URL. That is where a toolset earns its keep: surfacing structural issues at scale.

Keyword research is another strength, but I treat the numbers as directional. I once targeted a keyword that a tool rated as 'low difficulty', only to find the entire first page taken by Amazon product pages. The tool had not accounted for SERP intent. So I use keyword data to generate hypotheses, not guarantees. A volume estimate of 500 searches a month is a rough guide, not a contract.

Where tools help me most is prioritisation. I export crawl data and filter by 'high impact' (pages with traffic but missing H1s) vs 'low effort' (fixing 30 redirects). That decision rule alone saves hours. For a client site, I fixed 50 broken links on the top 10 landing pages first because the tool showed they were losing 15% of referral traffic. Without the data, I would have started with the homepage.

I also automate rank tracking for 20 to 30 pages that matter. The tool sends a weekly email. I keep the list small because tracking everything is noise. Used deliberately, a toolset accelerates diagnosis and prioritisation. But you have to know which levers to pull.

Rank tracking is the biggest lie in most SEO toolsets. I have run side-by-side checks between a third-party rank tracker and a fresh incognito search from a different location. The tool claimed position #7 for a phrase; I saw #12. The discrepancy is usually 3 to 5 spots because tools use aggregated data, not your actual personalised SERP. I now treat rank trackers as trend indicators, not truth. If a keyword drops from #4 to #6 over a month, I act. A one-day shift I ignore.

Measuring real business impact is another blind spot. A tool might show that a page's organic traffic doubled after an optimisation, but it cannot tell you whether conversions moved. I once optimised a landing page that increased traffic by 40% and cut signups by 30% because the content attracted the wrong audience. The tool made the change look like a success. You need analytics or heatmaps to connect traffic to outcomes. I pair SEO tools with behaviour analytics to see what visitors actually do.

Local SEO nuance is poorly served by general tools. They typically pull desktop rankings from a central datacentre, not localised results for 'plumber near me'. For a small roofing client, a tool showed local pack rankings in the top three. When I checked from a real phone in their service area, they were not in the pack at all. The tool was using a fixed location that was 50 miles away.

Edge case: small sites. If your site has fewer than 500 pages, most keyword volume and backlink metrics are statistically meaningless. The sample sizes are too small. I wasted weeks on a niche blog with 200 pages, chasing keywords that had zero real impressions. Now I skip paid tools entirely for sites that size and rely on free alternatives.

I admit I used to buy a yearly subscription to a premium all-in-one tool before I understood its limits. Now I rely mostly on a free crawler and first-party data.

I have settled on a small stack that covers most of what I need. First, a crawler: Screaming Frog SEO Spider is my go-to. It is free for up to 500 URLs and fast. I use it for technical audits: finding broken links, duplicate titles, and orphan pages. For sites over 10,000 URLs, the paid licence is worth the £149 a year.

For competitive research, I rotate between [Semrush](/semrush/) and [Ahrefs](/ahrefs/). I prefer Ahrefs for backlink analysis because its index feels more complete for the niches I work in (UK and Thai markets). Semrush is better for keyword gap analysis — comparing my site's keyword coverage against five competitors at once. I only subscribe when I have an active project that needs competitor data.

[Bing Webmaster Tools](/bing-webmaster-tools/) is underused. It gives you keyword data straight from Bing, including search volume and average position. I have found queries there that Google Search Console didn't surface. And it is free.

For user behaviour, [Hotjar](/hotjar/) or [Fathom Analytics](/fathom-analytics/). I use Hotjar for heatmaps when I need to see where people click; Fathom for lightweight privacy-focused analytics. Both fill the gap that SEO tools leave between 'traffic went up' and 'why'.

If I need a free alternative for keyword ideas, [Ubersuggest](/ubersuggest/) works for quick brainstorming. Its volume data is less accurate than the paid tools, but it costs nothing.

My decision rule: for any new site, start with [Screaming Frog](/screaming-frog/) and the free Google tools. Only add a paid tool when you have a specific problem that free tools cannot solve. Most SEO tool subscriptions are sold as insurance; I prefer to buy them as medicine.