Intercom Fin alternative
By Saurav | Founder of saavos | Building in public toward $10k MRR
[!TLDR] Intercom Fin charges $0.99 per resolved conversation, on top of a $29/seat/month Essential platform fee. At 200 chatbot resolutions a month — normal for a small SaaS with a few hundred active users — that is $228/month before you add a second seat. For a founder at $500 MRR, that is 46% of revenue going to a support chatbot. Flat-rate alternatives (saavos at $19/month, Chatbase at $32/month annual, SiteGPT at $39/month) cover the same job for 8–17% of that cost. Here is when Fin is genuinely the right call, and when the math does not close.
I built saavos partly because I watched this pricing model play out on a few products I worked on before going indie.
The pitch for per-resolution billing sounds reasonable: you only pay when the bot actually solves a problem. You are not funding capacity you do not use. Performance-aligned pricing.
The problem is what "resolved" means in practice, and who decides.
Intercom Fin costs $29/seat/month as a platform fee plus $0.99 per resolved conversation. At 200 resolutions per month — realistic for a SaaS with 300–400 active users — that is $227/month from a single seat. The cost scales directly with how well your bot performs, which means a better-tuned Fin gets more expensive, not cheaper.
Intercom's current pricing (confirmed from intercom.com/pricing, May 2026):
A "resolved" conversation is one where Fin answered the question without needing human escalation. Intercom determines this automatically based on whether the visitor closed the chat without clicking "speak to a person" and without sending additional messages after the bot's last reply.
The mechanism matters because it is not purely opt-in. A visitor who reads Fin's answer, closes the chat, and emails you directly still counts as a resolution. You pay $0.99 regardless of whether the visitor got what they needed.
At different monthly resolution volumes, here is what Fin costs (single seat, annual billing):
| Monthly resolutions | Fin outcome fees | + Essential seat | Total/month |
|---|---|---|---|
| 50 | $49.50 | $29 | $78.50 |
| 100 | $99 | $29 | $128 |
| 200 | $198 | $29 | $227 |
| 500 | $495 | $29 | $524 |
| 1,000 | $990 | $29 | $1,019 |
For context: 200 resolutions/month is a realistic number for a SaaS product with 300–400 active users and a decently-trained bot. Not a high-traffic site. A normal-sized indie SaaS.
At 200 resolutions, Intercom Fin costs about $2,724/year. saavos Business at $199/month (15,000 messages, no resolution cap) costs $2,388/year. You can handle 75 times the message volume for less money — because saavos counts messages, not resolutions, and charges flat.
Fin makes financial sense in three specific situations: you already pay for Intercom's full platform (so you're comparing add-on cost, not full tool cost), you're measuring the bot against human agent cost rather than other chatbots, or you're past $1M ARR and need the CRM routing and user-segmentation context that standalone chatbots don't have.
I want to be honest about this because there are real situations where Fin's pricing structure makes sense.
You already have Intercom for everything else. If your team uses Intercom for CRM, ticketing, user segmentation, and live chat, Fin is an add-on to a platform you are already paying for. The incremental cost compares against spinning up a separate chatbot tool, not against a standalone chatbot price.
Your alternative is a human agent. Intercom targets the comparison against human support cost, not against other chatbots. If your fully-loaded cost per human-handled ticket is $15–$25, paying $0.99 per Fin resolution is a 94–96% reduction per resolved ticket. That math is excellent. The comparison breaks down when you compare Fin against other chatbots — which also handle resolutions, just at a flat monthly rate.
You are past $1M ARR and need the integrations. Intercom's routing logic, user segmentation, conversation data, and CRM integrations are genuinely useful at scale. Fin is not a standalone chatbot; it is a feature inside a CRM platform. The per-resolution fee buys you that context. If you are past Series A and your support team is routing tickets across multiple channels with user data attached, Fin fits inside a workflow that cheaper tools cannot replicate.
For everyone else — specifically for founders in the $0–$50k MRR range who need a chatbot to deflect FAQ traffic — the comparison is wrong. You are not comparing Fin against human agents. You are comparing Fin against a $19–$39/month tool that does the same deflection job.
The three strongest flat-rate alternatives to Intercom Fin are saavos ($19/month), Chatbase ($32/month annual), and SiteGPT ($39/month). All three train on your existing docs, embed on your site with a script tag, and charge a fixed monthly fee with no per-resolution variable. None of them have Fin's CRM routing — but for FAQ deflection on a marketing site or docs page, they don't need to.
| Tool | Entry price | Messages/month | AI model | Free tier | Multi-channel | CRM context |
|---|---|---|---|---|---|---|
| saavos | $19/mo | 1,000 | Claude Sonnet 4.6 | Yes (permanent) | No | No |
| Chatbase | $32/mo annual | 500 credits | GPT-4o, Claude, GPT-5 | Yes (50 msg/mo) | WhatsApp, IG | No |
| SiteGPT | $39/mo | 4,000 | GPT-4.1 | No (7-day trial) | No | No |
| Intercom Fin | $29/seat + $0.99/resolution | Unlimited resolutions | GPT-4 class | No | Yes (full) | Yes (full) |
All pricing confirmed from vendor pages, May 2026.
saavos ($19/month Starter, $49/month Pro): Claude Sonnet 4.6, 1,000 messages/month at Starter, 2 bots, API access at $19. No per-resolution fees. Forever-free tier (50 messages/month, no credit card, no expiry) so you can test with real visitors before committing. Website-only — no CRM, no live chat routing, no user segmentation. The job it does: answer visitor questions from your training sources, cite the original pages, hand off to email when it cannot answer. That is the whole product.
Chatbase ($32/month Hobby on annual billing, $40/month monthly): Multi-model selection (GPT-4o, Claude, Gemini, Meta, DeepSeek), 500 message credits/month at Hobby — where one credit equals one user-bot exchange, not one conversation. WhatsApp and Instagram integrations on paid plans. More integration breadth than saavos; tighter message cap at the entry tier. Model flexibility is the real differentiator if you have a preference.
SiteGPT ($39/month Starter): 4,000 messages/month — the highest message volume in the $20–$50 tier range. GPT-4.1 and GPT-4.1-mini model options. 95+ language support. No free tier (7-day trial only). If your site gets significant traffic or your audience is multilingual, SiteGPT's Starter gives the most messages per dollar of any platform in this range.
None of these have Intercom Fin's CRM context, routing logic, or multi-channel inbox. If those things matter to you, the comparison is not relevant. But if the job you are hiring for is "answer the questions my docs page already answers, 24/7, so I do not have to," all three do that job at a fraction of Fin's per-resolution cost.
For a SaaS with 400 monthly active users and a 15% chat engagement rate — about 60 chatbot sessions, 240 messages — saavos Starter costs $19 flat. Intercom Fin at the same traffic level costs $64 or more depending on how many of those 60 sessions resolve. The gap is structural, not incidental: per-resolution pricing means a better bot costs you more every month.
Say you have 400 monthly active users. Maybe 15% of them open the chat widget in a given month — 60 visitors. Average session is 4 messages. You get 240 chatbot messages per month.
At saavos Starter ($19/month), 240 messages costs you $19. Simple.
If 60% of those 60 conversations resolve without escalation — 36 resolutions — Intercom Fin charges $29 (seat) + $35.64 (36 × $0.99) = $64.64 that month. Three times more expensive for the same number of resolved conversations.
The gap widens as your bot gets better. A better-tuned bot resolves more conversations. On flat-rate pricing, that is just a better product. On per-resolution pricing, a more effective bot directly increases your monthly bill.
This is the structural problem with outcome-based pricing when the tool is working. The vendor's revenue grows with your bot's success. Your costs are unpredictable. For a bootstrapped founder managing a tight budget, unpredictable costs are a real operational problem — not just a math preference.
Use a flat-rate alternative if you are pre-$1M ARR and mainly need FAQ deflection. The specific pick: saavos at $19 for under 1,000 messages/month, SiteGPT at $39 or saavos Pro at $49 for 1,000–4,000 messages, Chatbase at $32/month annual if you need model selection or WhatsApp. Stick with Fin only if you already pay for Intercom's platform or your support workflow genuinely needs CRM context that standalone bots cannot provide.
Use Fin if: You already pay for Intercom's platform, you are comparing against human agent cost (not other chatbots), or your support workflow needs CRM context, user segmentation, and multi-channel routing that standalone chatbots cannot provide.
Use a flat-rate alternative if: You are pre-$1M ARR, you mainly need FAQ deflection on your marketing site or docs, and you want a predictable monthly cost. At $19–$39/month you get a well-trained chatbot running a capable model, zero per-resolution surprises, and enough messages to cover a realistic indie SaaS traffic load.
The specific pick within the flat-rate tier:
If you are already paying for Intercom's full platform and just want to compare the Fin add-on cost to what you would pay if you hosted the chatbot separately: the break-even is around 20 resolutions per month. Below 20 Fin resolutions/month ($19.80 in Fin fees), Fin is cheaper than running a separate tool. Above 20, a standalone flat-rate chatbot starts costing less.
Most sites with enough traffic to care about a chatbot are above 20 resolutions. Which is why most founders I talk to who switched away from Fin did it not because the product was bad, but because the math stopped working as their bot improved.
Get the next post in your inbox
Honest writing on building, embedding, and shipping AI chatbots. No spam. Unsubscribe anytime.
Only if you are already an Intercom customer. The $29/seat platform fee plus $0.99/resolution makes Fin expensive relative to standalone chatbots for low-to-medium traffic. A solo founder paying $228/month for 200 Fin resolutions could handle the same volume with saavos Business at $199/month — more messages, lower cost, no resolution-based variable.
A conversation is counted as resolved when the visitor closes the chat without escalating to a human agent and without sending additional messages after Fin's last reply. This is determined automatically. A visitor who reads Fin's answer and emails you directly still counts as a resolution.
Yes. saavos runs as a chat widget embedded on your site, independent of Intercom. Some founders use saavos on their marketing site — where FAQ deflection is the job — and Intercom for in-app support where CRM context and routing matter more. The two tools serve different parts of the support surface.
Export your Intercom conversation logs (CSV from Reports → Conversations). Pull the 30 most common first-message questions — these are your training priorities. Set up your new bot, train it on those question topics, embed it, and run it for two weeks alongside the Intercom widget before removing Fin. This way you see real deflection rates before cutting over. The switch itself takes under an hour.
saavos Starter at $19/month is the cheapest flat-rate Intercom Fin alternative for indie SaaS. It includes Claude Sonnet 4.6, 1,000 messages/month, 2 bots, and API access. No per-resolution fees. For higher message volumes, Chatbase Hobby ($32/month annual) or SiteGPT Starter ($39/month, 4,000 messages) are the next options. All pricing confirmed from vendor pages, May 2026.
Below about 20 resolved conversations per month, Intercom Fin's $0.99/resolution fee ($19.80) plus the $29 Essential seat totals less than the cost of adding a separate chatbot tool on top of an existing Intercom plan. Above 20 monthly resolutions, standalone flat-rate chatbots start undercutting Fin's total cost. For sites with enough chatbot traffic to care about resolution rates, 20 resolutions/month is typically crossed within the first few weeks of a working bot.
Builds tools for solopreneurs and small SaaS teams who don't have an afternoon to spare.
Paste your URL. Train your bot. Drop one script tag. No credit card.