Landbase
Landbase is a B2B data platform that finds and qualifies prospects from a plain language description of the buyer. Agentic search turns a prompt into structured criteria across more than 300 million contacts and 1,500 enrichment fields, lookalikes find companies that share traits with a customer's best accounts, signals flag events such as hiring or funding, and AI qualification scores each account with a written reason. Contact details come through a waterfall of more than 20 providers powered by BetterContact, with emails checked by SMTP and inbox tests and phones checked for active lines.
The models behind it belong to Landbase's own GTM Omni family, which the company says was trained on tens of millions of past campaigns and refined with human and performance feedback, with a third version announced in September 2026. Although the company now describes itself as a data layer, its Execution module still runs AI personalized email, LinkedIn and dialer campaigns from provisioned, warmed sending domains, and a command line tool lets Claude Code and Codex search, enrich and launch campaigns. Results sync to Salesforce, HubSpot and Outreach and export as CSV, JSONL or Parquet.
Plans start with 1,000 free credits and run from $499 a month for 15,000 credits to $4,999 for 200,000, with searches and misses free, verified emails about three cents and verified mobiles about 33 cents. Landbase, Inc. was cofounded by chief executive Daniel Saks, a cofounder of AppDirect, works from Palo Alto and Toronto, and has raised more than $42.5 million, including a $30 million Series A in June 2025 led by Sound Ventures and Picus. Named customers include the logistics software company Octup and the outbound agency Martal Group.
Capability Axes
Capability grades
17 of 17 axes rated · 9 graded A or B
Agentic search turns prompts into structured criteria, AI qualification scores and explains each account, lookalikes extract traits from a customer's best accounts and research agents query more than 50 sources. These run on Landbase's GTM Omni models, a domain language model with action, planning and reward models refined on campaign outcomes. The contact data underneath comes from a provider waterfall and is sold by the credit on its own.
A campaign cannot launch until every contact has a full message set, sends are throttled by inbox and LinkedIn limits and a send window, and launching a teammate's campaign needs explicit confirmation. Enrichment and imports ask for confirmation, records that cannot be verified go to manual review, and the command line installer asks whether an AI assistant may run commands unprompted, defaulting to no. How a CRM push matches and overwrites records, and whether it can be previewed or undone, is not described.
The GTM Omni family pairs a domain model with action, planning, generator and reward models refined with human and performance feedback, though no base model or outside AI provider is named, and customer data is shared with unnamed third party AI models to write emails. Qualification shows weighted scores, disqualifying reasons and plain language explanations, and lookalike traits are weighted by revenue, conversion and retention. Training figures range from 40 to 50 million campaigns or messages.
Octup reports qualification 30 percent more accurate than its contractors and 50 meetings in its first weeks, and Martal Group reports replacing 13 tools with one and saving $350,000 a year, with client campaigns booking 107 and 139 meetings. GTMBench, Landbase's own test of 26 prompts judged by an OpenAI model, puts its precision at 76.7 percent against 47.9 for Clay and 36.0 for Apollo, with no independent check. Counts vary widely, from 150 paying customers in June 2025 to thousands of revenue teams, and from 150 million to 3 billion records.
The terms put CAN-SPAM, CASL and TCPA compliance on the customer, require an opt out in every campaign and state that Landbase has obtained no consents on the customer's behalf. Nothing describes Do Not Call scrubbing for the dialer or mobile numbers, an opt out for the people in the database or a data broker registration.
Privacy terms unchanged since 2023 cover Landbase's own users, place hosting in the US with transfers by consent and share customer data with third party AI models that are not named. No lawful basis or rights process is described for the hundreds of millions of people in the database, a DPA is promised where GDPR applies but not public, and no subprocessor list is available.
Contact data routes through more than 20 providers in a waterfall powered by BetterContact, among them ZoomInfo, Apollo, Clearbit and Lusha, and research agents draw on BuiltWith, Wappalyzer, job boards, SEC filings and LinkedIn activity. Emails pass SMTP and inbox checks with catch all detection, and phones are checked for carrier and active lines. Licensing terms with these providers are not described, and record counts range from 300 million contacts to 3 billion records.
LinkedIn profile links serve as lead inputs, LinkedIn activity feeds qualification, premium datasets add LinkedIn posts and comments, and campaigns can send connection requests and messages from LinkedIn accounts linked in the web app. How LinkedIn data is collected is not described, though Landbase's own benchmark rows were scraped fresh from LinkedIn, and the terms simply require customers to follow third party terms.
Whether customer campaigns and data train the GTM Omni models is not stated, though the company says the models learn from live campaign outcomes and improve with use. The terms leave title to customer data with the customer while granting Landbase perpetual use of anonymized usage information, telemetry from the command line tool is opt in, and customer data is deleted after a 30 day export window at termination.
Execution sends AI personalized email from provisioned inboxes, LinkedIn messages and dialer calls, and partner agencies run full service outbound on the platform. Whether recipients are told a message was written by AI is not addressed, and no position on Article 50 of the EU AI Act is taken.
Salesforce and HubSpot sync and pushes to Outreach sequences are offered, with Slack, Gong and Salesloft also connecting, but the 100 or more integrations claimed go unlisted and sync behavior is not described. The command line tool offers browser or API key sign in, batches of 100 leads at about ten a minute and exports to JSONL, CSV and Parquet, while webhooks, an MCP server, Clay and Zapier are absent from the integration set.
Data is hosted in the United States. The cloud provider, regions and residency choices are not described.
SOC compliance is asserted without a report type, and the Vanta trust center gives an outside reader no certification type, auditor or date. Encryption, single sign on and multi factor sign in are not described.
Monthly plans cost $499 for 15,000 credits, $1,999 for 75,000 and $4,999 for 200,000, with enterprise priced on request and a free start of 1,000 credits that never expire. Searches, lookalikes and misses cost nothing, AI qualification costs a fifth of a credit, a verified email about three cents and a verified mobile about 33 cents. The free start is also put at $49 of credits, dataset creation runs 25 to 50 credits at dollar figures that do not reconcile, and no annual pricing is offered.
Contracts renew for a year unless notice is given at least 30 days before the term ends, cannot be ended for convenience and are non refundable unless an order form says otherwise. Customers have 30 days after termination to export their data in CSV, JSONL or Parquet before it is deleted, while rights to enriched contact data after the contract ends are not addressed. California law and San Francisco courts govern, and liability is capped at twelve months of fees.
Execution provides provisioned sending domains and managed inboxes that are warmed and monitored, and for Martal Group Landbase bought domains, set up DMARC and ran warmup. Sends are throttled by inbox limits and a send window, the terms penalize excess bounces and complaints, and emails are verified with a stated 97.4 percent deliverability and charged only when verified. Daily volume limits and what happens when a domain's reputation drops are not specified.
Revenue operations, sales, marketing and growth teams are the buyers, in software, logistics technology and cybersecurity, with enterprise and private equity named for the newest model and coverage described as global. Named customers are Octup and Martal Group, with testimonials from Vangates and Ordesa AI and agency partners such as Atlantic Growth Solutions, and the company counted 150 paying customers in June 2025.
Alternatives to Landbase
Vendors in the same categories as Landbase whose graded profiles look most alike, with a line on what each product does.
Similar is not the same, so check that each one does the job you need. No vendor pays for placement.
AI Ark is a B2B data platform built around AI search.
Cleanlist is a B2B enrichment and prospecting platform that runs a waterfall across more than 25 contracted data providers instead of keeping its own contact database.
Clodura Systems Private Limited, trading as Clodura.AI, was founded in 2015 in Pune and now positions itself as a single go to market platform running the whole motion from database to closed deal.
Czech prospecting platform combining autonomous discovery agents, a named enrichment waterfall and multichannel email and professional network sequencing in one product.
SalesIntel sells B2B contact and company data checked by a research team of more than 2,000 people, covering over 200 million contacts and 54 million mobile numbers.
SyncGTM is a go to market orchestration platform built around a table in which every column is an action, in the style of Clay.
Ordered by how closely the graded profiles match across the seventeen axes, inside the same categories. See how each axis is graded in the methodology.
Pricing
What this vendor charges, what it commits to in writing, and where the bill can move. Figures the vendor publishes itself are labeled Vendor Published. Figures labeled Estimated come from other sources and the vendor has not confirmed them.
- ›Landbase sells monthly credit plans at $499 for 15,000 credits, $1,999 for 75,000 and $4,999 for 200,000, with enterprise priced on request. Searching costs nothing, a verified email costs about three cents and a verified mobile about 33 cents, and new accounts start with free credits.
How the price works
What you are charged for, and what makes the bill go up.
Monthly credit plans of $499, $1,999 and $4,999; credits charged per action; searches and misses free; enterprise quoted
What the contract says about your data
What the vendor commits to in writing once your data is in the product.
No BAA tier listed; a DPA is offered where GDPR applies but not published.
Getting started
What it costs and what is included before the product is running.
No setup fee listed.
What to watch for
Where this pricing can surprise a buyer who has not read it closely.
A free start gives 1,000 credits with no card and no expiry, also put at $49 of credits. Verified emails cost about $0.033 and verified mobiles about $0.33, AI qualification costs 0.2 credits, and no annual pricing is offered. Order form contracts renew for a year unless cancelled 30 days ahead and are non refundable.