SE Ranking's paid trial paywalled everything. Semrush's free tier found a real bug instead.
Quick context same as always: I test AI/SEO tools firsthand against my own real site (solostacklab.net) before writing anything - actual signups, actual trials, not just reading pricing pages. Disclosure: I've had affiliate relationships with both tools mentioned here (SE Ranking's is active, Semrush's application was actually declined after I wrote the review) - this isn't about promoting either one, it's one specific thing I noticed running both against the same domain.
I ran a full site audit on both, same domain, close together. Access level was different by design: Semrush's audit needs no signup and no card at all. SE Ranking needs a 14-day trial signup (which, separately, rejected my business Gmail address and made me set up a domain email just to get in).
Semrush's completely free tier found a real bug: www.solostacklab.net had no DNS record at all - not a redirect, not a 404, nothing. Confirmed it myself with nslookup, fixed it same session. The audit told me exactly which URL had the problem.
SE Ranking's trial - the one that made me jump through a signup hoop - gave me a Health Score and issue counts by category ("No inbound links: 6," "Internal links to 3XX redirect pages: 7"). Clicking into the actual issue report to see which pages had those problems replaced every single detail row with "Available by subscription" and an "UNLOCK AUDIT" button. Despite being on an active trial, not even the free tier, I got zero actual URLs.
So the tool that needed no signup gave me a specific, fixable bug. The tool that made me clear a signup hurdle gave me a category and a count, then asked for more money to see which page. That's backwards from what the friction difference would predict.
Same shape I flagged with 4 other tools back in August (Surfer, SE Ranking again, Semrush's own CWV widget, Headline Studio) - free tier shows you there's a problem, paywalls the part that tells you what to actually do about it. What's new this time is the direction: I'd have guessed the tool that made me work harder to sign up would give me more, not less.
Full writeup with both audits side by side: solostacklab.net/comparisons/semrush-vs-se-ranking/
Anyone else found a "more friction, less actual data" pattern like this, or is this just these two specific tools?
I haven't tested these two, but your comparison suggests a better trial limit: one fully actionable issue, including a rescan after the fix. Then charge for coverage and ongoing monitoring. A health score asks me to trust the diagnosis before I've been able to verify even one finding.
That's a sharper design than either tool actually implements. Semrush's free tier got close by accident - it gave us one fully specific, real finding (the DNS issue) - but there's no built-in rescan step; we had to verify the fix ourselves with nslookup outside the tool entirely. A native "here's the issue, fix it, here's confirmation it's resolved" loop would prove the diagnosis is trustworthy before asking for a card, which is a different (and arguably more honest) pitch than "we found N problems, pay to see which ones."
Having to leave the tool to verify the fix is the missing step I meant. Would completing that loop have made you more willing to pay for ongoing monitoring, or would it still have been a one-off need? That seems like a useful distinction between a trial that proves value and a problem that supports a subscription.
Honestly, closing that specific loop would still have been a one-off need for us - once the DNS record existed and we confirmed it resolved, there's nothing left to keep monitoring about that particular finding. What would actually justify a subscription isn't re-verifying an old fix, it's catching new issues as the site changes - more pages, more integrations, more chances for something to break quietly. So I'd draw the line differently than "complete the loop, then people pay": the loop earns trust in the diagnosis, but the subscription's value proposition has to be about ongoing new-problem detection, not confirmation of the one you already fixed. We haven't felt a need for continuous audit monitoring yet mostly because the site's still small - that's a data point about our stage, not a verdict on the model.
Thanks, that answers what I was wondering. The trial can earn trust even when the site doesn't yet need ongoing monitoring. Your point about the stage of the site helps explain why a useful trial can still end without a subscription.
Exactly - appreciate you pushing on this, it sharpened something I hadn't articulated clearly before.