Slack Chatbot Pricing Calculator

Why a Slack chatbot has no single price

The first thing to clear up: putting a chatbot inside Slack is free on Slack's side. You install an app, grant it scopes, and it posts to channels or answers in a direct message — Slack charges a workspace for human seats, not for the bots those humans add. Every dollar on the recurring invoice comes from the chatbot or AI-agent platform doing the answering. That is the number this calculator estimates.

The catch is that those platforms do not agree on how to charge. Across the categories we review, four billing structures dominate, and the same support workload can be the cheapest or the most expensive option depending on which one a vendor uses. Comparing sticker prices is meaningless until you translate each into the same unit — cost for your monthly volume — which is exactly what the tool above does.

The four pricing models, and who each one punishes

1. Per agent seat

You pay a flat monthly rate for every human agent with access, regardless of how many conversations come in. This is the classic customer-service helpdesk model. Its cost is volume-independent, so it gets cheaper per conversation the more the bot deflects — it wins once your volume is high and your containment rate lets you hold a small, stable agent count. It punishes small teams who are forced onto a minimum number of seats they do not need.

2. Per AI resolution

You pay once for each conversation the bot closes on its own. This is the model behind most modern AI-agent add-ons, and it is the most honest at low volume: zero conversations resolved means zero spend. But it scales in lockstep with success — the better your bot gets, the more resolutions you pay for — so a high deflection rate that would be pure savings under a seat model becomes a rising bill here. Forecast it against your busiest month, not your average one.

3. Per message (metered)

You get an included message allowance and pay per inbound message after that. Predictable for chatty, low-volume support, but it charges for length, not outcomes: a long back-and-forth thread that never resolves still runs up the meter. Conversations that need many turns — troubleshooting, multi-step flows — make this model expensive fast, which is why handing off cleanly instead of looping matters to the bill as well as to the customer.

4. Flat fee plus bring-your-own-LLM

You pay a flat platform subscription and supply your own model, so you also own the per-conversation token cost. Often the cheapest at mid volume if you can run an efficient model, but you inherit the LLM bill and its variability — model price changes, longer prompts, and retries all land on you. Our BYOLLM explainer covers when owning the model is worth the operational overhead, and the LLM API cost calculator estimates that token cost across providers so you can feed a real number into the field above.

How to read the crossover table

The most useful output is not today's cheapest model — it is the volume sensitivity table, which recomputes the winner at half, one, two, and four times your current volume. Volume-based models (per-resolution, per-message) climb as you grow; fixed models (per-seat, flat fee) stay put. The row where the winner changes is your crossover point: the volume at which staying on your current model starts costing more than switching. If that crossover sits just above where you expect to be in a year, negotiate the switch into your contract now rather than after the overage.

Three checks before you trust any quote

• Replace every default rate.The numbers preloaded here are typical starting points to make the comparison legible, not a specific vendor's price. Paste in the figures from your own pricing page or sales quote before you act on the result.
• Confirm what counts as a "resolution." Vendors define a billable AI resolution differently — some bill any answered conversation, some only ones the customer confirms were solved. That definition can double or halve the per-resolution column.
• Check whether the AI tier is bundled or an add-on. A low flat fee that excludes the AI capability you actually need is not a low price. Read what the base plan includes before comparing it to an all-in per-resolution rate.

From pricing model to business case

This tool answers "which billing model is cheapest at my volume." It does not tell you whether the bot pays for itself — that depends on the labor it offsets. Once you have a monthly platform cost, drop it into the chatbot ROI calculator to project payback against agent salaries, and read the three pricing models guide for the full decision framework on matching a model to your business shape. If you are reporting upward, the KPIs a board cares about guide translates these costs into the metrics that get budget approved.

Related Chatbotscape tools and resources

• Chatbot ROI calculator — does the bot pay for itself
• LLM API cost calculator — price the BYOLLM token cost
• Pricing an AI chatbot: three models — the decision framework
• Chatbot ROI guide — the full business-case methodology
• Intercom review — a per-resolution AI pricing benchmark
• SendPulse review — a flat-fee, multi-channel point of comparison

FAQ

Does Slack charge anything to run a chatbot?

No. Slack does not bill per bot or per app — installing a chatbot app on a workspace is free. A workspace still pays Slack for its human members under its own plan, but that cost is driven by your team size, not by the bot, so this calculator leaves it out and estimates only the chatbot platform's recurring cost.

Where do the default rates come from?

They are typical starting points we see across published pricing pages in the chatbot-builder, helpdesk, live-chat, and AI-agent categories — chosen to make the four models comparable out of the box, not lifted from one vendor. Treat them as placeholders and replace each with your real quote. The result is only as accurate as the rates you enter.

Is anything I type sent to a server?

No. The calculation runs entirely in your browser as client-side JavaScript. Your volumes and rates are not stored, logged, or transmitted — closing the tab discards everything.

Why does the cheapest model change as volume goes up?

Two of the models bill on usage (per resolution, per message) and two are effectively fixed (per seat, flat fee). At low volume the usage-based models are cheap because you barely use them; as volume climbs they overtake the fixed models. The crossover table makes that tipping point visible so you can plan the switch before the overage hits.

Can I embed this calculator on my site?

Yes — free. Copy the iframe snippet from the embed section below. The embedded version strips Chatbotscape navigation and keeps the calculator plus the attribution badge.

About this tool

Built and maintained by the Chatbotscape editorial team. We built the model to match the pricing structures we normalize when comparing platforms for review, so the four columns mirror how vendors actually invoice rather than a simplified average. The defaults are deliberately generic placeholders — the tool is a structure for yournumbers, not a quote. Found a billing structure these four models can't represent? Email corrections@chatbotscape.com and we'll review it.