I've been in ecommerce long enough to see the parallels between search engine rankings and Amazon marketplace dynamics. When a site gets hit with 70k injected pages, clearing the technical mess is step one. But Google's trust is like Amazon's account health score - once you've been flagged, the memory lingers. Six months of clean indexing is promising, but it's still early days. Impressions still trickling in? That's not a dead domain. A dead domain flatlines. Those accidental clicks? They're like a suppressed product getting sporadic views - the algorithm hasn't fully written you off.
i'd check the same things I'd audit for a suppressed ASIN: are the old hacked URLs still returning 200s? Sounds like you've got them returning 410s. good. Toxic backlinks? None. That's a win. Crawl budget? On Amazon, wasted budget kills your best sellers' visibility. Same here - make sure junk parameter URLs aren't sucking up Google's attention. You've noindexed tags and categories, that's smart.
The bit that really resonates is the low-competition keywords not ranking. In my world, that's like targeting long-tail search terms with a listing that has a tarnished account health flag. The algorithm holds site-level quality signals against you, even for easy queries. It's not about the keyword difficulty - it's about domain trust.
Given your clean index, growing impressions, and that Wikipedia backlink, I wouldn't kill the project. I'd treat it like a suspended account that's been reinstated: push quality content consistently, tighten technical SEO, and give it another 6-8 months. If after that period the pages still can't break top 50 for genuinely easy terms, then you start questioning whether the domain is burned. But right now? Too early to pull the plug. Focus on building the trust signals - internal linking, fresh content, no broken paths - and let time do its work.