The context
Before buying a laptop, people search for what goes wrong after they buy it. Warranty terms, service turnaround, whether the brand honours a claim. Reddit and Quora hold most of those answers, and AI engines lean on both heavily when a prompt turns to reliability or after-sales.
This brand’s own community record was working against it.
The challenge
Roughly one in three comments about the brand’s after-sales was negative.
We audited live Reddit and Quora conversation covering the brand and four competitors. After-sales dominates the category everywhere, so the first finding was context: this is not a single-brand problem, and roughly two in five comments across the whole dataset carried negative sentiment.
The brand’s own share of it was harsher. Negative comments outnumbered positive ones on Reddit, and on Quora the question titles themselves carried the damage. Three complaint clusters dominated: refund and replacement handling, warranty coverage and denial, and repair turnaround and parts availability.
The age of the content was the real problem. Quora answers up to eight years old were still indexed, still ranking and still being cited. In the brand’s parallel LLM audit, sentiment trailed the category leaders and the recommendation rate on brand-trust queries was zero.
The strategy
Treat it as a standing function, not a campaign
Reddit and Quora reward aged accounts and consistent participation. A short push reads as astroturfing, gets called out, and leaves the brand worse off. The engagement was scoped as always-on from the start: the first year to reset the record, continuous presence to hold it.
Respond on-thread, publicly, and close the loop
Official, human responses on live negative threads — acknowledge, resolve, and post the resolution where the complaint sits. A thread that ends in a fix reads very differently to an engine than a thread that ends in silence.
Three voices, not one
An official brand account for AMAs, launches and complaint resolution. Two to three aged category-expert accounts posting reviews and comparisons, with roughly four in five posts unrelated to the brand. Five to seven helper accounts answering ‘help me pick’ threads, never first in a comment chain.
Add new signal at volume, at a 4:1 value-to-promotion ratio
Long-term usage reviews, price-band answer threads, service-experience posts and structured comparisons — the content shapes engines quote — published steadily across the target subreddits and the highest-ranking Quora questions.
The execution
Monitoring ran daily across both platforms with escalation alerts before a thread gained momentum, and negative threads were answered on-platform rather than routed to a support inbox.
Content ran in three phases across the six months. Ground building for the first two: account seeding, karma, no promotion. Content flooding through months three to five: weekly comparisons, long-term reviews, the first AMA. From month six, monitoring what the engines actually returned and scaling the thread formats that were being cited.
The outcome
Over six months the brand’s AI sentiment score closed 28% of its gap to the category leaders, and brand-trust queries — which had returned the brand as a recommendation in none of the audited prompts — began returning it in four months.
On-platform, the negative share of the brand’s community conversation fell to 20%. The old threads are still there. They are no longer the only thing an engine finds.
Baseline from a mid-2026 audit of the brand’s and four competitors’ Reddit and Quora threads, coded for sentiment and complaint type. Closing figures measured on the same basis.
What we learned
Negative community content does not decay. The eight-year-old Quora answers in this dataset were being surfaced as current. Every new search and every new AI prompt re-surfaces the same threads until something outweighs them.
Deletion is not available and volume is the only lever. Stop the programme and the old threads climb back.
- Social & sentiment audit
- Reddit and Quora community strategy
- Always-on monitoring and response
- YouTube creator programme