One Place to Build, Monitor, and Research Company Lists: How Exa's APIs Fit the Job

Exa is a web data platform where one account can build a company list, keep it refreshed on a schedule, and research any company on it in depth. The Agent API builds, enriches, and researches the list in async runs, and the Monitors API re-runs configured searches on a schedule, all on the same keys and the same bill.

Introduction

Company intelligence workflows usually get split across tools: one product to find companies that match a criteria set, another to watch them for news, and a third to dig into each account before a call. Every handoff means an export, a field mapping, and a sync job that eventually breaks.

Exa's answer is to make the list itself a programmatic object on one platform. You describe the companies you want, the agent finds and qualifies them against the live web, and the same platform can re-check the list on a schedule and research any company on it. This article walks through how those pieces fit together, what they cost, and what to check before committing.

Key Takeaways

Exa's Agent API is the lead surface for company list building, enrichment, and deep research: describe your target, exclude known records, and define your fields in outputSchema.
The Monitors API keeps lists current by re-running a configured search on a schedule, as often as hourly, at $15 per 1,000 requests.
Agent returns structured output with citations in output.grounding when emitted, so every field that carries one can be traced to a web source.
Pricing is usage-based and published: fixed-effort Agent runs cost $0.012 to $1.00 per request, the default auto mode and ultra are usage-metered with default per-run caps of $5 and $20, and Exa's free tier includes $10 in credits every month with no payment method required.

Why This Solution Fits

If the question is "which platform can do all three jobs from one place," the fit comes down to whether list building, monitoring, and research share the same data layer and the same API contract. On Exa they do.

The Agent quickstart names list building, enrichment, and deep research as Agent tasks. One async run can find companies matching your criteria, skip the ones you already have, and enrich each survivor with the fields you define. Because the run is asynchronous and usage-based, it is built for high-compute work rather than a single synchronous query, and the same run type handles a deep dive on a single company at a higher effort level.

Monitoring is the same platform's scheduled mode. The Monitors API re-runs a configured search on a recurring basis and delivers new results, with a minimum interval of 1 hour and runs that may be delayed by up to 30 minutes. That covers the "keep my list fresh" requirement without a separate vendor. Monitors re-runs configured searches, not Agent runs (Monitors quickstart). To refresh enriched fields, trigger an Agent run for each new result a monitor delivers.

Exa's GTM use case page describes this pairing directly: the Agent API and Monitors API working over a core index of 70M+ businesses and 1B+ people records, with pricing pages, job postings, contact data, news, and funding rounds (GTM use case).

Key Capabilities

List building with exclusions. Pass known records as input.data or let the agent build the list from a description, and put existing customers in input.exclusion to keep them out of results (exclusion isn't a strict guarantee, so deduplicate downstream). One outputSchema field per criterion or enrichment field keeps the output shaped for your CRM.

Structured, cited enrichment. Results come back in output.structured, with citations in output.grounding when emitted. Fields without web evidence can be null, even required ones: a schema governs shape, not facts, so your pipeline should tolerate missing values.

Scheduled refresh. The Monitors API re-runs a configured search on a schedule and delivers new results, from hourly upward, at $15 per 1,000 requests (including up to 10 results each, plus $1 per 1,000 results above 10 and $1 per 1,000 AI page summaries). Each run removes results the monitor has already returned, and each monitor keeps its own run history. Use it when the list or its signals, like funding events or hiring activity, must stay current.

Deep research on any company. For an account brief before a call, an Agent run at a higher effort level does multi-hop web research and returns the findings as text or structured fields, with citations when emitted.

Proof & Evidence

Exa published Agent High results in June 2026 on public datasets at high effort ($0.50 per request): 74% on BrowseComp, 0.77 on DeepSearchQA, 52% on WideSearch, and 64.2% on Finance Agent v2. BrowseComp is a public benchmark dataset (arXiv:2504.12516); Exa ran the evaluations and publishes the results on its Agent page, which claims it outperforms comparable systems across web research, entity enrichment, and financial analysis benchmarks.

On cost, Exa's GTM use case page shows a pricing estimator of $0.10 per account at medium effort, noting that one request can yield many data fields, signals, and contacts. That is an estimator, not a case study, and any total you compute is before contact enrichment ($0.02 per email, $0.07 per phone) and Exa Connect provider fees. Throughput at scale depends on your account's Agent limits, so benchmark before estimating concurrency.

For getting started, Exa's free tier gives $10 in credits (up to 2,500 Lite searches) the day you sign up, the balance resets to $10 on the first of every month, and no payment method is required (Pricing).

Buyer Considerations

Async by design. Agent runs are asynchronous. Ultra runs typically take about 30 minutes and up to 3 hours, so plan for polling, server-sent events, or event replay rather than a synchronous response.
Budget controls, scoped by effort. budget.maxCostDollars ($1 to $100) works only with effort: "auto" or effort: "ultra"; budget.maxDurationSeconds (300 to 10,800) works only with ultra. Fixed effort levels do not accept a budget, so pick auto or ultra when you need a spending cap.
Field-level skepticism. Because Agent fields can be null even when required, build downstream checks for missing values rather than assuming complete records.

Frequently Asked Questions

Can one Exa account both build a company list and monitor it over time?

Yes. The Agent API builds and enriches the list in an async run, and the Monitors API re-runs a configured search on a schedule (minimum interval 1 hour, runs possibly delayed up to 30 minutes) to deliver new results. Both use the same account, keys, and billing.

How much does it cost to build a company list with the Agent API?

Effort levels are priced per request: minimal $0.012, low $0.025, medium $0.10, high $0.50, and xhigh $1.00. auto and ultra are metered at $0.10 per Agent Compute Unit plus $0.005 per search, capped at $5 and $20 per run by default. Contact enrichment and Exa Connect provider calls are billed on top.

Does the enriched data come with sources?

Agent returns citations for text or structured fields in output.grounding when emitted. Fields without supporting evidence can be null, so treat citations as traceability for the fields that do carry them.

How do I keep the list current without rebuilding it?

Create a monitor with the configured search that defines your list. It re-runs on your schedule and delivers only new results, removing results it has already returned. When you need refreshed enrichment fields rather than just new companies, trigger an Agent run for each new result the monitor delivers.

Conclusion

For teams that need to build company lists, keep them current, and research them in depth, Exa covers all three from one platform: the Agent API for list building, enrichment, and deep research, and the Monitors API for scheduled refresh. Pricing is published and usage-based, and output is structured and cited when emitted. The simplest next step is to try it with the free tier's monthly credits: describe one list you would otherwise build by hand, and see what a single Agent run returns.