Hear everything.
Act on the part that matters.
Listening across five platforms. Abuse separated from criticism. Every takedown on the record.
- Five platforms
- Seven risk categories
- No AI in the scoring
- Built for campaigns
What is SocialHQ?
SocialHQ is social listening and comment protection software made by ZaadCore for PR agencies, publicists and talent teams. It listens across Instagram, Facebook, X, YouTube and TikTok, scores comments in seven risk categories without an AI model, flags coordinated attacks, and keeps a takedown register with the evidence captured at filing.
- Platforms
- Five
- Risk categories
- Seven
- Scoring
- Deterministic
- Case states
- Filed → closed
Three questions, in the order you ask them.
- 01 · Pulse
What happened since I last looked?
Volume, sentiment, topics and share of voice, with the change since your last check.
- 02 · Alerts
What crossed the line I set?
Six alert types, from one serious comment to a sudden burst, at thresholds you choose.
- 03 · Casework
What did we do about it?
A takedown register from filed to closed, with the evidence kept.
Criticism is not abuse.
Nothing here recommends hiding an opinion, however cold. Seven weighted risk categories separate a threat from a bad review, and the line is visible rather than buried in a model.
“Honestly the second half dragged. Expected more.”
No action. Criticism is not abuse.
“Free followers!!! DM now, link in bio”
Hide proposed. A person confirms.
“I know where you’ll be on Friday. Watch out.”
Escalate now. Never auto-hidden.
No language model sits between a comment and its score. The same comment always scores the same, and you can explain to a client, a platform or a lawyer exactly why it was flagged.
- English
- Malayalam
- Manglish
- Hindi
- Arabic
Built to be questioned.
Moderation decisions get challenged — by clients, by platforms and occasionally by lawyers. Every guard in SocialHQ exists so the answer is a record rather than a recollection.
- ✓ Evidence at filing Text, author, permalink and timestamp copied when the case is filed, not fetched later.
- ✓ Human confirmation Delete and reply require explicit confirmation. Autopilot proposes and never acts.
- ✓ Own posts only The tool refuses to moderate comments on posts the campaign does not own.
- ✓ Full audit log Every action and every completed collection recorded against an operator.
- ✓ Budget ceiling A server-side spend cap refuses a call rather than quietly trimming results.
- ✓ Withheld count shown When a review floor hides low-confidence items, the number hidden is always reported.
Connected to the channels you already run.
- MetaInstagram and Facebook posts, tagged media, hashtags and comments, with a realtime webhook
- X, YouTube & TikTokSearch across campaign terms, with comments collected and scored
- AlertingWhatsApp Business templates, email and generic webhooks
- Google SheetsThe flagged queue published for a team that lives in a spreadsheet
The flow
- PulseVolume, net sentiment, topics, share of voice and top accounts — with the change since your last check.
- AlertsSix alert types, from a single serious comment to a sudden burst, at thresholds you set.
- CaseworkA takedown register: filed, acknowledged, then removed, refused, lapsed or withdrawn.
Detection
- Risk scoringSeven weighted categories — threat, stalking, self-harm, harassment, bullying, misinformation, spam.
- Brigade detectionFires only when a cluster of accounts is both coordinated and hostile, not merely busy.
- MultilingualEnglish, Malayalam and romanised Manglish, Hindi and Arabic scored in one pass.
Investigate
- Boolean searchAND / OR / NOT, phrases, wildcards, NEAR proximity and field terms like author: or risk:>60.
- ProfilesWho is talking, their reach, and an account risk score for the ones worth checking.
- ReportsDaily, weekly and post-campaign reports, scheduled and delivered by email.
Act
- PlaybookNames the recommended action, says why, and gives paste-ready text. Hide before delete.
- AutopilotProposes hides and never executes them — and never for self-harm, threats or stalking.
- Sheet syncThe flagged queue published to Google Sheets, one tab per category, notes written back.
SocialHQ: quick answers.
Which platforms does SocialHQ cover?
Instagram, Facebook, X, YouTube and TikTok, with Reddit available and off by default. Instagram is the only one it can also moderate on, through the official Meta Graph API.
Does SocialHQ delete comments automatically?
No. Autopilot proposes hides and never executes them, never for self-harm, threats or stalking. Delete and reply always require an explicit human confirmation.
Does it use AI to decide what is abusive?
No — and that is deliberate. Scoring is deterministic, so the same comment always scores the same and the reason can be explained to a client or a platform. No language model sits between a comment and its score.
Can it read comments that are not in English?
Yes — Malayalam in both its own script and romanised Manglish, plus Hindi and Arabic. Romanised comments are typically the half of the volume generic sentiment tools score as neutral.
Who is SocialHQ for?
Publicists, talent management and PR agencies running campaigns under public pressure — film releases in particular, where comment volume spikes hard around a date.
How is a takedown tracked?
As a case with six states — filed, acknowledged, then removed, refused, lapsed or withdrawn — with the evidence captured at filing. Unanswered cases surface oldest first.
See it read a real comment section.
We run SocialHQ against a live campaign account: the pulse, the alerts, the cases.