Zanus AI's public Front Office pricing page says its solutions are priced per company rather than per seat and offers a seven-day sandbox trial without a credit card. The actual price list loads dynamically, so this guide does not quote a number. Check the live page and ask for the terms that apply to your account before comparing a plan.

The more useful comparison is the cost of the complete phone workflow: calls, usage, included capabilities, implementation, support and the business actions the agent is expected to complete.

Scope: This article summarizes public Zanus product information checked on 7 October 2026. It is a buyer checklist, not an independent product test or endorsement. Pricing, features and plan contents can change.

What the public pricing page says

Zanus presents Front Office as a company-level subscription, not a per-seat plan, and says its trial runs for seven days without a credit card. The price page also notes that Back Office pricing is being finalized and that on-premises AI servers are quoted separately. These details apply to the public page at the time checked; confirm the exact product, scope and terms in your own quote.

The price list is loaded live into the page. Because the amounts may vary by selected solution and change over time, use the current Zanus Front Office pricing page rather than relying on an old screenshot or a third-party article.

Compare the same call workload

Before asking for a quote, write down the expected monthly workload. Include inbound and outbound calls, average connected duration, peak concurrency, phone-number requirements, languages, transfers and any planned automated campaigns. Then ask which of these are included, usage-based or separately priced.

Zanus's AI Phone page describes answering calls, transfers, call records and outbound Auto-Call. It also says that some Auto-Call scheduling is included inside a solution. Treat these as vendor descriptions, then confirm which capabilities and limits apply to the exact plan you are considering.

Cost areaAsk the vendorRecord for your comparison
SubscriptionWhich Front Office solution and company scope does the quote cover?Current quote, billing term and renewal basis
UsageHow are connected minutes, outbound attempts and peak concurrency counted?Expected and high-usage scenarios
TelephonyAre numbers, transfers and any carrier charges included?Number count, regions and transfer paths
CapabilitiesWhich phone, booking, CRM and Auto-Call actions are enabled?Included features and any add-ons
People and supportWhich team access, onboarding and support are included?Named users, service scope and response terms
Change and exitWhat happens when usage, needs or the subscription changes?Overage, cancellation and data-export terms

Use the trial to test an outcome

If you try the sandbox, choose one real but low-risk task and define what counts as success before testing. For example, ask the agent to capture a new enquiry, offer an available appointment, and confirm the booking only after the calendar has accepted it. Also test an unavailable time, an unclear answer and a transfer request.

After each call, inspect the system of record. Did the appointment exist? Was the right contact created? Did a failed action become a clear follow-up? A natural-sounding call is not proof that a business task completed.

Keep a short log with the test date, configuration, transcript, final record and any human work needed to complete the task. That gives you a repeatable basis for comparing a trial with another vendor or with your current process.

Calculate cost per completed task

Once the quote and trial assumptions are clear, estimate the monthly cost for the same workload across every option. Include subscription, usage, telephony, setup, integration work and ongoing human review. Divide by the number of correctly completed outcomes, not by the number of attempted calls alone.

This does not predict savings. It gives your team a like-for-like method for checking whether a plan fits the work it is expected to handle. Run low, typical and peak-volume cases; confirm every assumption with the vendor.

For a broader view of the operating model, compare the vendor details with Sysevo's voice agents, caller memory, CRM and business platform. Sysevo and Zanus describe different product scopes; use your call scenarios and current written quotes to determine fit.

Mapping the Phone-Agent Workflow Before You Evaluate Any Quote

A pricing comparison for a business phone agent depends heavily on how clearly a business has documented its own workflow before any vendor conversation begins. Without that documentation, two quotes can appear to describe the same service while covering entirely different scopes of work. The exercise below is designed to produce the internal clarity that makes any vendor comparison—including one involving Zanus AI—genuinely actionable rather than superficially numerical.

Begin by drawing the full path a call takes today, from the moment a phone rings to the moment a business outcome is recorded. That path typically involves more steps than teams initially count: greeting, intent identification, data collection, verification against a calendar or record, confirmation to the caller, creation or update of a system record, any follow-up message, and a handoff note if a person needs to continue the work. Each of those steps is a potential scope boundary in a vendor quote, and a capable buyer should know which steps they expect the agent to own before reading a price.

Distinguishing the Call from the Task

One of the more consequential distinctions in evaluating a phone agent is separating call volume from task volume. A call is a connected audio session. A task is a defined business outcome with a verifiable result—a confirmed appointment in a calendar, a contact record with correct details, a support request with enough context for a staff member to act. These are not the same unit, and pricing models may reflect one, the other, or neither cleanly.

Consider an illustrative scenario. A fictional home services business receives calls that divide roughly into new enquiries, existing customer changes, supplier queries and wrong numbers. Only the first two categories result in a task the agent might be expected to complete. If a quote is structured around total call volume, the business is effectively paying for agent handling of calls that require no business action. If the quote is structured around completed tasks only, the business needs to define what counts as completion and what happens to calls that do not reach that state.

Neither model is inherently better. What matters is that the evaluation team defines task types before the vendor conversation, assigns each a clear completion condition, and asks each vendor to price the full set. Without this, a comparison becomes a comparison of summaries rather than of scoped services.

Identifying Who Owns Each Step

Phone-agent workflows involve at least three distinct parties whose roles are worth separating explicitly: the caller, the business, and the agent. A caller presents a request and provides information. The agent verifies, routes and records. The business—meaning a staff member—owns any decision that cannot be resolved through the agent's configured logic. Blurring these roles in evaluation creates risk later.

A practical way to surface ownership gaps is to walk through a set of exception cases before the trial begins. What happens when a caller provides a name that partially matches an existing record? What happens when the requested appointment slot is unavailable and no alternative is within the caller's stated constraints? What happens when a caller asks a question the agent cannot answer and the relevant staff member is unavailable? Each of these cases has a correct resolution—but the resolution may involve human judgment rather than agent logic, and the evaluation should test whether that handoff is clean.

Staff who receive transferred calls or follow-up tasks are a useful but sometimes overlooked source of evaluation feedback. They experience the downstream quality of an agent interaction: whether context arrived with the transfer, whether the record was created correctly, whether the caller was appropriately set up to continue the conversation without repetition. Collecting structured feedback from those staff members during a trial period adds a perspective that call transcripts alone do not provide.

A Sample Evaluation Record for a Single Call Type

The following format illustrates how a team might document one call type for evaluation purposes. It is presented as a structure, not as a claim about any vendor's output format.

  • Call type label: New appointment request, no existing record
  • Trigger condition: Caller provides a name not found in the system
  • Expected agent actions: Collect name, contact number, service type, preferred time; check calendar availability; offer available slot; confirm with caller; create contact record; create appointment; send confirmation message
  • Completion condition: Appointment exists in calendar with correct time, service and contact detail; caller received confirmation; no staff action required to complete the booking
  • Exception cases to test: No availability in preferred window; caller corrects a detail mid-call; connection drops after agent confirms but before record is written
  • Evaluation evidence to inspect: Calendar entry, contact record, outbound message delivery log, call transcript, any follow-up task created for staff

Repeating this structure for three to five call types before a trial begins gives the evaluation team observable, consistent criteria rather than subjective impressions of how natural the agent sounded.

Questions That Should Change a Purchase Decision

Some questions in a vendor evaluation are clarifying. Others are genuinely decision-altering. The following are worth treating as the second kind, because the answers directly affect whether a phone-agent subscription fits the work it will be asked to handle.

  • If the agent cannot complete a task—because a system is unavailable or a caller's request is outside configured scope—what does the caller experience, and what record is created for staff?
  • If the business changes its service area, adds a calendar, or restructures its call routing, does that change the price, require a new configuration engagement, or both?
  • If call volume is significantly lower than forecast in the first months, can the subscription be adjusted without forfeiting setup investment?
  • If the vendor changes the capabilities included in a plan after contract signing, what is the notice period and the business's remedy?
  • When the subscription ends, in what format and timeframe can the business retrieve call records, transcripts, contact data and appointment history?

These questions are not designed to find fault with any specific vendor. They are worth asking of every option because the answers reveal how a vendor thinks about the ongoing operating relationship rather than the initial sale.

Handover Arrangements Worth Specifying in Writing

Before signing any agreement, a business evaluation should specify in writing how the agent handover process works for calls that exceed its scope. This includes the trigger conditions for a transfer, the information passed to the receiving staff member, the fallback if no staff member is available during business hours, and the after-hours handling arrangement. These details affect caller experience directly and are often not fully addressed in a public pricing or capability description.

If a staff member receives a transferred call, the quality of the handover—context, caller name, reason for transfer, any information already collected—determines whether the agent added value or simply delayed the human interaction. That quality is testable during a trial and worth recording as a named evaluation criterion alongside the technical completion of a booking or enquiry record.

Documenting these handover specifications also provides a useful baseline for the renewal review. If handover quality degrades after a configuration change or a product update, the written specification gives the business a concrete reference point for a conversation with the vendor rather than a vague sense that something has changed.

The most durable outcome of a structured phone-agent evaluation is a written record that covers the original workload assumptions, the trial results against defined criteria, the final quote with its scope clearly marked, and the open questions answered by the vendor in writing—giving any team member who inherits the account a clear basis for the next renewal decision without starting the assessment from scratch.

Sources

  • Zanus Front Office pricing, official pricing page, checked 7 October 2026. The live price list is dynamically loaded; confirm current amounts and terms on the page.
  • Zanus AI Phone, official capability description, checked 7 October 2026.

Why a headline Zanus price is not enough

For a business phone agent, a quote should make clear what is included in the subscription and which costs vary with use. Ask whether the quoted amount covers the business phone service, AI receptionist, calling minutes, phone numbers, CRM records, appointment scheduling, SMS, support and onboarding. If the price is described per company, per location, per user or per minute, write down that unit beside the number. Those models answer different questions and cannot be compared by looking at the monthly total alone.

Ask for a sample invoice using your expected workload. It should show the recurring platform charge, any included usage, what happens at the limit, taxes, one-time setup and any optional services. Confirm whether the same terms apply to all locations and whether an additional line or department changes the bill. If a free trial is offered, get the trial duration, included usage, conversion date, cancellation method and whether a payment method is required in writing.

Zanus’s public pricing page is dynamically presented, so a buyer should verify the live amount and its billing basis directly before relying on a screenshot or an older article. Record the date and currency of the quote. If an item is not publicly stated, treat it as unknown rather than filling it in from a different vendor’s pricing model.

Build a like-for-like phone-agent comparison

Start with a common workload that reflects your business: monthly inbound calls, after-hours share, average connected length, routine questions, booking requests, transfers and calls requiring staff follow-up. Use the same workload for Zanus and every alternative. Then ask each vendor to price that case and explain any exclusions. This avoids comparing one vendor’s base plan with another vendor’s higher-capacity bundle.

Quote itemWhat to recordWhy it matters
Billing unitCompany, location, seat, minute or completed actionShows what makes the bill grow.
Included usageMinutes, calls, numbers, locations or actionsDefines the real allowance for your workload.
OverageRate, increments, alerts, cap and pause behaviorPrevents surprise usage and service interruption.
TelephonyNumbers, inbound, outbound, transfers and destinationsSeparates carrier cost from AI subscription.
Setup and supportIncluded onboarding, integration work and response termsSurfaces deployment effort and ongoing help.

Do not compare only “cost per call.” Calls differ in duration and outcome. A short call that captures a name is not equivalent to a confirmed appointment with a CRM record and reminder. Calculate monthly cost per valid completed task, and report the task definition beside the result.

Use a trial to check a complete customer journey

Before the trial starts, choose one narrow task and prepare a small test set. Include the normal request, an unavailable appointment, a caller with a correction, an unclear question, a transfer request, and a failed external action. Ask the agent to identify what it can verify, collect only necessary details, and hand off when the task is outside its allowed scope.

After every call, inspect the system of record. Verify the appointment time zone, contact details, service, notes and confirmation. If the agent says it sent a message, check the actual delivery status. If it transfers the call, ask the receiving employee whether the context was sufficient to continue without making the caller repeat the entire story. Keep a log of each defect and whether it was fixed by configuration, data cleanup or a product limitation.

Use a second person to review a sample of calls without seeing the configuration changes made by the builder. This reduces the chance that the team declares success because it already knows what the agent was supposed to mean. Ask a few real employees to test the handoff experience too; they will notice missing context and confusing follow-up work that may not be visible in the caller transcript.

Calculate the business case without promising savings

Estimate the value of the work the agent may complete, but keep the assumptions separate. For a call-capture workflow, measure calls that currently go unanswered, the share that are qualified enquiries, the share that become customers, and the value or gross margin of those customers. For a booking workflow, measure confirmed appointments and attendance rather than counting every scheduled slot as revenue.

Compare that value with the full cost: subscription, telephony, implementation, integration maintenance, human review, failed calls and follow-up. Use three cases—conservative, expected and high volume—and change only one or two assumptions at a time. The result is a sensitivity analysis, not a revenue promise. It helps show which unknown deserves a real pilot measurement.

Keep a short procurement record with the quote, source URL, date checked, included allowances and written answers to your questions. Recheck the live Zanus pricing page and AI Phone overview before approval. A structured comparison is more useful than an unsupported “cheapest” claim, and it remains useful if a vendor changes its plans.

A quote request you can send to a vendor

Ask Zanus to price one documented workload, not a vague estimate. Provide the number of locations, current phone setup, monthly calls, average duration, peak simultaneous calls, expected bookings or enquiries, and the human handoff route. Request the recurring price, billing unit, included minutes or calls, overage rate, phone-number costs, setup, integrations, support and contract terms. Ask the vendor to show the total for a normal month and a peak month.

Request a written explanation for the statement “per company” or “per account.” Confirm whether a second location, phone number, team or calendar changes the amount. Ask whether the trial has usage limits, whether data is retained after the trial, how to cancel, and whether the quote remains valid after any promotional period. Save the answer and date alongside the live pricing page.

Use the same request with each competing vendor. If one vendor cannot price a particular item until discovery, mark it unknown and keep it out of the headline comparison. Do not treat unknown as zero. A transparent quote table will make the real difference between an integrated business phone system and a usage-priced agent easier to see.

Set a renewal checkpoint

Pricing comparisons should continue after launch. Put a 30-, 60- or 90-day review on the calendar and compare actual invoices with the original workload assumptions. Check whether call mix, duration, locations or staff follow-up changed. If usage is above forecast, identify the cause before moving to a larger package. If usage is below forecast, ask whether the plan can be resized without losing required features.

Use that review to decide whether to expand, adjust routing or keep the service limited. Document any vendor changes to price, terms or feature availability before renewal. This makes the budget a living record rather than a one-time sales estimate.

Related buying guides