Comparison
Belowdecks and Relevance AI compared point by point
Relevance AI is a platform to build specialized AI agents yourself. With Belowdecks, our team installs AI agents for you and monitors their work. This page compares the two on the points that matter to a small business.
Two ways to hand work to AI
Relevance AI is a no-code platform to build, test and manage specialized AI agents. It targets sales, marketing, customer support, operations and HR teams, and presents itself as built for large companies. A built-in agent, Invent, helps create the agents, their tools and their triggers. Its public pricing page now shows only the Enterprise offer, while its documentation still describes self-service Pro and Team plans.
With Belowdecks, our team installs AI agents in your business that do the repetitive work, then monitors their work and adjusts them every month. An agent does a piece of work from start to finish, such as answering a quote request received by email. Before an important action, the agent waits for your go-ahead, and it asks your team when information is missing.
Relevance AI therefore suits an organization that has people to build and maintain its agents, or that buys the support included in its Enterprise offer. Belowdecks suits a business that prefers to hand this work to our team.
Relevance AI and Belowdecks, line by line
The facts about Relevance AI come from its public pages and documentation, checked on October 6, 2026. Several features depend on the plan you choose.
What you get
- Relevance AI
- A platform to build your own agents
- Belowdecks
- Agents installed and monitored by a team
Who builds and maintains it
- Relevance AI
- You, with Invent; a deployment team on Enterprise
- Belowdecks
- The Belowdecks team, from diagnostic to follow-up
What starts the work
- Relevance AI
- Integrations, Gmail, schedule, webhook, API
- Belowdecks
- An email, a schedule, another app, a new row in a list
Your go-ahead before an important action
- Relevance AI
- Approval set tool by tool
- Belowdecks
- Set up at installation to wait for your approval
Where approval happens
- Relevance AI
- Tasks page, conversation or Slack; alerts by email or Teams
- Belowdecks
- In your Belowdecks workspace, or from an email
When information is missing
- Relevance AI
- An “Escalate to Manager” tool you turn on
- Belowdecks
- The agent asks your team and waits
Assistant for your customers
- Relevance AI
- Public widget, WhatsApp, outbound calls
- Belowdecks
- Assistant on your site; the customer confirms the request
Where data is hosted
- Relevance AI
- US, UK or Australia, chosen once
- Belowdecks
- Your Belowdecks workspace; region agreed at the diagnostic
AI bill
- Relevance AI
- Vendor credits, or your own key
- Belowdecks
- Billed in your name by the AI provider
Monitoring the work
- Relevance AI
- Tasks page, error and approval alerts
- Belowdecks
- A result for each task, a monthly report
Help and support
- Relevance AI
- Documentation in English; certified partners
- Belowdecks
- The Belowdecks team
Pricing structure
- Relevance AI
- Actions and credits; Enterprise on request
- Belowdecks
- From $449 a month, setup quoted separately
Choosing
The situation where each one fits best
Both options meet real needs. The right choice depends mostly on the time and skills your team has.
Relevance AI fits better if
Your organization has people to build and maintain several specialized agents, or buys the Enterprise support. You need outbound phone calls, WhatsApp, agent evaluations or a vendor audited for SOC 2 Type II.
Belowdecks fits better if
You want to hand over a complete piece of work, from the request received to the finished work, without building the agents yourself. Important actions must wait for your go-ahead, and our team monitors the agents’ work every month.
Common questions about Belowdecks and Relevance AI
Relevance AI can ask for approval too. What changes with Belowdecks?
In Relevance AI, each tool of an agent can run on its own, wait for approval or let the agent decide. Pending requests appear in the Tasks page or in the conversation, and an alert can notify you by email, Slack or Microsoft Teams. A task started by a trigger can also wait for approval, and you configure this setting yourself, agent by agent.
With Belowdecks, our team sets up with you, at installation, which actions must wait for your approval. The person you choose can answer from an email. Before it is turned on, each agent is tried on real cases with the instruction to send nothing.
Can we offer our customers an assistant in both cases?
Yes. Relevance AI offers a chat widget embedded with a tag, as well as WhatsApp and Telegram, plus outbound phone calls on the Team plan and above. Its documentation states that an embedded agent becomes public and indexable by search engines, that each use is billed to the account’s credits and that the widget does not support connecting integrations on behalf of each user, an Enterprise feature.
The Belowdecks assistant can also be added to your site. The customer sees only their own conversation, confirms the summary of their request themselves and reads the privacy notice you write below the conversation.
Where is the data kept in each case?
Relevance AI hosts data in the region chosen at sign-up: the United States (Virginia), Europe (London) or Australia (Sydney). That choice cannot be changed in the account, and no Canadian region is offered. With Belowdecks, your data is hosted in a region agreed at the diagnostic, which may be outside Quebec.
In both cases, the text read by the AI goes through its provider, which may process it outside Quebec. With Belowdecks, other data also leaves your Belowdecks workspace, such as technical error reports, notification emails, the monthly usage report and the agents’ web browsing. The Security page lists them, and your privacy impact assessment must take this into account.
How do the costs compare?
Relevance AI counts two things: Actions, counted each time a tool runs, even when the tool fails, and vendor credits, which pay for model usage and some tools. With your own model key, model usage no longer goes through these credits. The free plan has been withdrawn, and the public pricing page points to the sales team. Add the time of the people who build and maintain the agents.
With Belowdecks, plans start at $449 a month with Crew, which includes 15 million tokens, and the setup is quoted separately. It includes the agents our team designs for you, and the number of agents does not change the price. The AI provider bills you directly for what the model consumes. The Pricing page details the three plans.
Sources
The information about Relevance AI comes from its public pages and documentation, checked on October 6, 2026: Enterprise offer, plans, security and regions, tool approvals, alerts and escalation, how Actions are counted and embedding agents and Tasks page. Relevance AI’s features and prices change often, so check them before you decide.
Relevance AI is a trademark of OnSearch Pty Ltd. Belowdecks is not affiliated with Relevance AI.
First call
Find out whether some of your work suits agents
A free 20-minute call to describe the work that takes up your time. We will tell you whether Belowdecks fits, or whether a platform like Relevance AI suits you better.