AI Summary
About
Front is a customer operations platform built around a shared inbox: it unifies email, live chat, SMS, WhatsApp, Slack, and social channels into a single collaborative ticketing workspace for support, sales, and account-management teams. On top of the core inbox and workflow-automation layer, Front has layered a set of “Front AI” products — Copilot (an always-on assistant for human agents), Autopilot (an omnichannel AI agent that can triage, hand off, or fully resolve conversations), Smart QA, and Smart CSAT — that are sold both bundled into its top plan and as metered add-ons on lower tiers.
Front markets itself as an all-in-one alternative to point tools for shared inboxes, ticketing, live chat, and knowledge bases, and says it is trusted by 9,300+ businesses, including Uber Freight, ClickUp, Culture Amp, Navan, Branch, and Baggu. It competes most directly with Zendesk, Intercom, Freshdesk, Help Scout, Hiver, Missive, and Pylon — a comparison set Front calls out explicitly in its own site navigation.
Front is a Y Combinator (S14) company. It raised a $65M Series D led by Salesforce Ventures and Battery Ventures on 28 June 2022 at a $1.7B valuation, bringing total funding to over $200M, and third-party trackers put its annual recurring revenue in the $100M range — which makes it a mid-scale challenger rather than a Zendesk-sized incumbent, and helps explain why its pricing moves are packaging-led rather than volume-led.
For the most current information on Front’s pricing and market position, visit Front.
Pricing summary : How Front’s pricing model works
Front’s core plans (Starter $25, Professional $65, Enterprise $105, all per seat/mo billed annually) are a straightforward seat-based ladder gated by channel count, automation-rule limits, and workspace/SSO features. Layered on top is a family of AI add-ons priced on two different mechanics: Copilot, Smart QA, and Smart CSAT are flat per-seat monthly add-ons ($20, $20, and $10/seat/mo respectively), while Autopilot is pure outcome-based usage — billed per conversation, at one of three prices depending on how far the AI agent got (Triage $0.05, Handoff $0.39, Resolution $0.89), and only ever charged once per conversation at the highest level reached.
- Seats — the core unit for Starter/Professional/Enterprise plans and for Copilot/Smart QA/Smart CSAT add-ons. See how the wider corpus uses per-seat pricing.
- Conversations (outcome-based) — Autopilot’s meter; priced by automation level, not by time or token volume. This puts Front in the outcome-based pricing cohort alongside Intercom’s Fin.
- Billing cadence — monthly or annual, both paid upfront; annual saves roughly 24% on the core plans (Enterprise is annual-only).
- Metered overages — a $200 per +100 requests/min API rate-limit increase, and Meta-billed WhatsApp costs plus a 20% admin fee for the Native WhatsApp Channel add-on.
What makes this different: most seat-based helpdesk vendors sell their AI layer as one flat per-seat upcharge; Front instead splits its AI portfolio in two — assistive/QA features (Copilot, Smart QA, Smart CSAT) stay per-seat, while its most autonomous product (Autopilot) is priced purely on delivered outcomes, so a team doing high-volume automation pays roughly in proportion to what Autopilot actually resolves rather than how many seats it has.
Pricing by product
Core plans
| Tier | Price | Included | Key mechanics |
|---|---|---|---|
| Starter | $25/seat/mo billed annually ($35/seat/mo billed monthly); up to 10 seats | Shared inbox & ticketing, single channel type (email, chat, or SMS); AI Topics + up to 10 automation rules; basic analytics; no-code public knowledge base | Self-serve “Try for Free”; one-license minimum; single channel type only |
| Professional | $65/seat/mo billed annually ($85/seat/mo billed monthly); up to 50 seats | Everything in Starter, plus omnichannel (email, SMS, social, etc.); macros + up to 20 automation rules; advanced analytics; multiple workspaces, SSO & SCIM | Sales-led “Request a Demo”; the plan a 14-day free trial runs on (capped at 15 teammates, 5 workspaces, 30 channels, 5,000 messages) |
| Enterprise | $105/seat/mo, billed annually only (“Only available on annual”) | Everything in Professional, plus smart rules, unlimited rules/macros, multi-language knowledge base, custom roles & permissions; Copilot, Smart QA, and Smart CSAT bundled at no extra cost | Badged “MOST POPULAR” on Front’s own pricing page; sales-led, uncapped seats; no license minimum disclosed (sales-quoted) |
AI add-ons (Autopilot, Copilot, Smart QA, Smart CSAT)
| Add-on | Price | Included | Key mechanics |
|---|---|---|---|
| Autopilot | Outcome-based: $0.05 (Triage), $0.39 (Handoff), $0.89 (Resolution) per conversation | Omnichannel AI agent that triages, hands off to a teammate, or fully resolves a conversation end-to-end via Playbooks | Charged once per conversation at the highest automation level reached; pay-as-you-go, invoiced monthly; not charged if a teammate cancels the automation, the customer asks for a human, or Autopilot has a technical failure; 30-day free trial with up to 1,000 Resolutions |
| Copilot | $20/seat/mo as an add-on on Starter/Professional (and legacy Growth/Scale plans) | Always-on AI assistant: summarizes conversations, drafts replies, surfaces relevant history and knowledge | Included at no additional cost with the latest Enterprise plan; rate-limited to 200 drafting requests/user/day |
| Smart QA | $20/seat/mo as an add-on on Starter/Professional (and legacy Growth/Scale plans) | Automated, AI-powered QA scoring of agent conversations | Included at no additional cost with the latest Enterprise plan; bundled with Smart CSAT for $25/seat/mo total; 30-day free trial |
| Smart CSAT | $10/seat/mo as an add-on on Starter/Professional (and legacy Growth/Scale plans) | AI-inferred CSAT scores to supplement low customer-survey response rates | Included at no additional cost with the latest Enterprise plan; ~90% alignment with customer-submitted CSAT per Front’s own reporting; 14-day free trial; not retroactive to past conversations |
Other add-ons & usage fees
| Add-on | Price | Included | Key mechanics |
|---|---|---|---|
| Native WhatsApp Channel | Meta-billed WhatsApp conversation costs, plus a 20% admin fee | Direct connection of WhatsApp Business numbers to Front | Available on Professional and Enterprise only; sales-led (“Talk to Sales”) |
| API Rate Limit Increase | $200 per +100 API requests/min per month | Raises the default API throughput ceiling for complex workflows | Available on all plans; sales-led |
| Success Services / onboarding | Quoted (Professional Onboarding, Enterprise Onboarding, and Success Services project-hour packages) | Guided setup, solution design, platform optimization, training/change management | An onboarding package is required for contracts over $25k |
Sales motions across products: PLG / self-serve for Starter; sales-led for Professional, Enterprise, and all AI and usage add-ons.
Hidden costs : What Front users actually pay
Front’s headline per-seat number is the smallest part of a mature bill. Three things inflate it: the AI add-ons are per-seat and stack on top of the plan, Autopilot is uncapped pay-as-you-go usage, and the plans carry hard seat ceilings that force a tier jump rather than an overage charge.
Archetype 1 — a 12-seat Professional team running the full AI stack
A 12-person omnichannel support team on Professional, with Copilot and the Smart QA + Smart CSAT bundle enabled for everyone, and Autopilot touching about 5,600 conversations a month (3,500 triaged only, 900 escalated to a human handoff, 1,200 fully resolved):
| Line item | Monthly cost |
|---|---|
| Professional plan — 12 seats × $65/seat (annual rate) | $780.00 |
| Copilot add-on — 12 seats × $20/seat | $240.00 |
| Smart QA + Smart CSAT bundle — 12 seats × $25/seat | $300.00 |
| Autopilot Triage — 3,500 conversations × $0.05 | $175.00 |
| Autopilot Handoff — 900 conversations × $0.39 | $351.00 |
| Autopilot Resolution — 1,200 conversations × $0.89 | $1,068.00 |
| Estimated total | $2,914.00 |
That is an effective $242.83 per seat per month — 3.7× the $65 headline — and Autopilot alone ($1,594) is 55% of the bill. Two secondary lessons hide in the table. First, this team is over-paying for its own packaging: the three per-seat AI add-ons cost $45/seat on Professional, so it is spending $110/seat while Enterprise bundles all three for $105/seat and adds smart rules, unlimited macros, custom roles, and a multi-language knowledge base. Second, the Autopilot line is the only one that moves with demand — the seat lines are fixed all year, which is exactly the cost-unpredictability profile that catches finance teams out when a seasonal volume spike lands.
Archetype 2 — the eleventh hire (the Starter seat cliff)
Front does not sell overage seats. Starter is capped at 10 seats and a single channel type; Professional is capped at 50. When a team outgrows a cap, the whole account re-prices at the next tier:
| Line item | Monthly cost |
|---|---|
| Starter — 10 seats × $25/seat (plan ceiling reached) | $250.00 |
| Hire #11 arrives — Starter cannot hold an 11th seat | — |
| Professional — 11 seats × $65/seat (forced tier move) | $715.00 |
| Net change for one hire | +$465.00/mo (+186%) |
The same cliff repeats at the top of Professional: 50 seats costs $3,250/mo, the 51st forces Enterprise at 51 × $105 = $5,355/mo, a 65% jump. And because Front’s billing documentation states that payments are nonrefundable and that removing a license produces “no refund or credit for that license”, a team that over-buys seats in anticipation of growth carries the cost to the end of the term. Downgrades — seat decreases, plan changes, add-on removal — only take effect at renewal.
Two further line items are easy to miss when modelling: the Native WhatsApp Channel bills Meta’s messaging costs plus a 20% admin fee rather than a fixed rate, and an API Rate Limit Increase costs $200 per additional 100 requests/min per month. Front also requires a paid onboarding package on any contract over $25k.
Want to estimate your own Front bill? Use the Front pricing calculator to model your monthly cost based on seat count, AI add-ons, and Autopilot conversation mix.
Pricing evolution : Front pricing history and changes
Cadence
| Quarter | Price changes | Product / SKU additions | Notes |
|---|---|---|---|
| 2021 Q3 | 0 | 0 | Baseline: Starter $19, Prime $49 per person/mo billed annually; Enterprise contact-sales only. |
| 2021 Q4 | 0 | 0 | Seat minimums published for the first time — 2 users on Starter, 5 on Prime. |
| 2022 Q2 | 1 | 0 | Enterprise gets a public price at $99/seat/mo (10-user minimum) and the MOST POPULAR badge; Starter capped at a 10-user maximum. |
| 2023 Q1 | 1 | 3 | Prime → Growth and Enterprise → Scale rename with no price movement; Premier added at $229/seat/mo (50-seat minimum) and the Scale minimum doubled from 10 to 20 seats; first paid channel add-ons — Telegram and Dialpad SMS at $100/mo each, Scale only. |
| 2023 Q2 | 1 | 0 | Growth raised $49 → $59/seat/mo (+20%); Scale and Premier unchanged. |
| 2023 Q3 | 0 | 1 | Google Business Messages joins the $100/mo channel add-on rail. |
| 2023 Q4 | 1 | 0 | Growth seat minimum cut 5 → 2; automation-rule ceilings published on the page (20 rules Growth, 200/workspace Scale). |
| 2024 Q2 | 0 | 1 | AI Answers launches at $0.70 per resolution — Front’s first outcome-priced SKU; page redesigned to a four-tier grid with a monthly/annual toggle headlined “Save up to 34%“. |
| 2024 Q4 | 1 | 2 | Premier’s $229 list price replaced by “Contact Us”; premium add-on rail lists WhatsApp at $70 per 1,000 conversations and Dialpad SMS at $100/mo; Front for Startups offers up to 6 months free. |
| 2025 Q2 | 0 | 1 | Smart QA ships at $20/seat/mo, having been listed COMING SOON in the March 2025 snapshot. Premium add-ons remain restricted to Growth plan or higher. |
| 2025 Q3 | 4 | 2 | The repackage. Four tiers become three: Starter $19 → $25 (+32%), Growth $59 → Professional $65 (+10%), Scale $99 → Enterprise $105 (+6%), Premier retired. Native WhatsApp moves from a flat $70/1,000 conversations to Meta-billed cost + 20% admin fee; an API Rate Limit Increase add-on appears at $200 per +100 requests/min/mo. |
| 2025 Q4 | 1 | 1 | AI Answers retired — its help-center article now reads “AI Answers is no longer available for purchase” — and Autopilot takes over the outcome meter at $0.89 per resolution. |
| 2026 Q1 | 1 | 0 | Autopilot’s published rate is pulled from the pricing page, leaving a “Talk to Sales” card with no visible price. |
| 2026 Q2 | 1 | 3 | Autopilot Playbooks, Autopilot Resolve, and a rebuilt Copilot announced 2026-04-28; the full three-level outcome ladder — Triage $0.05 / Handoff $0.39 / Resolution $0.89 — returns to the pricing page. |
| 2026 Q3 | 0 | 0 | Stable at Starter $25 / Professional $65 / Enterprise $105; annual-discount headline now reads SAVE 24%, down from SAVE 34%. |
Tracked range: July 2021 – September 2026. Quarters not listed above showed no price change and no SKU addition in the snapshots reviewed. Coverage is thinnest between June 2022 and December 2022, where no snapshot was captured — any change inside that window is dated to the first snapshot in which it appears.
Notable changes
- 2022-04 — Enterprise stops being a “Contact Sales” card and lists at $99/seat/mo with a 10-user minimum, making all three tiers publicly priced for the first time (front.com/pricing, archived).
- 2023-01 — Prime and Enterprise are renamed Growth and Scale with no price movement; the page’s social proof rises from 7,000+ to 8,000+ businesses.
- 2023-03 — Premier is introduced at $229/seat/mo with a 50-seat minimum, more than doubling the top of the ladder, and the Scale minimum jumps from 10 to 20 seats.
- 2023-05 — Growth rises from $49 to $59/seat/mo, a 20% increase and the only clean single-SKU list-price rise visible in the 2021–2025 archive.
- 2024-05 — AI Answers launches at $0.70 per resolution, Front’s first outcome-priced SKU and its entry into the per-resolution pricing cohort that Intercom’s Fin defined.
- 2024-11 — Premier’s $229 price is replaced by “Contact Us”, moving Front’s top tier back behind a sales conversation after 20 months of public pricing.
- 2025-08 — The repackage (see below).
- 2025-11 — AI Answers is retired and Autopilot takes over at $0.89 per resolution; the legacy help-center article carries a “no longer available for purchase” banner.
- 2026-04-28 — Front announces Autopilot Playbooks, Autopilot Resolve, and a rebuilt Copilot on its blog, moving Autopilot from a chat-deflection product to a multi-step process agent.
- 2026-06 — The three-level Triage / Handoff / Resolution ladder is published on the pricing page as “Starting at $0.05 /conversation”.
The 2025 repackage in detail
The single largest pricing event in Front’s archived history happened between the 2025-07-08 and 2025-08-05 Internet Archive snapshots of front.com/pricing, with no accompanying blog or changelog post that we could locate. In one move Front:
- Collapsed four tiers into three — Starter, Growth, Scale, Premier became Starter, Professional, Enterprise — retiring Premier entirely.
- Raised every surviving price: Starter $19 → $25 (+32%), Growth $59 → Professional $65 (+10%), Scale $99 → Enterprise $105 (+6%). The steepest increase landed on the cheapest, most price-sensitive tier.
- Replaced Growth’s “no seat minimum” with a 50-seat ceiling on Professional, and kept Starter’s 10-seat ceiling — turning both plans into capacity boxes rather than open-ended ladders.
- Changed the WhatsApp economics from a predictable $70 per 1,000 conversations to a pass-through of Meta’s messaging cost plus a 20% admin fee, converting a fixed rate into a variable margin on someone else’s price list.
- Introduced the API Rate Limit Increase add-on at $200 per additional 100 requests/min per month — a rare example of throughput itself being sold as a SKU in a helpdesk product.
- Quietly cut the annual-billing headline from SAVE 34% to SAVE 24%, which is a real increase in the effective annual price on top of the list-price rises.
For existing customers, Growth, Scale, and Premier became “legacy plans” — Front’s own help-center articles for Copilot, Smart QA, and Smart CSAT still price the add-ons “for the latest Starter/Professional plans and legacy Growth/Scale plans”, confirming legacy accounts were kept on their old seat prices while the add-on rate card was unified across both generations.
What’s unique : Front’s distinctive pricing mechanics
1. The top tier is priced below its own à-la-carte AI bundle. Enterprise costs $105/seat/mo and includes Copilot, Smart QA, and Smart CSAT. Buying the same three on Professional costs $65 + $20 + $25 (the Smart QA + Smart CSAT bundle) = $110/seat/mo — five dollars more for fewer platform features. Most vendors price the flagship tier at a premium and let add-ons soften the gap; Front inverts it, so every AI add-on sold on Professional is an argument for upgrading. The add-on rate card is deliberately not a profit centre — it is a ramp.
2. One conversation, three prices, charged once. Autopilot’s meter is not per message, per token, or per minute; it is per conversation, resolved to whichever of three automation levels the agent actually reached — Triage $0.05, Handoff $0.39, Resolution $0.89. A conversation that gets triaged, then runs a Playbook that ends in a human takeover, bills $0.39 and not $0.44. Front also publishes what it will not charge for: a teammate cancelling the automation, the customer asking for a human, or a technical failure on Front’s side. And it defines “resolved” operationally — no teammate follow-up within 24 hours on Resolve, SMS, WhatsApp, and custom channels, or 72 hours on email and the portal. That is a far tighter contract than “AI resolution” as most vendors define it.
3. Buying the usage SKU unlocks a platform entitlement. Front’s help centre states that the Autopilot add-on grants access to dynamic variables for Smart Rules and custom Connectors “regardless of your pricing plan” — capabilities otherwise gated to the Enterprise tier. A Starter customer who buys a per-conversation meter therefore gets a seat-tier feature for free. It is a genuine cross-axis leak in an otherwise clean seat ladder, and it makes the outcome meter more attractive than its headline rate suggests.
4. Seat ceilings, not seat overages. Starter stops at 10 seats and a single channel type; Professional stops at 50. Front does not sell an eleventh Starter seat at a higher rate — the account re-prices at the next tier. Most seat-based SaaS gates features and lets headcount run; Front gates headcount, which produces unusually violent step functions (see Hidden costs) but also an extremely legible upgrade trigger for its sales team.
Strengths & weaknesses
| Strengths | Weaknesses |
|---|---|
| Outcome definitions are unusually precise. Autopilot publishes three named automation levels, a once-per-conversation charging rule, an explicit no-charge list, and a 24h/72h silence window that defines “resolved”. | Hard seat ceilings create violent cliffs. An 11th hire takes a Starter team from $250 to $715/mo (+186%); the 51st seat on Professional forces a 65% jump to Enterprise. There is no overage seat. |
| The AI upgrade path is arithmetically obvious. Enterprise at $105/seat bundles Copilot, Smart QA, and Smart CSAT, which cost $110/seat if bought on Professional — so the upsell needs no persuasion. | No self-serve spend cap on Autopilot. Front’s own docs tell customers to “Contact us to review a maximum limit of conversations”, so the only ceiling on the usage line is a sales conversation. |
| Every current price is on the public page. All three seat tiers, all four AI add-ons, the QA/CSAT bundle, and the API rate-limit SKU carry visible numbers — rare in a category that mostly hides the top tier. | Top-of-ladder transparency has regressed before. Premier carried a public $229 from March 2023, went to “Contact Us” in the November 2024 archive, and was retired in 2025; Autopilot’s rate was pulled from the page for roughly four months in early 2026. |
| Billing UX is genuinely self-serve. A Bill summary panel previews the new price before you confirm, add-ons can be activated from billing settings, and invoices are prorated then aggregated into one consolidated bill. | Payments are nonrefundable and downgrades are renewal-only. Removing a license yields “no refund or credit”; seat decreases, plan changes, and add-on removals only take effect at renewal, so over-buying is a term-length mistake. |
| Trials are real, not teasers. A 14-day Professional trial needs no credit card, Autopilot ships a 30-day trial that includes up to 1,000 Resolutions, and Smart CSAT has its own 14-day trial. | The AI SKU line has churned. AI Answers, sold at $0.70/resolution from May 2024, is now marked “no longer available for purchase” with the only path forward a migration to Autopilot Resolution at $0.89 — a 27% higher rate for the same job. |
Billing UX : Front billing controls and transparency
- Current plan tab (self-serve Billing settings) — view and edit plan tier, seat count, billing recurrence, and active/inactive add-ons directly from company settings; changes apply immediately to the subscription.
- Bill summary panel — a live-updating price preview shown alongside any plan, seat, or add-on change before it’s confirmed.
- Upcoming plan at renewal tab — where downgrades (seat decreases, plan-type changes, add-on removal) are staged; Front explicitly will not apply downgrades before the renewal date.
- Payment details and invoices tab — manage card, ACH, or wire-transfer details, billing contact email, Tax ID, and download past invoices (view all / per-invoice download icons).
- Prorated, aggregated invoicing — new teammates are billed pro-rata for the remainder of the billing period, then rolled into one consolidated monthly invoice rather than billed the instant they’re added.
- Billing frequency toggle (Monthly ⟷ Annual) — a switch on the public pricing page itself that changes displayed per-seat prices in real time (Starter/Professional only; Enterprise is annual-billing only).
- Free-trial usage gate — the 14-day trial runs on Professional-tier limits (15 teammates, 5 workspaces, 30 channels, a 10-recipient message cap, and a 5,000-message cap, with Sequences/guest invites/auto-reply locked) and requires no credit card.
- Autopilot usage view (Billing settings) — an estimated month-to-date overview of Autopilot conversation usage; note the caveat in Front’s own docs that setting “a maximum limit of conversations” or getting a detailed per-conversation breakdown requires selecting Contact us, so there is no self-serve spend cap.
- Per-feature free-trial buttons — Front exposes a Free trial control inside the settings menu of premium add-ons, and the length varies by product: 14 days for Smart CSAT, 30 days for Smart QA, and 30 days for Autopilot with up to 1,000 Resolutions included; after an Autopilot trial ends, any Autopilot rules, Playbooks, and Resolve channels stop running unless you actively opt in to the paid version.
- Pre-paid volume commitments — Autopilot is pay-as-you-go by default and invoiced monthly on actual usage; Front’s documentation states that “volume discounts are pre-paid commitments”, so the discounted rate is only reachable through a sales-quoted prepay.
- [email protected] — a dedicated billing contact channel for ACH/wire setup on annual contracted deals over $12K.
- Nonrefundable-by-policy — worth naming as a billing-UX constraint rather than a fine-print item: monthly and annual payments are nonrefundable and are not issued as account credit, and removing a licence or add-on gives no refund or credit — you keep the unused capacity until the end of the term.
Strategic wins : Why Front’s pricing decisions worked
1. Splitting the AI portfolio by autonomy, not by brand
Most helpdesk vendors sell “AI” as one upcharge. Front split its four AI products along the line that actually matters to a buyer’s budget: the ones a human drives (Copilot, Smart QA, Smart CSAT) stay per-seat and predictable, and the one that acts on its own (Autopilot) is billed per delivered outcome. That mapping means the meter matches the mechanism — a team with 12 agents and low automation pays 12 × $45, while a team automating 5,000 conversations a month pays for those conversations regardless of headcount. It is the cleanest expression of the shift away from per-user licences in the customer-service category, and it avoids the trap of charging seats for work that no longer requires a seat. Front’s own framing in its help centre is the giveaway: “Autopilot automates business processes; Copilot assists teammates with their work.”
2. Making the outcome definition falsifiable before selling it
Front did not ship “pay per resolution” and leave the definition to the invoice. It published three automation levels with worked examples, a once-per-conversation charging rule, a no-charge list (teammate cancels, customer asks for a human, Front-side failure), and a time-boxed test for what counts as resolved — no teammate follow-up within 24 hours on chat/SMS/WhatsApp channels or 72 hours on email and portal. That precision is the hardest part of choosing the right usage metric: a metric the buyer cannot audit becomes a trust problem the first time a bill surprises them. Compare it with the wider outcome-based pricing shift, where many vendors still leave “resolution” as a marketing word. Front also priced partial success, which most competitors do not — a triaged conversation that never reaches resolution still earns Front five cents and still saves the buyer a routing decision.
3. Using the repackage to raise prices without an announcement
The 2025 Q3 move from four tiers to three did four things at once: retired the $229 Premier tier, raised all three surviving prices, converted a flat WhatsApp rate into a pass-through-plus-20% margin, and cut the annual discount headline from 34% to 24%. Because the plan names changed, none of those increases read as a price rise on the page — a customer comparing “Growth $59” to “Professional $65” is comparing two different products, not the same one before and after. Legacy customers were left on their old seat prices, which removes the immediate churn risk while every new cohort lands on the higher card. It is textbook repackaging, and it worked in the sense that we could find no press coverage, changelog entry, or community backlash tied to it — the change is only visible by diffing archived snapshots.
Areas to improve : Gaps in Front’s pricing approach
Community-signal note: we searched Hacker News via the Algolia API, Reddit, G2, TrustRadius, and general press for pricing-specific reaction to Front. No trust event met our thresholds — the highest-scoring Front story on HN is a 94-point, 14-comment Show HN from 2014-06-09 about the original product, not pricing, and no HN thread above 50 points discusses Front’s pricing at all. Reddit’s search API and TrustRadius both refused our fetcher, so that channel is unverified rather than clean. The consistent criticism below comes from third-party pricing teardowns and review-aggregator summaries (Hiver, eesel, getmacha, Drag, Desk365, TrustRadius/G2 summaries), which converge on the same three complaints — per-seat cost escalation, AI sold as stacking add-ons, and contractual rigidity — and each is verifiable against Front’s own published prices and billing documentation.
1. Sell an overage seat instead of a cliff
The strongest and most repeated criticism of Front’s pricing in third-party reviews is that per-seat cost escalates faster than value as a team grows. Front’s ceilings turn that from a gradient into a step: the 11th seat on a 10-seat Starter account takes the bill from $250 to $715/mo, a 186% increase for one hire, and the 51st seat on Professional adds 65%. A team that hires one person cannot buy one person’s worth of software. The fix is not to abandon tiering — the ceilings are a legitimate segmentation device — but to price the crossing: a per-seat overage rate above the ceiling (say, Professional-rate seats bolted onto a Starter base) or a published, time-boxed blended rate for the quarter in which a team crosses. Either preserves the upgrade trigger without making a routine hire feel like a penalty.
2. Ship a self-serve spend cap for Autopilot
Autopilot is uncapped pay-as-you-go, invoiced monthly on actual usage. Front’s billing settings show an estimated month-to-date figure, but the help centre instructs customers to “Contact us to review a maximum limit of conversations” — meaning the only real ceiling is a sales conversation. For a meter that can be more than half the bill (55% in our 12-seat worked example), that is the wrong default. A self-serve monthly conversation cap plus configurable alert thresholds at, say, 50/80/100% of budget would close the gap; the mechanics are standard and covered in our guide to thresholding and alerting on usage-based pricing. Until then Autopilot is exactly the kind of meter that produces unpredictable AI product costs on the buyer’s side of the invoice.
3. Publish a deprecation and migration policy for AI SKUs
AI Answers was sold at $0.70 per resolution from May 2024. Its help-center article is now titled “AI Answers [legacy]” and opens with “AI Answers is no longer available for purchase”, pointing customers at Autopilot Resolve. The replacement outcome costs $0.89 — 27% more for the same job — and we found no public notice period, no grandfathering statement, and no migration guarantee anywhere on front.com or in the community announcements. Combined with a nonrefundable, renewal-only downgrade policy, that leaves a buyer with no way to model the risk of the SKU they just standardised on being retired. The fix is cheap and entirely editorial: a dated deprecation policy (minimum notice, price protection window, migration path) published alongside the pricing page. Vendors that make outcome pricing durable will be the ones that make its lifecycle predictable, not just its per-unit rate.
Monetization stack & signals : how Front builds & buys its revenue engine
Buys 5 Builds 1 2 signal roles
Buys its GTM stack — Salesforce, Snowflake, Gong — then builds an in-house 'GTM Engineering' AI-agent layer on top, now large enough to need its own platform-security and cost owner. No billing or metering vendor is named anywhere on its board.
-
“Front is building a serious amount of software for itself. GTM Engineers are shipping internal tools and AI agents on top of Salesforce, Gong, Snowflake, Workato, Notion and Claude — automations that our revenue teams now depend on to do their jobs.”
-
“GTM Engineers are shipping internal tools and AI agents on top of Salesforce, Gong, Snowflake, Workato, Notion and Claude — automations that our revenue teams now depend on to do their jobs.”
-
“Own the cloud infrastructure our internal tools and AI systems run on AWS & Snowflake: compute, networking, data stores, DNS, certificates.”
-
“GTM Engineers are shipping internal tools and AI agents on top of Salesforce, Gong, Snowflake, Workato, Notion and Claude — automations that our revenue teams now depend on to do their jobs.”
-
“Hands-on experience owning a marketing automation platform and CRM integration ... Required experience in HubSpot.”
-
“Maintain a pipeline of prospects through Gong/Salesforce to forecast revenue accurately”
-
Front buys the CRM, warehouse and revenue-intelligence layer, then runs an internal 'GTM Engineering' function shipping AI agents on top of it — mature enough to need a dedicated platform-security hire reporting into Corporate IT. The same role owns platform cost, attributing AI and warehouse spend to the teams driving it: chargeback discipline on Front's internal revenue stack, not on Autopilot's per-conversation margin.
“Front is building a serious amount of software for itself. GTM Engineers are shipping internal tools and AI agents on top of Salesforce, Gong, Snowflake, Workato, Notion and Claude — automations that our revenue teams now depend on to do their jobs.”
-
Titled 'Customer Success Manager' externally but 'Growth Specialist' in the JD body, sitting on the Self-Serve team — a self-serve upsell role that explicitly coordinates with a named internal 'billing' team on upgrade escalations, the human layer behind Front's hard seat-ceiling upgrade mechanic.
“Front is looking for a Growth Specialist to join our world-class Go-To-Market organization... guide accounts towards upgrading to higher-tier plans... Cross-functional Collaboration: Work with other teams - support, billing, operations - to manage escalations and staff for campaigns.”
1 more matched role — supporting evidence
- Director, Account Management Retention Aug 19, 2026
Signals reviewed · derived from public job posts
Job postings fill and close over time — once a posting is filled we keep it as a dated citation (the quoted evidence remains); use View open roles for current listings.
Key takeaways
- Price your top tier below the sum of its add-ons. Front’s Enterprise plan costs $105/seat and includes three AI products that cost $110/seat if bought separately on Professional. Every add-on your sales team sells then becomes an argument for the upgrade, which converts an à-la-carte rate card from a margin decision into a routing decision.
- Split the AI rate card by autonomy, not by product family. Assistive features that a human operates map cleanly to seats; autonomous features that act alone map cleanly to outcomes. Front runs both meters in parallel rather than forcing everything into one, and the resulting bill is legible to a buyer who can point at which half is fixed and which half moves.
- Publish what you will not charge for. Front’s Autopilot documentation lists three no-charge conditions and a time-boxed definition of “resolved” before it lists a price. A negative-space definition is worth more trust than a positive one, because it tells the buyer where the vendor absorbs the risk of its own product failing.
- Repackaging is the least-friction way to raise a price — and the least accountable. Renaming Growth to Professional while moving $59 to $65 means no customer can diff the same product across the change. It works, but it also means your only honest record of your own pricing history lives in someone else’s archive. Publish a changelog before an archive becomes the primary source.
- Seat ceilings are a segmentation tool with a customer-experience bill attached. Capping Starter at 10 seats gives you a clean upgrade trigger and a clean ICP, but it means a routine hire can triple a small team’s invoice. If you use ceilings, price the crossing explicitly — see choosing the right usage metric for how to reason about step functions versus gradients.
UBP implications
- Outcome pricing is fragmenting into outcome ladders. Intercom’s Fin established the single-price resolution meter, and the customer-support AI pricing cohort largely copied it. Front’s Triage $0.05 / Handoff $0.39 / Resolution $0.89 ladder is the next step: partial automation is a billable outcome too, at roughly 1/18th the price of a full resolution. Expect more vendors to monetise the work an agent does before it finishes, because the alternative is doing that work for free.
- The credibility of an outcome meter lives in its exclusions, not its rate. A per-resolution price is only as trustworthy as the definition of resolution and the list of cases the vendor declines to bill. Front’s 24h/72h silence windows and its three no-charge conditions are the substantive part of the offer; the $0.89 is the easy part. UBP teams shipping outcome pricing should write the exclusion list first and derive the rate from it, not the other way round.
- Hybrid pricing survives because the fixed half is what finance can actually plan. Front did not replace seats with outcomes — it kept a fully seat-based platform and bolted an outcome meter on top, and in our worked example the meter is 55% of the bill while the seats are 45%. That split is the practical shape of the hybrid pricing model in incumbent SaaS: the seat line funds the platform and anchors the budget, and the usage line captures the automation upside. Pure-usage conversions in this category remain rare for exactly that reason.
Sources
- Front pricing page — all three seat tiers, the four AI add-ons, the Smart QA + Smart CSAT bundle, Native WhatsApp, API Rate Limit Increase, and the plan-comparison matrix (accessed 2026-09-04)
- Automate complex workflows with Autopilot — the Triage / Handoff / Resolution rate table, the once-per-conversation rule, the no-charge conditions, the 24h/72h resolution windows, the pre-paid volume-discount commitment note, and the 30-day / 1,000-Resolution trial (accessed 2026-09-04)
- Get real-time conversational assistance in your inbox with Copilot — $20/seat/mo add-on pricing, Enterprise inclusion, and the 200-drafting-requests-per-user-per-day rate limit (accessed 2026-09-04)
- Smart QA — $20/seat/mo add-on pricing, Enterprise inclusion, and the 30-day free trial (accessed 2026-09-05)
- Smart CSAT — $10/seat/mo add-on pricing, Enterprise inclusion, the 14-day trial, and the ~90% alignment claim (accessed 2026-09-04)
- Front’s plans and pricing — trial limits, licence minimums, accepted currencies and payment methods, and the refund policy (accessed 2026-09-04)
- Manage your billing and invoices — the Current plan tab, Bill summary panel, Upcoming plan at renewal tab, and licence-removal rules (accessed 2026-09-04)
- Understanding your invoice — upfront payment, proration, aggregated billing, and the no-credit rule on removals (accessed 2026-09-04)
- Front Help Center — Billing category — cancellation, sales tax, and multi-instance guidance (accessed 2026-09-04)
- AI Answers [legacy] — the “no longer available for purchase” banner and the historical $0.70-per-resolution rate (accessed 2026-09-05)
- From request to resolution: How Front AI runs the work, not just the reply — the 2026-04-28 Autopilot Playbooks / Resolve / Copilot announcement and its add-on entitlement statements (accessed 2026-09-05)
- Front Raises Series D at $1.7B Valuation — the June 2022 $65M Series D and valuation figures cited in About (accessed 2026-09-05)
- Internet Archive snapshots of front.com/pricing, July 2021 – August 2026 — the basis for the Cadence table and every dated entry in the pricing timeline (accessed 2026-09-05)
More company teardowns like this one are indexed in the pricing blueprint.
Bottom line
Front runs one of the most legible hybrid rate cards in customer service: three public seat tiers, four AI products split cleanly between per-seat assistance and per-outcome automation, and an Autopilot meter whose exclusions and resolution windows are documented before its price is. The mechanics are good — pricing the top tier five dollars below its own à-la-carte AI bundle is a genuinely elegant upsell, and the Triage/Handoff/Resolution ladder is a more honest outcome model than the single-price resolution meters it competes with. What lets it down is contract shape rather than rate shape: hard seat ceilings that turn one hire into a 186% bill increase, a usage meter with no self-serve spend cap, nonrefundable prepayment with renewal-only downgrades, and a 2025 repackage that raised every price and cut the annual discount without a single public note. Buy Front for the model; negotiate it for the terms.
Want to compare Front against other customer-service platforms? See Intercom Fin, Zendesk AI, and Gorgias, or browse the full pricing blueprint.
Pricing timeline : Major events on a vertical axis
Each milestone below corresponds to a public pricing change, product launch, or material adjustment. Major events use a filled marker; minor adjustments use a faded one.
Current state: three seat tiers plus a four-product AI add-on portfolio
Starter $25, Professional $65, and Enterprise $105 per seat per month billed annually ($35 / $85 monthly; Enterprise annual only). Copilot $20, Smart QA $20, and Smart CSAT $10 per seat are add-ons on Starter and Professional and bundled free into Enterprise; Autopilot remains outcome-priced on every plan. The annual-discount headline now reads SAVE 24%, down from SAVE 34% in 2024–2025.
Three-level outcome ladder published: Triage $0.05, Handoff $0.39, Resolution $0.89
By the 2026-06-12 snapshot Autopilot was back on the pricing page as 'Starting at $0.05 /conversation' with the full ladder documented in the help center — $0.05 Triage, $0.39 Handoff, $0.89 Resolution — each conversation charged once at the highest level reached.
Autopilot Playbooks, Autopilot Resolve, and the rebuilt Copilot announced
Front announced the next phase of Front AI on its blog: Playbooks for multi-step processes, Resolve as an AI-native self-serve channel, and a rebuilt Copilot. Playbooks and Resolve ship inside the Autopilot add-on and must be enabled by Front; Copilot stays an add-on on Starter and Professional and is included in Enterprise.
Autopilot's price pulled from the pricing page
The $0.89 figure vanished between the 2026-01-19 and 2026-02-05 snapshots, leaving Autopilot as a Talk to Sales card with no published rate for roughly four months.
AI Answers retired; Autopilot takes over at $0.89 per resolution
Between the 2025-09-09 and 2025-11-05 snapshots the $0.70-per-resolution AI Answers SKU disappeared from the pricing page and Autopilot appeared in its place at $0.89. Front's help-center article for AI Answers is now titled 'AI Answers [legacy]' and carries the banner 'AI Answers is no longer available for purchase.'
Repackage: four tiers become Starter $25 / Professional $65 / Enterprise $105
Between the 2025-07-08 and 2025-08-05 Internet Archive snapshots Front collapsed its ladder from four tiers to three and raised every surviving price: Starter $19 to $25 (+32%), Growth $59 to Professional $65 (+10%), Scale $99 to Enterprise $105 (+6%), with Premier retired. The Native WhatsApp Channel switched from a flat $70 per 1,000 conversations to Meta-billed costs plus a 20% admin fee, and an API Rate Limit Increase add-on appeared at $200 per additional 100 requests/min per month.
Smart QA ships at $20 per seat per month
Smart QA moved from COMING SOON in the March 2025 snapshot to a live premium add-on at $20 per seat per month, joining AI Answers ($0.70/resolution) and the WhatsApp channel ($70 per 1,000 conversations). Premium add-ons remained restricted to Growth plan or higher.
Premier's $229 list price replaced by Contact Us
Front withdrew the published price from its top tier, leaving Starter $19, Growth $59, and Scale $99 as the only visible per-seat prices. The premium add-on rail listed AI Answers at $0.70/resolution, WhatsApp at $70 per 1,000 conversations, and Dialpad SMS at $100/month, alongside a Front for Startups offer of up to 6 months free.
AI Answers launches at $0.70 per resolution — Front's first outcome-priced SKU
AI Answers appeared on the pricing page marked NEW at $0.70 per resolution, resolving Front Chat inquiries from help content. The same redesign restored a four-tier grid (Starter $19, Growth $59, Scale $99, Premier $229) with a monthly/annual toggle headlined Save up to 34%.
Growth seat minimum cut from 5 to 2; rule caps published
Front lowered the Growth entry point to a 2-seat minimum and started publishing automation-rule ceilings on the pricing page — up to 20 rules on Growth and up to 200 per workspace on Scale.
Growth raised from $49 to $59 per seat (+20%)
The mid tier took a 20% list increase while Scale held at $99 and Premier at $229. It is the only straight list-price rise visible on a single Front SKU in the 2021–2025 archive.
Premier added at $229/seat; Scale minimum doubles to 20 seats
Front introduced a fourth tier, Premier, at $229 per seat per month with a 50-seat minimum, raised the Scale minimum from 10 to 20 seats, and pushed Starter behind a FOR SMALLER TEAMS toggle. The same snapshot introduced Front's first paid channel add-ons — Telegram and Dialpad SMS at $100 per month each, Scale plan only.
Prime becomes Growth, Enterprise becomes Scale
A pure rename with no price movement — Starter $19, Growth $49, Scale $99, all per person per month billed annually. The page claimed 8,000+ businesses, up from 7,000+ in April 2022.
Enterprise gets a public price at $99/seat
Front published a price for its top tier for the first time in the captured record: Enterprise at $99 per person per month billed annually with a 10-user minimum, badged MOST POPULAR. Starter was capped at a 10-user maximum in the same snapshot.
Three-tier Starter / Prime / Enterprise ladder
The archived pricing page shows Starter at $19 and Prime at $49 per person per month billed annually, with Enterprise as contact-sales only. By December 2021 the page also published seat minimums: 2 users on Starter, 5 on Prime.
- · Front's Enterprise plan at $105/seat/mo is priced $5 below the same AI stack bought a la carte on Professional — $65 plan + $20 Copilot + $25 Smart QA/CSAT bundle = $110/seat — so the AI add-ons function as an upsell ramp into the top tier rather than a standalone margin line.
- · Autopilot charges each conversation only once, at the highest automation level it reached, and Front publishes a no-charge list: a teammate cancelling the automation, the customer asking for a human, or a technical failure on Front's side are all free.
- · Front's first outcome-priced SKU was AI Answers at $0.70 per resolution, launched in May 2024. Its help-center article is now titled 'AI Answers [legacy]' and carries a 'no longer available for purchase' banner — the successor, Autopilot Resolution, costs $0.89, a 27% increase.
Questions & answers
- How much does Front cost per seat?
- Front costs $25 per seat per month on Starter, $65 on Professional, and $105 on Enterprise when billed annually. Billed monthly, Starter is $35 and Professional is $85 per seat; the Enterprise plan is annual-billing only.
- How does Front's Autopilot pricing work?
- Autopilot is outcome-based and billed per conversation at one of three levels: $0.05 for Triage, $0.39 for Handoff, and $0.89 for a full Resolution. Each conversation is charged only once, at the highest automation level Autopilot reached, and Front does not bill when a teammate cancels the automation, the customer asks for a human, or Autopilot suffers a technical failure on Front's side.
- Does Front have a free tier?
- No. Front has no free plan. It offers a 14-day free trial on Professional-tier features with no credit card required, capped at 15 teammates, 5 workspaces, 30 channels, a 10-recipient message limit, and 5,000 messages, with Sequences, guest invites, and auto-reply rules locked. Autopilot has a separate 30-day trial that includes up to 1,000 Resolutions.
- Is Front AI included in the plan price?
- Only on Enterprise. The latest Enterprise plan bundles Copilot, Smart QA, and Smart CSAT at no additional cost; on Starter and Professional they cost $20, $20, and $10 per seat per month, with a $25 per seat Smart QA + Smart CSAT bundle. Autopilot is a paid add-on on every plan, including Enterprise.
- What are the hidden costs in a Front bill?
- The big ones are the seat ceilings (Starter stops at 10 seats, Professional at 50, each forcing a tier jump), a required onboarding package on contracts over $25k, the Native WhatsApp Channel which passes through Meta's messaging costs plus a 20% admin fee, and API rate-limit increases at $200 per additional 100 requests per minute per month.
- Can you get a refund if you cancel or remove Front seats?
- No. Front's billing documentation states that monthly and annual payments are nonrefundable and are not issued as account credit, and that removing a license gives no refund or credit — the seat stays usable until the end of the term. Downgrades, including seat decreases and add-on removal, only take effect at renewal.