Searching for Zendesk Answer Bot pricing now means checking Zendesk's AI agent pricing and the billing model attached to your account. A seat price alone does not tell you the full cost, but there is useful public information: Zendesk publishes its resolution tier definitions, included allowances and several base subscription prices.
Reviewed October 4, 2026 against Zendesk’s published pricing and product documentation. Examples below are planning exercises, not customer results or Zendesk quotes.
Table of Contents
- Check which billing model your account uses
- What the published resolution tiers mean
- Included allowances and base subscription prices
- A worksheet for 500 monthly conversations
- Compare the complete support setup
- Questions to ask before buying
Check Which Billing Model Your Account Uses
Do not assume every Zendesk customer has already moved to the same pricing model. Zendesk's documentation for the earlier automated resolutions platform says existing accounts remain on that platform until they move to resolution tiers. On the earlier model, the published baseline is 5 resolutions per paid agent per month for Team, 10 for Growth/Professional, and 15 for Enterprise, subject to a 10,000-resolution annual allocation cap. Legacy Support plans are an exception: they can access AI agents without an included resolution allocation.
Your dashboard helps identify the model. The traditional version has an Automated resolutions dashboard. The evolved version has Resolution usage and Resolution allowance dashboards; the latter is restricted to the account owner and billing admins. Zendesk explains the distinction in its account identification guide.
Before comparing quotes, record your exact plan, subscription term, billing currency and current model. Keep renewal terms separate from today's invoice. A screenshot of another company's dashboard is not a substitute for your own service order.
What the Published Resolution Tiers Mean
Zendesk introduced these definitions on May 18, 2026:
| Tier | What it describes | Allowance treatment |
|---|---|---|
| Assisted escalation | AI helped before a human completed the interaction. | Does not consume resolution allowance. |
| Contained resolution | The interaction ended without another assistance request, but did not pass Zendesk's LLM resolution verification. | Does not consume resolution allowance. |
| Verified resolution | The interaction passed verification that the customer's request was resolved. | The billable resolution category. |
These are published outcome definitions, not an undisclosed split between simple and complex questions. See Zendesk's automated resolution tier documentation.
The monetary rate is a separate question. Zendesk's upgrade guide explicitly labels prices in its sample invoices as placeholders and directs readers to sales for pricing details. Do not copy a screenshot's example amount into a budget as an actual rate.
For a trial, keep a small review set: straightforward questions, ambiguous requests and conversations needing a person. Record the answer, the outcome assigned in the dashboard and whether the customer actually received useful help. Billing classification and your own quality review answer different questions.
Included Allowances and Base Subscription Prices
Under the resolution allowance model, the allowance is a currency pool, rather than a fixed bucket of conversations. Zendesk publishes these monthly defaults per paid agent seat:
| Plan | Suite allowance | Support allowance |
|---|---|---|
| Team | US$2 | US$2 |
| Growth | US$5 | Not applicable |
| Professional | US$5 | US$2 |
| Enterprise | US$10 | US$2 |
| Enterprise Plus | US$10 | Not applicable |
Light agents do not contribute to the calculation. The default allowance has a US$5,000 annual cap; unused monthly allowance does not roll over. These figures apply to the allowance model, not the earlier count-based allocation. Source: Managing resolution allowances.
Separately, the public Zendesk pricing page displayed these US-dollar subscription prices when reviewed:
| Subscription or add-on | Per agent per month, paid yearly |
|---|---|
| Support Team | US$19 |
| Suite Team | US$55 |
| Suite Professional | US$115 |
| Suite Enterprise + Copilot | Contact sales |
| Copilot add-on | US$50 |
The page states that AI agents are included in Suite and Support plans. That does not make usage unlimited or give each plan identical channel capabilities. Copilot is the separate assistant for human support staff. Check add-on eligibility and your purchasing path before assuming a quoted package includes it.
Eligible sales-assisted accounts can buy more allowance through their billing controls. Managed or self-service accounts are directed to sales. The allowance purchase instructions explain who can make those changes.
For standard allowance accounts, the documented overage choice is to continue with additional billing or pause paid AI functionality at the limit. Forethought plans have a different restriction: their overage setting cannot be disabled. Review the applicable overage controls before launch.
A Worksheet for 500 Monthly Conversations
Consider an illustrative team with three paid Suite Team seats and 500 monthly conversations. This is a budgeting example, not a prediction of its automation rate.
Seat calculation: 3 × US$55 = US$165 per month equivalent, on annual billing. If all three seats are eligible for and purchase the US$50 Copilot add-on, that adds US$150, giving US$315 before other charges. Annual pricing is not a promise of monthly payment terms. These calculations use the published subscription prices.
Included allowance: Under the current currency model, three Suite Team seats provide 3 × US$2 = US$6 monthly allowance, based on the published defaults. Do not confuse this with six conversations.
Usage inputs: Suppose your measured period eventually contains 200 verified resolutions. That figure is an assumption for the worksheet. It is not a claim that a typical team will achieve that result. Obtain the applicable rate, prepaid allowance terms and overage rate before assigning it a dollar value.
Build your estimate with these separate inputs:
- Subscription commitment: seats, plan and billing term.
- Optional purchases: the add-ons the team actually needs.
- AI usage: measured billable outcomes and the applicable contracted rate.
- Allowance accounting: included funds, purchased funds and any additional usage charges, without counting the same prepaid usage twice.
- Other costs: channel usage, integration work, taxes and operational time where applicable.
Run the worksheet for an ordinary month and a busier month using your own traffic. Also test the business consequence of your spending limit: if automation pauses, who handles the waiting conversations? A budget cap is useful only when the support team can handle the resulting workload.
Compare the Complete Support Setup
The right comparison starts with the work your team needs to do. If you already operate Zendesk, include the effort of moving ticket history, routing rules and staff habits before treating another subscription as a saving. If you only need website self-service, test that narrower workflow before choosing a broader support suite.
Use the same questions and source material in each evaluation. Include a stale policy, a missing answer, a request requiring authentication and a handover to a person. Record setup time and unresolved cases alongside cost. Avoid treating an unanswered question as a successful resolution simply because the visitor left.
LoopReply is one option to evaluate for a website Agent using uploaded documents, CSV/XLSX data, crawled pages and configured PostgreSQL or MySQL snapshots. Its shared inbox supports human takeover, and workflows can connect to external services when the necessary endpoints and permissions are configured. Those capabilities are not a claim that every Zendesk channel or integration has a native LoopReply equivalent. See the product facts and knowledge source requirements.
Compare the current LoopReply plans and the allowances shown in your Workspace against your own usage. Message limits and resolution counts are different units; neither establishes which product is cheaper without a workload and a complete quote.
Test your own support questions
Configure a LoopReply Agent with your knowledge, check its answers and human handover, then review the plan allowances for your workload.
Questions to Ask Before Buying
Send these questions with your expected volume so the reply can become a usable comparison:
- Which billing model applies now, and what changes at renewal?
- What is included for our paid seat count and subscription term?
- What rate applies to additional usage, and does prepayment change it?
- Which requested channels, knowledge connections and staff tools require another purchase?
- Can we see a sample invoice using our assumptions, with placeholders clearly identified?
- Who receives usage alerts, and what happens operationally when the budget is exhausted?
Keep the written answers with your trial results. That gives you an estimate you can check against the first invoice, instead of relying on an assumed industry rate or an unsupported promise of savings.
Keep the price question separate from migration scope
Use this article for the billing model and allowance calculation. Use the LoopReply versus Zendesk workflow comparison to decide which support operations you would retain or replace. The distinction matters: reducing an Agent subscription does not remove the cost of a helpdesk you still need.
Compare the same workload with the Intercom seat-and-outcome example and the business buyer guide. Preserve the account quote and allowance dashboard alongside the calculation so a later plan change can be traced.