Skip to content
‹ Back to IndustriesSoftware & Technology

How to choose between per-resolution and six-figure AI support contracts

Pricing for AI-native customer support platforms varies widely. Fin charges $0.99 per outcome, while Sierra contracts are estimated at $150,000 to over $350,000 a year, so support leaders should model their own resolution volume first.

This story was produced through MarketScale. See how Software & Technology teams put it to work with Executive Thought Leadership.

By MarketScale Newsroom · Ai Customer SupportCustomer Service Ai AgentsSierraDecagon
Share
How to choose between per-resolution and six-figure AI support contracts

Key takeaways

01

When 70%-plus resolution is a standard vendor claim, the pricing model is what separates AI support platforms: $0.99 per outcome at one end, an estimated $150,000 to $350,000-plus a year at the other.

02

Resolution can range from 64% to 94% inside a single platform depending on deployment maturity, and only 10% of support organizations report mature AI deployment, so year-one budgets belong at the low end of any vendor range.

03

A slow ramp shrinks a per-outcome bill but not a fixed annual contract, which makes expected resolution volume the first number to pin down before comparing quotes.

Free workspace

Turn your Software & Technology expertise into content.

Record interviews, organize footage, and write with AI on a free trial of the MarketScale platform for qualifying companies. No demo required, no credit card.

Try it Free

Dynamic Business has published a buyer's guide to AI-native customer support platforms, and it gives support leaders plenty to compare. For operators weighing a purchase, the harder question may not be which tool resolves the most tickets. It may be how each vendor charges for the work.

When every vendor arrives quoting the same headline number, that number stops helping anyone choose. The guide's own selection advice points elsewhere. Does the AI replace the helpdesk or sit beside it? Which channel carries the volume? Which certifications does the industry demand? How does the vendor charge? For the head of support who has to walk one shortlist into procurement, that last question is where the 44 options split furthest apart.

Replace the helpdesk or sit beside it

The guide sorts its field into autonomous AI resolution platforms, AI-first helpdesks, traditional helpdesks that have added AI agents, voice AI and contact center platforms, and chatbot builders.

AI-native support platforms by category (count of guide entries)
Dynamic Business guide, entries counted by MarketScale · © MarketScaleDownload chart

The line between the first two groups is the first real decision. Autonomous platforms such as Sierra, Decagon and Ada work alongside an existing helpdesk. AI-first helpdesks such as Intercom Fin, Gorgias and Kustomer replace it. Decagon is a clean example of the first model: it has no native helpdesk, is built to run next to Zendesk, Intercom or Salesforce Service Cloud, and uses what it calls AI Actions to carry out refunds, subscription changes and order lookups in Stripe, Shopify and Salesforce.

The layered platforms compete on how much control they hand the support team. Decagon defines resolution logic through Automated Operating Procedures, or AOPs, written in plain language, and runs a built-in QA layer called Watchtower that monitors live conversations for quality and compliance. Ada pairs a Reasoning Engine with Playbooks so teams can decide when the AI reasons freely and when it follows a set path, and it lists native integrations with Zendesk, Salesforce, Shopify and more than 50 other platforms.

Each camp pitches its architecture as the advantage. Fin, Intercom's AI agent, argues in its own ranked list of customer service agents that running an agent on a separate helpdesk adds integration cost and handoff friction. That page carries a disclosure that it's published by Fin and isn't an independent hands-on test. Fin also sells itself on top of Salesforce, HubSpot and Freshworks, so even the vendor making the single-system case offers the layered route.

The practical split comes down to sunk work. A team with years of macros, routing rules and reporting inside an incumbent helpdesk avoids a migration by layering. A team already planning to leave its helpdesk could fold two purchases into one by choosing an AI-first replacement.

Three ways the bill gets built

When resolution rates above 70% are a standard vendor claim, the number that separates platforms is on the invoice: $0.99 an outcome at one end, an estimated $150,000 to more than $350,000 a year at the other. The high end is Sierra, which the guide describes as enterprise-only, with no self-serve or trial tier. The low end is Fin used with a customer's current helpdesk, at $0.99 per outcome with no separate platform, setup or integration fee, according to Fin.

Other price points on the table

~$50K
Decagon platform fee, per the guide
~$386K
Decagon median ACV, per the guide
$19/seat/month
Added to Fin's $0.99 per outcome when bought with Intercom Helpdesk
50 outcomes
Fin's monthly minimum when used with a current helpdesk

Dynamic Business; Fin (Intercom)

That's three billing logics: an enterprise contract, a platform fee plus contract, and a meter. Fin's list adds that Salesforce Agentforce and Zendesk AI charge per seat, per action or per conversation. The guide is blunt about the tradeoff. Outcome-based pricing rewards teams with high deflection rates, while seat-based pricing is predictable but doesn't scale with automation. Its advice is to know expected resolution volume before comparing costs at all.

An enterprise contract could also include services a per-outcome meter doesn't show, such as hands-on implementation and compliance work.

Sierra shows what that looks like. The guide describes white-glove implementation with typical go-live timelines of four to 10 weeks, alongside Agent Studio for no-code agent design and an Agent SDK for custom integrations. Fin pitches the opposite operating model: self-managed by CX teams with no engineering required, plus simulation and testing tools for safe deployment.

Fin's own evaluation criteria explain why the operating model belongs in the price comparison: a capable product can still be difficult or expensive to operate if every change requires a vendor or engineering team. That suggests counting the people who will maintain the agent next to the license fee, whichever billing logic a team picks.

The meter has its own exposure, too. A per-outcome bill grows with every resolution, which is the whole point when automation is taking work off human agents, but it also means spend moves with contact volume. A team with sharp seasonal peaks could find a flat contract easier to forecast, even if it costs more per resolution in a quiet month.

The guide says its entries were researched from vendor websites, independent analyst reviews and third-party pricing data. Sierra's figure is labeled an estimated annual contract value (ACV) and Decagon's a median, while Fin's come from its own published pricing table, so a side-by-side compares estimates with list prices.

The 70% claim meets the maturity gap

Resolution rates deserve a closer read than the headline gives them. The guide reports Sierra's rates at anywhere from 64% to 94% depending on deployment maturity, a 30-point spread inside a single platform. Fin reports a 76% average across 12,000 customers, improving about one point a month, with ecommerce brands at 70% to 84%. Both are vendor-reported, and both say the same thing: the rate a team sees depends on how far into the deployment it is.

Very few teams are far in. Intercom's 2026 Customer Service Transformation Report, which surveyed 2,470 support professionals worldwide in the fourth quarter of 2025, found that 82% of senior leaders had invested in AI for customer service over the prior 12 months and 87% planned to invest in 2026. Only 10% said they had reached mature deployment, defined as AI fully integrated into support operations and delivering measurable results at scale.

Put those together and a pattern suggests itself. The top of a vendor's range describes deployments that have had time to mature, and most organizations, by Intercom's count, aren't there yet. A support leader building a business case would be on safer ground modeling the low end of any range for year one and treating the high end as something the team earns through knowledge base work, workflow design and integrations.

That connects straight back to pricing. Under a per-outcome meter, a slow first year could mean a smaller bill. Under a fixed annual contract, the full cost lands whether the agent resolves 64% of conversations or 94%.

Some buyers may want a fallback while the agent climbs that curve. The guide lists Crescendo, delivered as cloud plus business process outsourcing (BPO), as a fit for teams wanting AI with a BPO safety net behind it.

What counts as a resolution matters as much as the rate. The guide separates simple FAQ deflection from complex multi-step transactions such as refunds, account changes and bookings, says they require different AI architectures, and advises matching a platform to the complexity of a team's top support drivers. Fin's criteria draw a similar line: deflection reduces ticket volume, but genuine resolution determines customer experience and operational value.

Academic work shows why that line carries cost. A 2024 Business Horizons article on generative AI customer service by Carla Ferraro and colleagues notes that even the highest-quality GenAI chatbots can misunderstand user input or produce inaccurate responses, resulting in service failure. Under a per-outcome contract, whether a conversation like that gets billed would turn on how the contract defines an outcome.

Ask each finalist for the resolution rate its customers hit in their first 90 days, not only the fleet average. Then get the contract's definition of an "outcome" or "resolution" in writing before comparing per-resolution prices side by side.

Certifications, deployment and new owners

Some filters remove platforms before price comes up, and compliance certification is one of them. Sierra lists SOC 2, ISO 27001, HIPAA, GDPR and ISO 42001. Fin lists ISO 42001, SOC 2 Type II, ISO 27001 and HIPAA.

The guide spells out where the bar sits by sector: financial services (FCA, PCI DSS, DORA), healthcare (HIPAA) and government (FedRAMP) each carry specific certification requirements. It flags compliance status for every platform with a blunt instruction: do not assume.

Deployment can matter as much as paperwork. Cognigy and Zowie are listed with cloud, on-premises and air-gapped options, the guide's answer for regulated contact centers and data sovereignty. Gradient Labs targets UK and EU fintechs regulated by the FCA, and Lorikeet targets fintech and healthtech companies.

Channel mix narrows the list further. If inbound phone volume is the main load, the guide steers buyers to platforms built for voice AI. Chat-and-email platforms with voice added belong in a different evaluation.

Even within voice, the fit differs. The guide matches PolyAI to large, voice-only contact centers and Parloa to global enterprises working in more than 130 languages. Sierra, by comparison, runs voice, chat, email, SMS and WhatsApp from a single agent configuration.

Ownership is the last filter. The guide notes M&A activity across 2025 and 2026 and names three platforms now under larger companies: Forethought with Zendesk, LivePerson with SoundHound and Cognigy with NICE. It advises factoring integration plans and roadmap direction into the evaluation. For a buyer, that's a contract question: an acquired product may gain a larger parent's ecosystem, and a written roadmap commitment is how to find out which way it's heading.

The guide's own summaries hint at that direction. It names Forethought AI Agents by Zendesk as best for existing Zendesk customers and LivePerson as best for existing LivePerson customers. A team outside those installed bases would do well to ask how much of the roadmap is aimed at it.

Three numbers to have before the demos

The guide's advice and the vendor figures point to a short piece of homework that changes how every sales conversation goes.

  1. Expected monthly resolution volume. The guide makes this the starting point for any cost comparison, and it sets the scale. At $0.99 an outcome, Sierra's estimated $150,000 floor works out to about 151,500 outcomes a year, roughly 12,600 a month. The bundled services differ, so this is a rough marker rather than a break-even, but it shows which side of the line a team sits on.
  2. A realistic year-one resolution rate. Start from the low end of a vendor's range, such as Sierra's 64%, and multiply by contact volume. That turns a percentage into the resolution count a per-outcome bill is actually built on.
  3. Human seat count. Seats still drive part of the bill in several models: Fin with Intercom Helpdesk adds $19 per seat a month, and Fin's list counts per-seat billing among the options from Agentforce and Zendesk AI. A team expecting automation to absorb work would want to model the seats it plans to keep, not the ones it has today.

Featured companies

Your experts belong here

Every story in MarketScale Software & Technology starts with a company putting its solutions engineers, product teams, and customer engineers on the record. Buyers are already reading this topic. The only question is whose experts they find.

Buyers ask AI engines who to consider, and published expert answers are what those engines cite.

Book DemoSee how it works15 minutes, straight to a calendar.

About the author

MarketScale Newsroom
MarketScale NewsroomEditorial Team, MarketScale

The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.

B2B Weekly

The week in Software & Technology, and sixteen other industries, every Monday.

Ten stories, one-line takes, five minutes. Free.

Software & Technology: are you visible to AI?

Before they reach out, Software & Technology buyers ask AI engines which vendors to trust. Explore how your experts, customers, and partners can become useful content for buyers and AI search.

Free Trial

You just read one Software & Technology expert. Your company is full of them.

This article was produced through MarketScale. The same platform turns your solutions engineers, product teams, and customer engineers into the articles, video, and social content Software & Technology buyers are searching for. Start a free trial and see it with your own people. For qualifying companies, no credit card, no demo required.

NPS +73 · 1,000+ creators · 38+ countries

What your free trial includes

Hands-on access to the MarketScale platform
Media requests to your crowd, remote recording, AI writing tools
No demo required. No credit card.
For qualifying companies. Company confirmation required.

More Software & Technology Insights

OpenAI and ServiceNow expand partnership to connect workflows with ChatGPT

OpenAI and ServiceNow expand partnership to connect workflows with ChatGPT

According to OpenAI, ServiceNow's Action Fabric applies each customer's existing business rules and permissions to what people can do in ChatGPT.

  • 01ChatGPT inherits a ServiceNow instance's existing business rules and permissions through Action Fabric, so years of accumulated role settings now govern a new conversational interface.
  • 02The ServiceNow MCP plugin is available in the OpenAI marketplace now; joint OpenAI-ServiceNow security tooling enters early access in January 2027 and general availability in the second quarter of 2027, which puts it squarely in 2027 security planning.
  • 03With OpenAI now partnered with both IBM and ServiceNow, model access is a weak basis for comparing platforms; the permission enforcement and record context a model plugs into is the sharper thing to evaluate.

Nov 16, 2026

HFS survey: only a third of AI-active enterprises hit their AI business case

HFS survey: only a third of AI-active enterprises hit their AI business case

Only 33% of 502 senior AI decision makers at AI-active enterprises said their biggest AI program met or beat its business case in an HFS Research–Rackspace survey. Success drops to 28% when AI spans several functions. HFS says the pattern points to a fixable gap: unclear ownership of live AI performance, cost and governance as one system.

  • 01Business-case success falls from 37% to 28% as AI moves from one function to several, and reaches 56% where AI is embedded in standard operations. The groups differ, so this is a pattern rather than proof.
  • 02Survey buyers signaled willingness to pay a premium for a single accountable operator—73% said they would pay for a contracting model tied to clear operational responsibilities and measurable outcomes. HFS says the appetite is strongest among companies that have already felt the ownership gap.
  • 03Before adding another function to a working AI program, settle who answers for its performance, cost and governance after go-live, and how each is measured.

Nov 15, 2026

IBM brings 6,000+ agents and assets to Imagine Enterprise

IBM brings 6,000+ agents and assets to Imagine Enterprise

Imagine Enterprise, the agent platform from IBM and Microsoft, pairs IBM Consulting's more than 6,000 AI agents and assets with Microsoft's AI, data, cloud and security foundation. Forward-deployed engineers from both companies support qualifying engagements. For regulated firms, a useful test could be how quickly one named process reaches production.

  • 01Imagine Enterprise combines more than 6,000 AI agents and assets, Microsoft's identity, security and AI governance tools, and forward-deployed engineers from both companies for qualifying engagements, which suggests buying it looks closer to a delivery engagement than a software download.
  • 02Microsoft products including Entra, Purview and Defender provide the identity, AI governance and security foundation, which could matter most for regulated firms already standardized on Microsoft.
  • 03The sharper pilot question is wall-clock completion time and success rate on one named process as more agents run at once, a yardstick Arm is pushing for agentic workloads.

Nov 15, 2026

Explore More Software & Technology Insights

Read more expert perspectives from across Software & Technology.

Browse Software & Technology Hub

About the Expert

MarketScale Newsroom
MarketScale Newsroom

Editorial Team

MarketScale

The MarketScale Newsroom reports on the companies, technologies, and trends shaping 16 B2B industries. It turns primary sources and expert commentary into clear, useful coverage for the people doing the work.

For B2B teams

Your experts could be publishing here

Stories like this one run on content MarketScale captures from real practitioners. See how your team's expertise becomes coverage in Software & Technology and beyond.

Book a Demo

Or call us. No forms required. We pick up. 214-945-2512