The Chatbase playbook: how it got to $863,632 a month

Last verified Jul 31, 2026 · TrustMRR listing removed Oct 25, 2026 · Facts checked Oct 21, 2026 · Every fact links its source and evidence label · How we verify

The numbers

Monthly recurring revenue (last verified, Jul 31 2026)	$863,632
Last 30 days (per listing, same sync)	$865,869
All-time revenue	$19,213,227
Active subscriptions	12,424
Revenue per subscription (our calc.)	~$70/mo
Signups from ChatGPT + Claude (founder, Oct 2026)	37% Self-reported

Source: TrustMRR, which says the Stripe key expired and data was last updated Jul 31, 2026. We'll update this page if the feed comes back.

How Chatbase makes money

Chatbase lets businesses build and deploy AI agents for customer support, sales and product guidance across chat, email, voice, WhatsApp and Slack, without building AI infrastructure themselves.

Plans listed on TrustMRR: Free, Hobby $40/mo, Standard $150/mo, Pro $500/mo and Enterprise (custom), with annual plans 20% less. At about $70 per subscription per month, Chatbase sits close to the median of the 24 verified AI SaaS businesses we track (about $52). Its size comes from volume, not an unusually high price.

It didn't start there. The first version was a consumer tool at $10 and $30 a month, and the founder says the biggest pricing impact came from moving up-market Self-reported source.

1. The order they did it in

1. Late 2022: idea while building other AI side projects; MVP took ~6–8 weeks. Self-reported source
2. Feb 2, 2023: launch tweet to 16 followers goes viral. Self-reported source
3. First paying customer ~30 minutes after the pricing page went live. Self-reported source
4. Feb 11: $400 MRR · Feb 16: $900 · Feb 28: $3,000. Self-reported source
5. Mar 15, 2023: $10,000 MRR. Self-reported source
6. May 13, 2023: $64,000 MRR. Self-reported source
7. $1M ARR in 117 days. Self-reported source
8. Jul 31, 2026: $863,632 MRR (last TrustMRR sync). Verified source

2. How they got customers

1. Tweeted the "ChatGPT for your PDFs" MVP to 16 followers on Feb 2, 2023; it went viral before any pricing page existed. Self-reported source
2. AI influencers then posted about it. Self-reported source
3. Launched on Product Hunt. Self-reported source
4. Posted to relevant subreddits; "a story works better than 'here is a tool I made'". Self-reported source
5. Active on Indie Hackers; submitted to AI directories. Self-reported source
6. No paid marketing in the first three months: building in public on Twitter, LinkedIn and subreddits, "100% organic". Self-reported source
7. Later: B2B sales calls/demos, SEO/AEO (one person dedicated), "warm outbound" to inbound traffic, and SF billboards for brand. Self-reported source

3. Stack and costs

1. 2023 stack: React, Next.js, Supabase, OpenAI API, Langchain, Pinecone, Stripe. Self-reported source
2. Early model costs were paid from personal savings until revenue covered them; profitable since. Self-reported source
3. Exact monthly tool costs: not disclosed. Self-reported source
4. Team: solo; two employees added by May 2023 (coding + support). Self-reported source

4. What didn't work (skip these)

1. Viral/influencer marketing "is not a good long-term strategy" once copycats appeared. Self-reported source
2. Early churn was high: many users were just experimenting with AI. Self-reported source
3. Started as a B2C tool priced at $10/$30; the biggest pricing impact came from moving up-market. "Maybe the biggest mistake was not experimenting more" with pricing. Self-reported source
4. Failed two final university classes to focus on Chatbase. Self-reported source

Doubts we flagged

• MVP build time is "about two months" (2023 interview) vs "six weeks" (2026 interview).

5. What you could copy

Our read of the facts above. Each line points back to its source.

1. Ship in public before the pricing page. Chatbase was tweeted to 16 followers and went viral before there was any pricing. The first payment came about 30 minutes after pricing went live. If you have an MVP, show it now and add the price when people start asking. founder interviews
2. Post a story, not a tool. The founder's own advice for subreddits: "a story works better than 'here is a tool I made'". Write what you built and why, with numbers, then link. Indie Hackers interview
3. Price for teams, not hobbyists. It started at $10 and $30 for consumers, and the founder says moving up-market had the biggest pricing impact. Today's plans run $40 to $500 a month, and the last verified figures work out to about $70 per subscription (our calc). EO interview + TrustMRR
4. Expect tourists to churn. Early churn was high because many users were just experimenting with AI. Count retention from month two, not signups from launch week. EO interview
5. Get named by AI answers. In Oct 2026 the founder said 37% of signups come from ChatGPT and Claude, and that he is hiring a growth marketer to focus on AEO (getting cited by AI answers). Pages that answer the exact question your buyer asks are the cheap version of this. founder on X
6. Don't copy the horizontal product. Copy it for one niche. Our verdict on Chatbase is "Hard": Category leader with a 3-year head start and a strong backlink profile (Ahrefs DR 81). The replicable lesson is vertical: an agent for one niche. The businesses below did exactly that. our verdict

Niche version	MRR	Subs	$/sub
Adelante A managed AI support agent for Shopify brands: "we set it up, then we run it".	$17,837	70	$255
ChatDash White-label platform for agencies selling AI voice agents to their clients.	$19,483	113	$172
notis.ai A one-person AI assistant inside WhatsApp that turns voice notes into tasks.	$13,394	443	$30

Verified via TrustMRR (TrustMRR, Oct 22 2026). $/sub is our estimate (MRR ÷ active subscriptions).

6. Copy-paste replication prompt

Paste this into ChatGPT or Claude. It only uses the sourced facts above, and it asks about your skills, audience and budget before it plans anything.

You are my AI co-founder. Help me replicate the playbook behind Chatbase, using ONLY these sourced facts (evidence label in brackets):

GROWTH METHODS (in order):
- Tweeted the "ChatGPT for your PDFs" MVP to 16 followers on Feb 2, 2023; it went viral before any pricing page existed. [self]
- AI influencers then posted about it. [self]
- Launched on Product Hunt. [self]
- Posted to relevant subreddits; "a story works better than 'here is a tool I made'". [self]
- Active on Indie Hackers; submitted to AI directories. [self]
- No paid marketing in the first three months: building in public on Twitter, LinkedIn and subreddits, "100% organic". [self]
- Later: B2B sales calls/demos, SEO/AEO (one person dedicated), "warm outbound" to inbound traffic, and SF billboards for brand. [self]

STACK & COSTS:
- 2023 stack: React, Next.js, Supabase, OpenAI API, Langchain, Pinecone, Stripe. [self]
- Early model costs were paid from personal savings until revenue covered them; profitable since. [self]
- Exact monthly tool costs: not disclosed. [self]
- Team: solo; two employees added by May 2023 (coding + support). [self]

TIMELINE:
- Late 2022: idea while building other AI side projects; MVP took ~6–8 weeks. [self]
- Feb 2, 2023: launch tweet to 16 followers goes viral. [self]
- First paying customer ~30 minutes after the pricing page went live. [self]
- Feb 11: $400 MRR · Feb 16: $900 · Feb 28: $3,000. [self]
- Mar 15, 2023: $10,000 MRR. [self]
- May 13, 2023: $64,000 MRR. [self]
- $1M ARR in 117 days. [self]
- Jul 31, 2026: $863,632 MRR (last TrustMRR sync). [verified]

WHAT DIDN'T WORK:
- Viral/influencer marketing "is not a good long-term strategy" once copycats appeared. [self]
- Early churn was high: many users were just experimenting with AI. [self]
- Started as a B2C tool priced at $10/$30; the biggest pricing impact came from moving up-market. "Maybe the biggest mistake was not experimenting more" with pricing. [self]
- Failed two final university classes to focus on Chatbase. [self]

Task: 1) Ask me 5 questions about my skills, audience, budget and niche. 2) Map each fact above to an equivalent action for my niche. 3) Produce a week-by-week plan to my first dollar, with the cheapest tool for each step. 4) Flag which steps depend on advantages I may not have (audience, timing, capital). Do not assume any fact beyond the list.

Copy the prompt

More on Chatbase: Chatbase case study · How much does Chatbase make? · AI revenue leaderboard