Stuff people
actually ask.

BABAV is an omnichannel, AI-driven CRM that automates social media content creation, posting, and engagement across Threads, Instagram, Facebook, WhatsApp, YouTube, and Gmail. Think of it as a tireless social ops team that reads, replies, posts, and updates your CRM in your voice — built on Gemini 2.5 Flash and VEO 3.1.

Solo creators with growing audiences (50K–2M followers), DTC brands chasing fast customer response times, and small agencies running 5–20 client accounts. If you're drowning in DMs, comments, or cold-email opportunities and can't justify a full social ops team, BABAV is built for you.

Those are scheduling tools. BABAV doesn't just schedule — it generates the content, reads incoming messages, replies in your tone, updates the CRM, and learns from your knowledge base. It's an autonomous agent, not a post-queue manager.

Private beta as of 2026. Reach out at support@babav.co for access. We're onboarding a few dozen brands and creators per month while we tune the AI and add features.

Kickoff call within 48 hours of your first invoice. We connect your channels, ingest your knowledge base, calibrate tone, and set silence rules — you typically have at least one channel running on autopilot inside 5 business days, with all six fully live by week 2. Onboarding is white-glove, we do it with you.

Not at all. The whole platform is a dashboard. You'll upload some documents, pick tone settings, set a few rules, and click "go live." If you can use Notion or Slack, you can use BABAV.

Yes — approval mode is one click. The agent drafts everything, queues it, and waits for your thumbs-up. As you trust it more, you can switch specific channels or topics to fully autonomous.

Yes. CSV upload for cold-email lists, and any CRM export (HubSpot, Pipedrive, Airtable) maps into our lead database. Existing conversation history from supported channels is auto-pulled during connection.

Text generation runs on Gemini 2.5 Flash via Google Vertex AI. Video generation uses VEO 3.1 in both fast and standard modes. We chose Gemini for the latency and cost profile — sub-second inference at high volume.

You upload sample replies, brand documents, FAQs, and tone guidelines. The agent grounds every output against this knowledge base and runs a final tone-match check. We don't fine-tune models — we ground them. That means you can update your voice in seconds, not retraining cycles.

The agent is configured to refuse to answer questions outside its knowledge base, rather than guess. If a customer asks something not covered, it either drafts for human approval or sends a polite "let me check with the team" reply. You can tune the strictness per channel.

Yes. Silence rules let you mute the bot on specific keywords (e.g. "refund", "lawsuit"), on certain hours (weekends, launch days), or on individual contacts. The bot routes those messages to humans automatically.

That's the central design goal. We pass every output through a tone-match scorer and a "giveaway phrase" filter that strips the hallmarks of bot writing (over-explanation, excessive emojis, robotic phrasings). Beta customers' audiences haven't flagged the replies as AI in months of testing.

All infrastructure runs on Google Cloud Platform, specifically Cloud Run in us-central1. Storage uses Firestore and Cloud Storage with encryption at rest. EU data residency is available on Enterprise.

All third-party tokens (Meta, Google, WhatsApp) are encrypted at rest using AES-256-GCM with a per-user data-encryption key. The DEK itself is encrypted by Google Cloud KMS. Tokens are never logged, never displayed in plaintext, and rotated on every refresh.

No. Under our Vertex AI configuration, your prompts and outputs are not used to train Google's foundation models. We do not train any of our own models on customer data. Your knowledge base, replies, and conversation logs stay yours.

SOC 2 Type II is in progress (audit completing Q3 2026). We can share the gap-assessment report with Enterprise prospects under NDA today.

Yes. Account deletion is one click — every record (account, configuration, conversation history, encrypted tokens) is purged within 30 days, with anonymized backups expiring within 90.

Pay the $500 retainer directly through Stripe and we'll reach out within 4 business hours via email to get you set up. Everything happens async unless you specifically want a call — in which case you can grab a 30-minute slot on our calendar. Same plan, same launch bonus either way.

$500/mo retainer covers the team and platform, including $50 of usage. Above that, real per-action rates ($0.01 per comment reply, $0.02 per DM, $0.05 per post, etc.) accumulate during the month and settle on your next Stripe invoice. See the full rate sheet. No prepaid credits, no top-ups, no expirations.

No. Usage is logged only after a successful, completed action. If the Meta or Google API returns an error, our retry logic handles it without billing you. Billing hooks are tied to the success response, not the request — built into the platform from day one.

Yes, from your dashboard or by emailing support@babav.co. Full retainer refund within 14 days of your first paid charge if you're not happy. Usage already consumed isn't refundable.

Yes. Set a monthly spending cap in the dashboard and autopilot pauses when usage hits it. Inbound messages still queue and you can release the cap with one click — no surprise bills.

For month one only: $100 of free usage on top of the $50 included in the retainer. So you get $150 of action runway in month one — enough to load all six channels and see what the platform actually does at scale.

Threads, Instagram, Facebook, WhatsApp Business, YouTube, and Gmail. All six are included in the BABAV Done For You plan — we hook them up during onboarding in week one.

LinkedIn is on the roadmap for next quarter. We're waiting on stable API access to LinkedIn's messaging endpoints — once Meta-grade reliability is achievable, it ships.

TikTok is later this year, X is under evaluation pending their API policy changes. We'd rather ship something solid than half-broken — Instagram and Threads cover most of the use cases X used to.

An API is in private beta. It exposes the core actions (send message, generate content, fetch leads, update CRM). General availability when we're confident in rate limits and SDKs. Get in touch if you want early access.

Not currently — we route everything through Vertex AI under our enterprise agreement so we can guarantee cost, latency, and data-handling SLAs. BYO keys are on the Enterprise roadmap for 2027.

Yes — two-way sync with HubSpot and Salesforce is live; Pipedrive and Attio are in beta; everything else can pipe through Zapier or our outbound webhooks today.