Forethought AI Pricing in 2026: The Full Guide

How Much Does Forethought AI Cost?

Forethought AI costs on average of about $74,483/year. On the low range it's around $35,670 and the high quote goes up to $151,400. Its three plans (Team, Professional, Enterprise) and its three add-ons all sit behind a “Get a Quote” button, with no free plan and no free trial. Buyers now purchase through Zendesk, which acquired Forethought in March 2026.

Pricing is a platform access fee plus outcome-based usage, so the bill tracks support volume, not headcount. Email, voice and Slack sit on the Professional tier, Forethought asks for 20,000+ historical tickets, and overage applies above your contracted band. Other conversational AI platforms price the same job very differently.

*Third-party benchmark data (Vendr), not Forethought list prices.

Who Sells Forethought Now, and Why It Changes Your Quote

Zendesk announced a definitive agreement to acquire Forethought on 11 March 2026, and confirmed the deal had closed on 26 March. It said the acquisition accelerated its own product roadmap by more than a year. Neither company disclosed a price, which fits Zendesk’s pattern: TechCrunch noted that the rare deals where Zendesk has published a figure were modest ones, $29.8 million for Zopim in 2014 and $45 million for BIME in 2015. Forethought had raised $115 million in total funding, and the deal was widely reported as Zendesk’s largest acquisition in nearly two decades. Anyone quoting a purchase price is guessing.

On 4 June 2026 the product went on sale as “Forethought AI agents by Zendesk.” Zendesk customers buy it as a Zendesk add-on. If you’re not a Zendesk customer, Zendesk says Forethought is still available to purchase and can be integrated with your platform across all supported channels.

So Forethought AI customer service pricing now runs through Zendesk sales rather than Forethought sales. Raise these on the first call:

• Whose paper the contract sits on, and which entity holds the service-level agreement (SLA)
• Whether the quote arrives bundled with Zendesk Suite licenses you don’t need
• How the outcome-based model gets reconciled with Zendesk’s own AI agent pricing over time

Zendesk has said publicly that existing Forethought customers can expect uninterrupted service, and that new customers can adopt the product with no requirement to use the Zendesk platform. Take that at face value, then ask for it in writing.

Nobody has confirmed the future of Forethought for Slack either way, so if Slack matters to your internal support, name it in the contract. Voice is worth pricing separately too, since Zendesk bills its own telephony product on its own terms; our breakdown of Zendesk Talk pricing shows what that side of the stack costs.

What the Zendesk Stack Underneath a Forethought Quote Costs

Forethought’s own price is invisible, but the platform it now attaches to publishes its. If your quote arrives bundled with Zendesk licenses, these are the numbers underneath it, per agent per month on annual billing.

Zendesk’s own AI agents changed shape in 2026. They’re now included in every plan with a small allowance of automated resolutions per agent per month, and metered per resolution above that allowance. Zendesk doesn’t publish the per-resolution rate; third-party teardowns cluster around $1.50 to $2.00, which is an estimate rather than a quote. Forethought sits on top of all of it as its own line item, which is why the seat price matters even when the AI is what you’re buying.

What Pricing Plans Does Forethought AI Offer?

Forethought AI pricing plans come in three tiers, all quote-only: Team, Professional and Enterprise. Forethought describes the model in its own pricing FAQ as a blend of platform access fees and an outcome-based pricing cost, with optional add-ons and a referral to the sales team for anything more specific. There’s no Forethought AI pricing per month at any tier and no per-seat rate; Forethought prices against the volume of conversations its AI handles.

According to TrustRadius’s 2026 B2B Buying Disconnect Report¹, transparent pricing has been software buyers’ number one request of vendors for four years running, since TrustRadius started asking in 2023. Forethought publishes none of it. You can read the entire Forethought AI customer service features and pricing matrix without ever seeing a number attached to it. Quote-only pricing is still common at the enterprise end of AI customer service, but it isn’t universal, and the comparison later in this guide shows how differently the same category can be priced.

The add-on lineup moved with the seller as well. Zendesk’s help center documentation for the product lists the add-ons as Assist, Discover, Agent QA and AI QA, a different set from the Multibrand, Analytics API and gap-detection add-ons still shown on Forethought’s own pricing page. Both lists are quote-only, and the Zendesk one is what a current quote gets built from.

Forethought Team: Chat and Mobile Deflection, Price on Request

The entry tier suits a single-brand support desk that wants AI agents answering chat and in-app messages, plus enough reporting to tell whether it’s working. Forethought doesn’t publish billing terms. Vendr’s procurement data for Forethought buyers indicates most contracts run as annual subscriptions billed monthly or annually, quoted either as a platform fee plus usage or as a flat annual fee against a defined ticket volume band.

What’s included, per Forethought’s plan card:

• AI agents for chat and mobile
• Autoflows and custom actions
• CSAT (customer satisfaction) collection and an insights dashboard
• Ready-to-use ticket triage models
• AI QA (quality assurance), capped at 1 rubric
• Brand management for up to 2 brands
• Standard security and standard support

Forethought Professional: Email, Voice and Slack, Price on Request

Forethought marks Professional as “Most popular,” and the channel matrix explains why. It’s the first tier where AI agents work on email, voice and Slack, which is the configuration most support organizations run. Pricing stays quote-only, and the outcome-based component means your invoice tracks the volume of conversations the AI resolves.

Everything in Team, plus:

• AI agents for email, voice and Slack
• AI-surfaced insights
• Multi-brand management for up to 20 brands
• Custom handoff models, up to 3
• AI QA, raised to 5 rubrics
• “Ask the expert” support

Forethought Enterprise: API Access and Governance, Price on Request

The top tier aims at multi-brand and regulated operations that need Forethought’s data inside their own systems, plus the security and governance controls a procurement review will ask for. It’s also the only tier with programmatic access to the resolution engine itself.

Everything in Professional, plus:

• Solve API
• Knowledge gap detection and AI article creation
• Autoflow gap detection and generation
• Analytics API
• Enterprise security and governance controls
• AI QA, raised to 20 rubrics, and up to 6 custom handoff models
• Advanced security and advanced support

Forethought AI Channel Availability by Plan

Two of the three Forethought AI plans hide your most common support channel behind an upgrade.

Before anyone discusses volume with you, your channel mix has already decided your tier.

What Will Forethought AI Actually Cost?

For the shape of the whole dataset: Vendr reports a median annual contract of $74,483 across 69 purchases, with observed contracts running from $35,670 to $151,400, and an average buyer saving of 22% off the opening number. Vendr is a SaaS procurement platform reporting what its own customers paid. Those numbers aren’t Forethought’s list price, and Vendr doesn’t claim they are; outside a sales call, nobody sees one.

• Scenario 1, step by step. Email decides the tier, so the quote starts at Professional rather than Team. Vendr’s smallest documented cohort, 5,000 to 15,000 tickets a month, lands at $30,000 to $70,000 a year. Add implementation at $10,000 to $50,000 and year one sits between $40,000 and $120,000. Check one thing first: a young company doing 8,000 tickets a month may still fail Forethought’s 20,000 historical-ticket requirement, and no budget line fixes that.

• Scenario 2, step by step. At 25,000 tickets a month you’re in Vendr’s mid-market band of $60,000 to $150,000 a year. Teams this size usually run a non-standard helpdesk or an internal system, so add a custom integration at $5,000 to $25,000 on top of implementation. Then add the escalator. Vendr reports annual uplifts of 3% to 7% written into contracts, so at 5% your year-two platform cost is $63,000 to $157,500 before your ticket volume grows at all.

• Scenario 3, step by step. This is where Forethought AI support automation pricing bends hardest. The ticket base matches Scenario 2 exactly, but Vendr puts Voice AI infrastructure work at $10,000 to $30,000 at deployment and the ongoing Voice AI add-on at $30,000 to $80,000 a year for 5,000 to 15,000 calls a month. Voice roughly doubles the cost of the same support operation. That’s the reason phone-first teams should price a dedicated voice platform alongside the quote instead of after it.

• Scenario 4, step by step. Above 50,000 tickets a month Vendr’s enterprise band starts at $150,000 and reaches past $300,000, and Enterprise becomes mandatory if you need the Solve API or the Multibrand add-on. You do get negotiating room back. This is the band where Vendr’s 22% average saving is most achievable, because you’re the size of account that gets a discount desk involved.

Two things hold across all four scenarios. The escalator compounds every year whether or not your volume grows, and the outcome-based component turns a good quarter for your business into a bigger invoice. For a sense of how differently a published per-seat model behaves at the same team sizes, our breakdown of how much call center software costs prices the same headcount without a quote.

Are There Hidden Costs in Forethought AI’s Pricing?

Forethought AI customer support pricing has a second layer underneath the license. Forethought confirms in its own FAQ that additional usage charges may apply if your usage exceeds the plan’s limits and purchase volume, without publishing a rate, and the rest of the cost stack sits in the implementation and infrastructure around the product.

The Data Floor That Decides Whether You Can Buy at All

Forethought’s FAQ states that its AI performs effectively with a historical ticket data volume of 20,000+ and requires at least 2,000 email or chat tickets per month to operate smoothly. That’s the vendor’s own number rather than a reviewer’s opinion, and it’s the most important line in this guide for anyone running a small support team. Teams below it that still want to automate customer service have to look at tools without a historical-data requirement.

Salesforce’s 2026 State of Service: AI Agents Edition² found that 72% of service operations professionals say data readiness is a major blocker to AI, against 59% of customer service leaders. The people closest to the data worry about it most, and Forethought has turned that worry into a published entry requirement.

The Gap Between Signing and Value

Forethought publishes an implementation window of 30 to 90 days. Set that against the same Salesforce report, which found that 70% of organizations with AI service agents observe measurable value within 60 days of deployment. A Forethought buyer can spend the entire industry time-to-value window still in setup, and that’s before the internal effort lands.

Forrester’s 2026 predictions for customer service³ put it plainly: 30% of enterprises will create parallel AI functions that mirror human service roles, including managers who onboard and coach AI agents and specialists who unblock them when they falter. Budget for those people alongside the platform, the same way you would for any deployment of AI in customer service that touches live customers.

What Do Buyers Say About Forethought AI’s Pricing?

What buyers can say publicly about Forethought AI’s pricing is limited, because Forethought publishes none of it. Two constraints follow from that. There’s no published number for anyone to react to, and we don’t reproduce review quotes, ratings or counts we haven’t verified ourselves. So nothing below is a paraphrased review. What follows is what the vendor’s own published terms and Vendr’s contract data structurally mean for a buyer. Forethought’s G2 review page is the place to check current sentiment for yourself.

• Transparency: you can’t build a business case without a call. All six purchasable items on Forethought’s pricing page carry a “Get a Quote” button and no number, and the vendor’s own answer to “How does pricing work?” is a referral to sales. For anyone comparing AI customer service solutions side by side, Forethought is the column of the spreadsheet you can’t fill in before a sales cycle, which is exactly the friction the TrustRadius data measures.

• Budget predictability: a good quarter costs you more. Outcome-based pricing means the invoice tracks the volume of conversations the AI resolves, and Vendr reports that contracts define a band with overage above it. Growth you’d normally celebrate arrives as an unbudgeted line item, unless you negotiate the overage rate and volume headroom before signing.

• Cost of entry: the license isn’t the bill. Forethought publishes a 30 to 90 day implementation window and paid support engineer services for complex setups. Vendr puts implementation at $10,000 to $50,000 and custom integrations at $5,000 to $25,000. Year-one total cost of ownership can run well above the subscription, which matters most for teams whose finance function approved the subscription figure alone.

• Who the model shuts out. The 20,000 historical ticket and 2,000 monthly ticket floor is a qualification gate published by the vendor, not a recommendation. No negotiation moves it, so a small support desk below the floor won’t get a quote at all, at any price.

• Upgrade pressure: the entry tier is narrower than it looks. Email, voice and Slack all sit on Professional, and QA rubrics step from 1 to 5 to 20 across tiers. Any team that answers customers by email, which is most of them, is on Professional from day one. Price the realistic tier, not the advertised entry point.

• Vendor relationship: you’re buying from Zendesk now. The commitments Zendesk made at acquisition are on the record and should be read at face value. What nobody has answered yet is how the outcome-based model will line up with Zendesk’s own AI agent pricing at renewal, or what happens to Forethought for Slack.

What Should You Ask For in a Forethought Quote?

Quote-only pricing cuts both ways for a Forethought buyer. You can’t budget without a sales call, but no list price anchors you either, and Vendr’s data shows buyers negotiate an average of 22% off. Ask for these before you sign:

1. Overage rate in writing: at or below your base per-unit rate.
2. Volume headroom: 20% to 30% built into the contracted band.
3. Escalator cap: a ceiling on the annual uplift, or its removal on a multi-year term.
4. Implementation fees: waived, or itemized and reduced.
5. Bundled pricing: across products rather than standalone line items, now that Zendesk holds the paper.
6. Auto-renewal: removed.
7. Runway: 60 to 90 days before your target go-live, since compressed timelines correlate with worse pricing in Vendr’s dataset.

Most buyers give away that last one by starting the conversation too late. If you’re building the evaluation from scratch, our call center software buyer’s guide covers the requirements questions that apply to any voice or AI vendor you shortlist.

Which Alternatives Are Better and Cheaper Than Forethought AI?

The three strongest alternatives to Forethought AI are CloudTalk for phone-first support and sales teams, Intercom Fin for per-resolution ticket deflection, and Freshdesk with Freddy AI for teams priced out of Forethought. All three publish what Forethought doesn’t. TrustRadius’s 2026 B2B Buying Disconnect Report¹ found that 83% of buyers shortlist three or fewer products, and a vendor who won’t name a price is easy to cut from a list that short. If you want a wider field, we also rank the best AI for customer service calls.

1. CloudTalk: Best for Phone-First Support and Sales Teams

What Is CloudTalk?

CloudTalk is the global AI business calling software built for SMB growth, trusted by 5,500+ businesses across 100+ countries. It covers virtual numbers in 160+ countries, IVR (interactive voice response), call queues, skill-based routing, call recording, the AI Sales Dialer, AI Conversation Intelligence and the AI Voice Agent (CeTe) for autonomous inbound and outbound conversations.

The category difference matters, because it decides whether CloudTalk belongs on your list at all. Forethought is a ticket-deflection platform for a helpdesk. CloudTalk is a calling platform with 100+ native integrations, including Zendesk, Freshdesk, Intercom, Help Scout, Front, Gorgias and Kustomer, so it sits alongside your helpdesk rather than replacing it. If your support and sales load runs through the phone, that’s the overlap worth pricing.

Why Is CloudTalk Better Than Forethought AI?

• You can build a budget without a sales call, because plan and add-on prices are published. That is the one thing buyers have asked vendors for more than anything else, four years running.
• There’s no data floor. Forethought publishes a 20,000 historical ticket and 2,000 ticket per month requirement; CloudTalk publishes no equivalent gate, so a 12-person team can start.
• Voice is the core product instead of a tier-gated add-on. Vendr’s data puts Forethought’s Voice AI at $30,000 to $80,000 a year on top of the base for moderate call volume.
• Agents get help on live calls. AI Agent Assist surfaces answers and next steps during the conversation rather than after it.
• You can test it yourself. CloudTalk runs a 14-day free trial instead of a vendor-led Proof of Value engagement.
• Time to value is shorter. Forethought publishes a 30 to 90 day implementation window, which eats most of the 60-day mark Salesforce found for measurable AI value.

What Does CloudTalk Cost?

• Starter: $25/user/month, billed annually. Includes IVR and skill-based call routing, the Call Flow Designer, unlimited call queues and the desktop and mobile apps.
• Essential: $29/user/month. Adds the Smart Dialer, the 100+ integration catalog, API and data sync, workflow automation and unlimited call recording storage.
• Expert: $49/user/month. Adds the Salesforce integration, the Power Dialer, wallboards, real-time call monitoring and advanced analytics.
• Custom: contact sales for enterprise SLAs, dedicated support and specific compliance arrangements.
• Add-ons: AI Voice Agents are always an add-on, starting at $99/month for 200 minutes. AI Conversation Intelligence is always an add-on at $9 per user/month billed annually. Power Dialer is included on Expert and sold as a $15 per user/month add-on on Starter or Essential. Parallel Dialer is an add-on at $39 per user/month.
• Free trial: 14 days.

What Is CloudTalk’s Rating on G2?

CloudTalk’s current G2 rating is 4.4/5 from 1,702+ reviews. Reviewers most often point to call quality, integration breadth and fast setup, the categories that matter most when voice is the channel carrying your support load.

2. Intercom Fin: Best for Ticket Deflection With a Published Per-Resolution Price

What Is Intercom Fin?

Fin is Intercom’s AI agent, built to resolve customer conversations without a human. It runs inside Intercom, and it also runs standalone on top of an existing helpdesk, so you can put it in front of your current ticket queue without buying Intercom seats. Teams that keep phone support separate often connect it through CloudTalk’s Intercom integration so calls land against the same customer record.

Why Is Intercom Fin Better Than Forethought AI?

Fin gives you the cleanest transparency contrast available. Intercom publishes the exact per-outcome price Forethought won’t disclose, defines in writing what counts as a billable outcome, and states plainly that no contract is required on its self-serve plans. Forethought asks you to enter a sales cycle before you learn the shape of the bill. Fin lets you model it in a spreadsheet in ten minutes.

What Does Intercom Fin Cost?

• Fin AI Agent: from $0.99 per outcome, on every plan.
• Fin standalone on your existing helpdesk: $0.99 per outcome with no seat costs and no setup, integration or platform fees, subject to a minimum monthly commitment.
• Intercom seats: Essential $29 per seat/month, Advanced $85 per seat/month, Expert $132 per seat/month.
• Copilot add-on: $29 per agent/month billed annually, or $35 billed monthly, with unlimited usage.
• Other add-ons: Proactive Support Plus at $99/month including 500 messages; the Pro add-on at $99/month including analysis of 1,000 conversations and 2,000 credits, with extra credits at $0.01 each.
• Contract required: no. Essential and Advanced sell self-serve, monthly or annual.

3. Freshdesk With Freddy AI: Best for Teams Priced Out of Forethought

What Is Freshdesk With Freddy AI?

Freshdesk is Freshworks’ helpdesk, and Freddy AI is the AI layer on top of it: an AI agent for customer-facing resolution, a copilot for human agents, and an insights product for analytics. Freshworks sells all of it on published per-agent pricing with a published AI session rate. If calls are part of the picture, CloudTalk’s Freshdesk integration logs them into the same ticket.

Why Is Freshdesk Better Than Forethought AI?

It answers the affordability question Forethought can’t. Forethought publishes a 20,000 historical ticket floor, and Vendr’s lowest documented volume band starts around $30,000 a year. Freshdesk starts at $19 per agent per month with 500 Freddy AI Agent sessions included, which is a different order of commitment for a team of eight.

What Does Freshdesk Cost?

• Growth: $19 per agent/month, billed annually. Freddy AI Agent with the first 500 sessions included.
• Pro: $55 per agent/month, billed annually. Same AI agent allowance.
• Enterprise: $89 per agent/month, billed annually. Adds Freddy AI Insights.
• Additional AI Agent sessions: $49 per 100 sessions.
• Freddy AI Copilot: $29 per agent/month billed annually, on Pro and Enterprise.
• Day passes: $2 on Growth, $7 on Pro, $12 on Enterprise.
• Connector app tasks: $80 per 5,000 tasks.

How Does Forethought AI Pricing Compare With the Alternatives?

One note for anyone who wants Forethought’s technology on a published plan price: Zendesk’s own AI agents come included in all Suite and Support plans with a small monthly allowance of automated resolutions, then meter per resolution above it, while “Forethought AI agents by Zendesk” stays a separate quote-based add-on on top of a seat price that runs $19 to $115 per agent per month. The published Zendesk stack is the floor under any Forethought quote.

Is Forethought AI Worth It, and Who Should Buy It?

Forethought is a serious product with a real qualification gate, and most of the disappointment around it comes from teams who were never in its market. Salesforce’s 2026 State of Service: AI Agents Edition² found that 66% of customer service organizations now use agentic AI, up from 39% a year earlier, so automating resolution is table stakes. The question is whether this vendor’s cost structure fits your volume.

One last frame for the business case. Deloitte’s State of AI in the Enterprise report⁴ surveyed 3,235 senior leaders in late 2025 and found that while 66% of organizations report productivity and efficiency gains from enterprise AI, only 40% report reducing costs. A Forethought AI subscription justified purely on headcount savings is the version of this deal most likely to disappoint at renewal. Build the case on resolution quality and response time instead, and hold the cost argument to what you can measure in the first 90 days.

Sources:

1. TrustRadius
2. Salesforce
3. Forrester Research
4. Deloitte