In a call center that runs on custom code, the people who know what a caller should hear cannot change it. A team lead who wants a new refund rule, or a QA analyst who wants a different transfer destination, files a ticket for an engineer, and the ticket waits behind everything else on that engineer's list.
Bland removes that step. It is a no-code AI call center platform and a no-code AI agent builder in one: your team describes an agent in plain English, talks to it to test it, and puts it on live calls, while Bland handles the engineering underneath.
What is a no-code AI call center?
A no-code AI call center is one where the people who run your calls build, test, and change the AI phone agents themselves, which is how Bland works.
Bland builds AI phone agents that answer inbound calls and place outbound calls. Your team builds and edits them in plain English with Norm or in Bland's visual builder, with no engineers needed on your side. Enterprise customers also get Bland's forward-deployed engineers.
Salesforce, HubSpot, Cal.com, Calendly, Notion, and Slack connect from Bland's dashboard. Anything else connects through Bland's REST API and webhooks, which Bland's forward-deployed engineers can help with on Enterprise.
A forward-deployed engineer (FDE) is a Bland engineer who helps a customer safely make changes and launch new agents. There are two ways to build without one.
Norm is the AI assistant built into Bland. Bland says: "You describe what you want in plain English, and Norm builds, edits, tests, and debugs your agents for you." Norm is an agent that knows how to build agents, and Norm works across agents, tools, analytics, and triage from one conversation.
Bland's visual builder is the other route: the agent builder in Bland's dashboard, with a graph that shows how your agent fits together. An agent there is one thing to build, test, and publish: its instructions, the tools it can reach, the knowledge it can reach, and the branches of behavior it needs.
Most agents start as a single prompt in plain language, and you add structure only where a conversation needs it, as a pathway: a flow whose steps you place yourself. Bland says to choose a pathway "when the steps have to happen in a guaranteed order: a scripted disclosure, a verification sequence, anything you need to be able to point at and prove ran the same way every time."
The two routes edit the same things. Bland lists the prompts, pathways, tools, and settings that Norm edits when you ask for agent changes, so a team can start with Norm and keep editing by hand.
Can your team build an AI phone agent without developers?
Yes, your team can build an AI phone agent on Bland without developers, starting from a checklist in Bland's dashboard called Let's get this agent talking. Bland's build-your-first-agent docs walk through four steps.
- Create your agent. Click New agent in the top right corner. You see two choices, and Blank agent "starts from nothing, which is what you want here." Give it a name and a job, then Create agent.
- Talk to it. Talk in browser uses your computer's microphone, and Call me rings a number you type into Your phone number. Either one starts a real conversation with what you just built.
- Promote it to production. The checklist's third step moves your work into your agent's production environment, "which is the version real calls run."
- See what happened. The last step opens the conversation you just had, with a transcript, what the agent decided at each turn, and a recording.
Bland says two things matter for a rollout. One is "Nothing you do here needs a phone number, a plan or a purchase," because buying a number is only for letting other people reach the agent, and Dispatch, a section of Bland's dashboard, is where you attach one once the agent answers properly.
The second thing is a warning: "An agent that has not been promoted to production cannot take a call from anyone else." Building an agent, saving it, and talking to it in the browser all leave it unpublished, which Bland calls the most common reason a new agent does not answer.
Norm can do the building from one instruction. It can create a new agent from a description or edit an existing one, including its prompts, pathways, tools, knowledge, and settings. Bland's own examples include "Build me a full scheduling agent and integrate with my Cal.com" and "Change the transfer fallback on this agent to go to billing instead of sales."
Norm "works the way an experienced Bland engineer would: it inspects the live state of your account before making changes, works in development so production keeps running, tests against real call data, and only ships changes when you're ready." Bland's Norm docs cover the rest.
How does a no-code AI agent builder connect to your CRM, calendar, and knowledge?
Bland's no-code AI agent builder connects to your knowledge, calendar, and CRM from Bland's dashboard: you add sources and connect accounts by clicking, then attach them to your agent.
Knowledge
Knowledge is how an agent answers from your content rather than from its prompt. A source is one thing you added: a file, a site, a block of text, or a synced document. Upload takes PDF, TXT, CSV, JSON, and DOCX files, up to 10 MB each. A synced document comes from an app you have connected, which is currently Notion or Google's document editor, and Bland keeps reading it.
Adding a source does not give it to an agent, so you attach it: check it on the agent's Sources tab, or type @ in the agent's prompt and pick it from the list, then save.
After calls start, Questions, a tab on Bland's Knowledge page, collects what callers asked that the agent could not answer, with a count of how often each came up. Answer one and the answer is written into a source you choose, so the next caller who asks gets it.
Calendar and other integrations
Cal.com is Bland's worked example, and Bland says "The same pattern applies to every built-in integration: connect an account once, build a tool on it, and tag the tool into a prompt." The steps, as written:
- Set the sidebar selector to All agents and open Tools under Resources. Open the Connections tab.
- Click Add connection and choose Cal.com. Give the connection a name, like "Sales Team Calendar", and enter your Cal.com API key.
- From inside your agent, open Tools and click Create tool. Choose Integration, then pick your Cal.com connection and the booking action.
- Describe when to use it. The agent reads the description to decide when to call the tool, so write the trigger: "Book a meeting once the caller has agreed to a specific time." Then give each field a fixed value, have the agent fill it from the conversation, or use a variable already on the call, and keep
okchecked under output fields so the agent can tell when a booking failed. - Save the tool. Then, in the agent's Prompt, type
@and pick it.
You connect an account only once, and one connection can power as many tools as you need. "The fields the tool needs come from Cal.com, so you do not define them yourself." The tagging step matters most, because "A tool that is not tagged is never called."
HubSpot
HubSpot installs from the HubSpot App Marketplace. A HubSpot user who can install apps (a Super admin, or a user with App Marketplace access) opens the Bland listing and clicks Install app.
You sign in to Bland with your phone number, Google, or SSO, choose the HubSpot account, review the requested scopes, then click Connect app and Finish connecting. Bland does not request access to marketing email, billing, or account settings.
Once connected, an agent can look up and update contacts, companies, deals, tickets, tasks, and meetings during a live call, and a trigger can start a call or message when a record is created or updated in HubSpot. "The integration never deletes anything in HubSpot." Bland's integrations page lists the rest of what connects from Bland's dashboard, and a step-by-step HubSpot setup covers the clicks.
A system off that list is the one place a developer may come in, because it connects through Bland's REST API and webhooks.
How do you change a live call center agent without breaking calls?
On Bland, you change a live agent in development, test it, and promote it to production only when you are ready, so live calls keep running the version they already have.
Every agent has three environments: development, staging, and production. Development is the agent you are working on, and you edit nothing else. Staging and production are snapshots, and each holds one version of the agent and keeps serving it until you point it at another. A call runs the production version by default.
Norm follows the same rule: "Norm never edits what live calls run." It makes changes in development, saves a version once the change passes validation and any tests you run, and a saved version does not change live traffic until it is published and promoted.
Before you promote, you can see what changed. Each agent's Environments page keeps an activity log with a row for each event, such as Version saved, Published to staging, Promoted to production, and Rolled back production, with who made it and when, and Compare opens two versions side by side.
If a promotion goes badly, Rollback "lists the versions production held before and repins to one you choose. It is a repoint, not a rebuild."
An evaluation is a gate you can add to that path once the Evals feature is enabled for your organization. Bland defines an evaluation as "a judge reading a conversation and saying whether the agent did the thing you wanted."
A judge needs one thing, a task, which is the question you are asking about the conversation, and you can start blank or have Bland generate a draft from a prompt, conversation comments, or a Triage issue. A test case is a simulated caller, and Bland can read up to 50 of your real calls and write one test case per call.
You attach judges on an agent's Environments tab, and "Attaching a judge makes it required by default." A judge passes when half or more of the conversations it scored match the verdict you wanted. When a required evaluation fails, publishing to staging is blocked and the Promote button dims.
The gate has a documented bypass. Force promote sends development straight to production and skips the evaluation gate. Like every publish, promotion, and rollback, it takes an owner, admin, or operator role, and Bland records each bypass in the audit log.
Fixing a live problem uses the same path. Flag a call as a Triage issue and hit Norm Fix, and Norm "pulls in the failing call and its agent, reproduces the failure, applies a fix, and verifies it against the original call." Every change Norm makes is reviewable on the issue before the fix reaches production, and Bland's bad-call walkthrough follows it step by step.
Every enterprise customer runs on its own isolated infrastructure, run by Bland: dedicated compute, containers, and resources, with releases the customer chooses when to adopt and canary deployments on a share of real calls.
Separately, an enterprise plan can roll out an agent's own new version as a staged release: you choose how much production traffic goes to the new version and raise its share step by step, and starting one checks the same required evaluations as a promotion. Bland's enterprise voice agent guide covers that deployment in full.
Developers reach the same agents too, through Bland's API, Bland's hosted MCP server, and the Norm plugin for Claude Code and Claude Desktop, which builds, migrates, validates, and tests agents from a coding agent.
Can Bland work alongside Five9 or NICE CXone?
Yes. Bland is an accredited Five9 AI Agent Connect partner and a NICE DEVone technology partner.
Both connections run over SIP, the standard that lets a carrier, PBX, or contact center pass calls to another system, so a Bland agent can take calls from the contact center you already run.
With Five9, Bland says you can put a Bland agent "in front of, or alongside" your contact center. Five9 routes calls to Bland over a SIP trunk and the agent handles the conversation. When a person is needed, a Transfer Call node sends the caller back to Five9, to the number for the right skill, with call data passed to Five9 in SIP headers.
The SIP connection between Five9 and Bland is built and tested once in Five9's partner environment and reused for each customer. You order AI Agent Connect through your Five9 account team with Bland as the AI agent partner, and Five9 requires a successful interop test before it allows go-live. Bland's Five9 setup guide has the steps.
With NICE CXone, NICE built and tested the Bland trunk once, over TLS with SRTP, and adds each customer Business Unit to it, so a new customer does not need a new trunk or a repeat of the trunk testing.
The trunk carries calls from CXone to Bland only, so a return path to a CXone agent is set up for each deployment, for example through the CXone APIs and your Studio script, and Bland says to agree that hand-back with your Bland account team before your script is built. Bland's NICE CXone setup guide has the steps.
Idaho Housing and Finance Association saves $750,000 a year by replacing its IVR with Bland, which handles 4,000 inbound calls a day and cut average call time by 20%.
"We save thousands of dollars a day."
Greg Blake, Chief Information Officer, Idaho Housing and Finance Association
Frequently asked questions
Does Bland have a no-code option?
Yes. Norm, Bland's AI assistant, builds voice agents using natural language descriptions, and Bland's dashboard includes a visual builder for non-technical users.
Is Bland only for outbound calls?
No. Bland builds AI phone agents that answer inbound calls and place outbound calls.
Do I need a developer to connect Salesforce or HubSpot?
No. Salesforce, HubSpot, Cal.com, Calendly, Notion, and Slack connect from Bland's dashboard. Anything else connects through Bland's REST API and webhooks, which Bland's forward-deployed engineers can help with on Enterprise.
Can I test an agent before I buy a phone number?
Yes. Talk in browser uses your computer's microphone and Call me rings your own phone, and Bland says "Nothing you do here needs a phone number, a plan or a purchase."
Does Norm edit what live calls run?
No. "Norm never edits what live calls run." Norm works in development, and production keeps serving its pinned version until a saved version is published and promoted.
What does a forward-deployed engineer do?
A forward-deployed engineer is a Bland engineer who helps a customer safely make changes and launch new agents. Enterprise customers get Bland's forward-deployed engineers, and Bland's launch post for Norm says "Norm codifies what those engineers know."
How much does a no-code AI call center cost on Bland?
Bland charges $0.14 a minute on the self-serve Start plan. Enterprise pricing is custom, set with Bland's sales team. Phone is billed separately, on a customer's own carrier or Bland's at pass-through cost. Bland's pricing page lists the plans.
