<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[Leadsets]]></title><description><![CDATA[Unlimited B2B Contact Data for AI Agents, Starting from $29/mo. Search, enrich and verify work emails over MCP or REST. Free: 1,000 records, no card.]]></description><link>https://leadsets.hashnode.dev</link><generator>RSS for Node</generator><lastBuildDate>Sat, 21 Nov 2026 15:39:09 GMT</lastBuildDate><atom:link href="https://leadsets.hashnode.dev/rss.xml" rel="self" type="application/rss+xml"/><language><![CDATA[en]]></language><ttl>60</ttl><item><title><![CDATA[Email verification API: what verified, catch-all, unknown and invalid mean]]></title><description><![CDATA[Short answer: an email verification API checks whether a mailbox can receive mail without sending anything to it, and returns a status. Verified means the…]]></description><link>https://leadsets.hashnode.dev/email-verification-api-what-verified-catch-all-unknown-and-invalid-mean</link><guid isPermaLink="true">https://leadsets.hashnode.dev/email-verification-api-what-verified-catch-all-unknown-and-invalid-mean</guid><category><![CDATA[email]]></category><category><![CDATA[api]]></category><category><![CDATA[sales]]></category><category><![CDATA[SaaS]]></category><category><![CDATA[AI]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Sat, 21 Nov 2026 06:47:43 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> an email verification API checks whether a mailbox can receive mail without sending anything to it, and returns a status. <strong>Verified</strong> means the mailbox exists and accepts mail, so it is safe to send. <strong>Catch-all</strong> means the domain accepts every address, so this particular mailbox cannot be confirmed. <strong>Unknown</strong> means the check could not decide, often because the server timed out. <strong>Invalid</strong> means the address does not exist or cannot receive mail, so do not send. The useful part is treating those four differently instead of rounding them into "good" and "bad". Leadsets returns all four, with a confidence score and a <code>checked_at</code> date, through a REST API and a remote MCP server for AI agents. <a href="https://www.leadsets.ai/signup">Get a free API key</a> to try it.</p>
<h2>Why verification matters</h2>
<p>Sending to addresses that do not exist produces bounces, and a high bounce rate hurts the sender reputation of your domain. Verification before sending is the cheap fix. It matters even more when an AI agent builds and sends lists, because an agent will happily email everything it is given unless the data tells it what is safe.</p>
<p>Verification answers "can this mailbox receive mail?". It does not tell you whether the person still works there or wants to hear from you. Freshness (when the record was last checked) and relevance (why the person is on the list) are separate questions.</p>
<h2>How verification works, briefly</h2>
<p>A verification service looks up the domain's mail servers, then talks to the receiving server the way a sending server would, stopping before any message is delivered. The server's answer tells the service whether the mailbox exists. Some servers give a clear yes or no. Some accept every address. Some do not answer in time. Those three cases are why there are more than two statuses.</p>
<h2>The four statuses</h2>
<h3>Verified</h3>
<p>The mailbox exists and accepts mail. Safe to send. In Leadsets the API value is <code>verified</code>.</p>
<h3>Catch-all</h3>
<p>The domain is configured to accept mail for any address, so the receiving server says yes to everything, including addresses that do not exist. The check cannot confirm this specific mailbox. The API value is <code>catch_all</code>.</p>
<p>Catch-all is common at companies, so throwing these away can cost you a real share of a list. Sending to all of them blindly is risky too. A sensible rule: send to catch-all addresses only when the rest of the record is strong (the person matched your filters and the data is recent), keep volume low, and watch bounces.</p>
<h3>Unknown</h3>
<p>The check could not decide, for example because the server timed out or the inbox is full. The API value is <code>unknown</code>. Hold these and check again later rather than sending now.</p>
<h3>Invalid</h3>
<p>The address does not exist or cannot receive mail, including bad syntax. The API value is <code>invalid</code>. Do not send, and remove it from your list.</p>
<h2>A simple decision table</h2>
<table>
<thead>
<tr>
<th>Status</th>
<th>Send?</th>
<th>What to do</th>
</tr>
</thead>
<tbody><tr>
<td><code>verified</code></td>
<td>Yes</td>
<td>Send.</td>
</tr>
<tr>
<td><code>catch_all</code></td>
<td>With care</td>
<td>Send only with a strong record and low volume, or hold for review.</td>
</tr>
<tr>
<td><code>unknown</code></td>
<td>Not yet</td>
<td>Hold and re-check later.</td>
</tr>
<tr>
<td><code>invalid</code></td>
<td>No</td>
<td>Drop.</td>
</tr>
</tbody></table>
<p>An agent can apply this table on its own. That is the main reason to prefer an API that never rounds uncertain results up to valid. Leadsets does not, and it does not serve invalid emails as contacts.</p>
<h2>Calling the API</h2>
<p>One email or up to 100 per call. Duplicates are checked once, and addresses are lower-cased before checking.</p>
<pre><code class="language-bash">curl https://api.leadsets.ai/v1/emails/verify \
  -H "Authorization: Bearer $LEADSETS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"emails": ["maya@northwind.io", "sales@stackfold.io", "old.address@parcelry.co"]}'
</code></pre>
<p>With the Python SDK (<code>pip install leadsets</code>):</p>
<pre><code class="language-python">import os
from leadsets import Leadsets

ls = Leadsets(os.environ["LEADSETS_KEY"])
res = ls.emails.verify(
    emails=["maya@northwind.io", "sales@stackfold.io", "old.address@parcelry.co"],
)
print(res.summary)
</code></pre>
<p>The TypeScript SDK (<code>npm i @leadsetsai/sdk</code>) has the same call as <code>ls.emails.verify({ emails: [...] })</code>.</p>
<h2>Reading the response</h2>
<p>Each entry in <code>results</code> has the <code>email</code>, its <code>status</code>, a <code>confidence</code> between 0 and 1, <code>checked_at</code>, and <code>mx_provider</code> (who hosts the mailbox, when known). In the example from the Leadsets docs, the first address comes back <code>verified</code> with confidence 0.97, the second <code>catch_all</code> with 0.5, and the third <code>invalid</code> with 0.</p>
<p>A <code>summary</code> counts each status plus <code>not_checked</code>. If you run out of records partway through a batch, the remaining emails come back as not checked, with a reason, and cost nothing. If no email in the call could be checked, the API answers <code>503 unavailable</code> and nothing is counted.</p>
<p>Use <code>checked_at</code> as well as <code>status</code>. A verified result from months ago is weaker evidence than one from this week.</p>
<h2>Verification inside search and enrichment</h2>
<p>You often do not need a separate verification step. When Leadsets returns a person from <code>search_people</code> or <code>enrich_person</code>, the email already carries a status and a <code>checked_at</code> date, and <code>has_verified_email: true</code> on a people search returns only people with a verified work email. Standalone <code>verify_email</code> is for lists you bring from elsewhere: a CRM export, event leads or an old spreadsheet.</p>
<h2>Using it from an AI agent</h2>
<p>Add <code>https://mcp.leadsets.ai</code> as a custom connector in Claude, Claude Code, Cursor, ChatGPT or any MCP client (<a href="https://www.leadsets.ai/docs/mcp">setup guide</a>). Then ask:</p>
<blockquote>
<p>Verify these emails with Leadsets. Put verified ones in a table, put catch_all and unknown in a second table labelled hold, and drop invalid ones.</p>
</blockquote>
<p>The agent calls <code>verify_email</code>, reads each status and sorts the list. Ask it to call <code>get_usage</code> first if the list is large; that call costs nothing.</p>
<h2>What it costs</h2>
<p>Each email that gets a status uses one enrichment record on your plan, including emails rejected for bad syntax. Emails that were not checked cost nothing. Leadsets plans are a flat monthly fee with no credits: Free (1,000 enrichments plus 100 search records, no card), Solo $29/mo, Pro $99/mo, Team $399/mo and Max $999/mo, with visible fair-use limits per rolling five hours and per week. See <a href="https://www.leadsets.ai/pricing">pricing</a> and the <a href="https://www.leadsets.ai/docs">docs</a>.</p>
<h2>FAQ</h2>
<p><strong>What does catch-all mean in email verification?</strong>
The domain accepts mail for every address, so the check cannot confirm that a specific mailbox exists. Treat it as "send with care", not as verified.</p>
<p><strong>Is an unknown result the same as invalid?</strong>
No. Unknown means the check could not decide, often because of a timeout. Re-check later. Invalid means the address cannot receive mail.</p>
<p><strong>Does verification send an email to the address?</strong>
No. A standard verification check talks to the receiving mail server and stops before any message is delivered.</p>
<p><strong>How many emails can I verify in one Leadsets call?</strong>
Up to 100 different emails per call. Each one that gets a status uses one enrichment record.</p>
<p><strong>Do emails from Leadsets search need separate verification?</strong>
Usually not. They already carry a status and a <code>checked_at</code> date. Use <code>verify_email</code> for lists that came from somewhere else. <a href="https://www.leadsets.ai/signup">Get a free API key</a> to start.</p>
]]></content:encoded></item><item><title><![CDATA[People search API: how to find B2B contacts by title, seniority and company size]]></title><description><![CDATA[Short answer: a people search API takes structured filters (job title, seniority, function, company size, industry, country) and returns matching business…]]></description><link>https://leadsets.hashnode.dev/people-search-api-how-to-find-b2b-contacts-by-title-seniority-and-company-size</link><guid isPermaLink="true">https://leadsets.hashnode.dev/people-search-api-how-to-find-b2b-contacts-by-title-seniority-and-company-size</guid><category><![CDATA[api]]></category><category><![CDATA[sales]]></category><category><![CDATA[SaaS]]></category><category><![CDATA[AI]]></category><category><![CDATA[mcp]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Sat, 21 Nov 2026 06:46:21 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> a people search API takes structured filters (job title, seniority, function, company size, industry, country) and returns matching business contacts as JSON, so a script or an AI agent can build a prospect list without anyone clicking through a database UI. The filters that matter most are fuzzy title matching, a fixed seniority scale, a headcount range and a "verified email only" switch. Leadsets is a B2B contact data API and remote MCP server built for this: Unlimited B2B Contact Data for AI Agents, Starting from $29/mo. You can <a href="https://www.leadsets.ai/signup">get a free API key</a> and run the examples below on the Free plan.</p>
<h2>What a people search API does</h2>
<p>A people search API answers one question: "who matches this description?" You send a set of filters and get back a page of people, each with a name, a title, a company and usually a work email. Good APIs also tell you why each record matched and how fresh the data is.</p>
<p>That is different from enrichment. Enrichment starts with someone you already know (a name and a company, or an email) and fills in the missing fields. Search starts with a description of the people you want and finds them. Most prospecting workflows use both: search to build the list, enrichment and verification to finish each record.</p>
<p>Well-known providers in this space include Apollo, ZoomInfo, People Data Labs and Lusha, and tools like Clay sit on top of several data sources. They differ in coverage, interface and pricing model, so compare them on the filters you actually need and on how cost behaves when an agent runs many searches.</p>
<h2>The filters that matter</h2>
<h3>Title, with fuzzy matching</h3>
<p>People write the same job in many ways: "VP Sales", "Vice President, Sales", "VP of Sales and Partnerships". An exact-match title filter misses most of them. Look for fuzzy title matching so one value catches the common variants. In Leadsets, <code>title</code> is fuzzy-matched, so "VP Sales" also finds "Vice President, Sales".</p>
<h3>Seniority and function</h3>
<p>Titles are messy, so a fixed seniority scale is the more reliable filter. Leadsets uses <code>c_level</code>, <code>vp</code>, <code>head</code>, <code>director</code>, <code>manager</code> and <code>ic</code>. A separate <code>function</code> filter (<code>revenue</code>, <code>marketing</code>, <code>engineering</code>, <code>product</code>, <code>ops</code>, <code>finance</code>, <code>hr</code>) lets you ask for "heads of revenue" without listing every sales title.</p>
<h3>Company size</h3>
<p>Headcount is the most common way to describe an ideal customer profile. Leadsets takes <code>employees</code> as a range, such as <code>"50-200"</code>.</p>
<h3>Industry, country and what the company does</h3>
<p><code>industry</code> takes slugs such as <code>"saas"</code>, and <code>country</code> takes ISO 3166-1 alpha-2 codes such as <code>"US"</code>. When an industry label is too broad, <code>company_keywords</code> matches whole words in the company name, description and specialties (for example <code>["gtm engineering"]</code>), and <code>exclude_company_keywords</code> removes companies you do not want.</p>
<h3>Verified email only</h3>
<p>If the list is going to a sequencer, you only want people you can actually email. <code>has_verified_email: true</code> returns only people with a verified work email. Every email in the response carries a status and a <code>checked_at</code> date.</p>
<h3>Freshness</h3>
<p>People change jobs. <code>max_age_days</code> limits results to records refreshed within a given number of days, and <code>exclude_stale</code> leaves out records that are due for a refresh.</p>
<h2>A first search</h2>
<p>Ten revenue leaders (VP or Head) at US SaaS companies with 20 to 200 employees, using the REST API:</p>
<pre><code class="language-bash">curl https://api.leadsets.ai/v1/people/search \
  -H "Authorization: Bearer $LEADSETS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"function": ["revenue"], "seniority": ["vp", "head"],
       "employees": "20-200", "country": "US", "industry": "saas",
       "limit": 10}'
</code></pre>
<p>The same call with the Python SDK (<code>pip install leadsets</code>):</p>
<pre><code class="language-python">import os
from leadsets import Leadsets

ls = Leadsets(os.environ["LEADSETS_KEY"])
res = ls.people.search(
    function=["revenue"],
    seniority=["vp", "head"],
    employees="20-200",
    country="US",
    industry="saas",
    limit=10,
)
print(res.items[0].match_reasons, res.usage)
</code></pre>
<p>And with TypeScript (<code>npm i @leadsetsai/sdk</code>), the method is <code>ls.people.search({...})</code> with the same fields.</p>
<h2>Reading the response</h2>
<p>Each item has a <code>person_id</code>, <code>full_name</code>, <code>title</code>, a <code>company</code> object, an <code>email</code> object with <code>address</code>, <code>status</code> and <code>checked_at</code>, and <code>match_reasons</code>, for example <code>["title", "size", "country"]</code>. Match reasons matter more than they look. An agent can drop records that matched on fewer criteria than you asked for, and a person reviewing the list can see at a glance why someone is on it.</p>
<p>The response also includes <code>next_cursor</code> for the next page, a <code>usage</code> object with what the call counted and what is left, and a <code>request_id</code> you can send to support if something looks wrong.</p>
<h2>Paging and sizing a market</h2>
<p><code>limit</code> goes from 1 to 100 (default 25; at most 10 per call on the Free plan). Pass <code>next_cursor</code> back as <code>cursor</code> to get the next page.</p>
<p>Before pulling a big list, size it. Set <code>count_only: true</code> and Leadsets returns only how many people match (counting stops at 10,000). A count uses no records, so an agent can try several filter combinations and pick the one that gives a list of the right size.</p>
<h2>Best contact at one account</h2>
<p>Account-based teams usually want "the best person at this company", not a list. The <code>priorities</code> parameter takes one <code>company_domain</code> and up to ten groups of titles, best first, for example <code>[{"title": ["founder", "ceo"]}, {"title": ["head of growth", "vp sales"]}]</code>. Results fill from the first group, then the next, never repeat a person and put verified emails first.</p>
<h2>Using it from an AI agent</h2>
<p>The same search is available as the <code>search_people</code> tool on the Leadsets remote MCP server. Add <code>https://mcp.leadsets.ai</code> as a custom connector in Claude, Claude Code, Cursor, ChatGPT or any MCP client (<a href="https://www.leadsets.ai/docs/mcp">setup guide</a>), then ask in plain language:</p>
<blockquote>
<p>Find 25 heads of marketing at US B2B SaaS companies with 50 to 500 employees, only people with a verified work email. Show name, title, company, match reasons, email and checked_at.</p>
</blockquote>
<p>The agent turns that into filters, calls the tool, and can page or refine on its own. Ask it to call <code>get_usage</code> first; that call costs nothing and shows what is left in the current window.</p>
<h2>What it costs</h2>
<p>Leadsets bills a flat monthly fee, with no credits and no per-record bill: Free (1,000 enrichments plus 100 search records, no card), Solo $29/mo, Pro $99/mo, Team $399/mo and Max $999/mo. Paid plans have visible fair-use limits per rolling five hours and per week. A search uses one search record per record returned and one request per call. An empty search uses no records. Details are on the <a href="https://www.leadsets.ai/pricing">pricing page</a> and in <a href="https://www.leadsets.ai/docs/limits">limits and costs</a>.</p>
<p>Coverage is work emails, roles and company data. Leadsets does not provide phone numbers.</p>
<h2>FAQ</h2>
<p><strong>What filters does a good people search API need?</strong>
Fuzzy title matching, a seniority scale, function, a company headcount range, industry, country, and a switch for verified emails only. Freshness filters help when job changes matter.</p>
<p><strong>What is the difference between people search and enrichment?</strong>
Search finds people who match a description. Enrichment fills in details for a person you already have. Most workflows search first, then enrich and verify.</p>
<p><strong>Can I search only for people with a verified work email?</strong>
Yes. In Leadsets, set <code>has_verified_email</code> to <code>true</code>. Each email still shows its status and <code>checked_at</code> date.</p>
<p><strong>How do I avoid paying for empty searches?</strong>
In Leadsets an empty result uses no records, and <code>count_only</code> returns the match count without using records.</p>
<p><strong>Can Claude or ChatGPT run people searches directly?</strong>
Yes, through the Leadsets MCP server at <code>https://mcp.leadsets.ai</code>. See the <a href="https://www.leadsets.ai/docs">docs</a> to connect, or <a href="https://www.leadsets.ai/signup">get a free API key</a> to use REST.</p>
]]></content:encoded></item><item><title><![CDATA[Contact and company enrichment APIs: how to pick one that works]]></title><description><![CDATA[Short answer: the best company or contact enrichment API is the one that fills the fields you need for your own target accounts, tells you how fresh and…]]></description><link>https://leadsets.hashnode.dev/contact-and-company-enrichment-apis-how-to-pick-one-that-works</link><guid isPermaLink="true">https://leadsets.hashnode.dev/contact-and-company-enrichment-apis-how-to-pick-one-that-works</guid><category><![CDATA[api]]></category><category><![CDATA[sales]]></category><category><![CDATA[AI]]></category><category><![CDATA[b2b]]></category><category><![CDATA[SaaS]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Fri, 20 Nov 2026 06:27:41 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> the best company or contact enrichment API is the one that fills the fields you need for your own target accounts, tells you how fresh and how certain each result is, and has pricing you can predict. Test two or three APIs on the same 100 records from your ICP before you commit. If you want enrichment inside an AI agent with flat pricing, Leadsets offers it over a remote MCP server and REST.</p>
<h2>What an enrichment API does</h2>
<p>An enrichment API takes something you already have and returns a fuller record.</p>
<ul>
<li><strong>Company enrichment:</strong> you send a company domain or name and get firmographic data back, such as industry, size and location.</li>
<li><strong>Contact enrichment:</strong> you send a person, for example a name plus company, and get their current role, company and a work email when one can be found.</li>
<li><strong>Search:</strong> you describe the people or companies you want and get a list of matches to enrich next.</li>
<li><strong>Verification:</strong> you send an email address and learn whether it is safe to send to.</li>
</ul>
<p>Most teams need all four in sequence: search, enrich, verify, then push to a CRM or sequencer.</p>
<h2>A checklist for choosing an enrichment API</h2>
<ol>
<li><strong>List the fields you will actually use.</strong> If routing uses company size and industry, and outreach needs a verified work email, those are your must-haves.</li>
<li><strong>Measure match rate on your own data.</strong> Take 100 real records, including small and obscure companies. Count how many come back with your must-have fields filled, not how many return any record at all.</li>
<li><strong>Ask how each result is qualified.</strong> A useful record tells you why it matched, whether the email is verified, catch-all, unknown or invalid, and when the data was last checked. Without that, you cannot tell a fresh record from a stale one.</li>
<li><strong>Model the bill.</strong> Per-record and credit pricing makes cost depend on volume and on how many lookups fail. Flat pricing makes cost fixed but comes with rate or fair-use limits. Price one normal month and one heavy month.</li>
<li><strong>Check the access path.</strong> REST for your own code, an MCP server for AI assistants and agents, or both. If agents will call the API, clear tool names and usage reporting matter.</li>
<li><strong>Read the rate limits.</strong> Know how many calls you can make per window before you design a bulk job.</li>
<li><strong>Plan for gaps.</strong> No provider covers everyone. Decide what your workflow does with a record that comes back empty or catch-all.</li>
</ol>
<h2>How to run a fair test in one afternoon</h2>
<ol>
<li>Pull 100 target accounts and 100 known contacts from your CRM.</li>
<li>Run the same inputs through each shortlisted API.</li>
<li>For companies, score the fields you listed in step one of the checklist.</li>
<li>For contacts, count verified work emails, and treat catch-all and unknown separately from verified.</li>
<li>Divide the monthly price by the number of usable records to get a real cost per usable record.</li>
</ol>
<h2>Where Leadsets fits</h2>
<p>Leadsets is a B2B contact data API and remote MCP server for AI agents: Unlimited B2B Contact Data for AI Agents, Starting from $29/mo (paid plans; fair-use rate limits apply).</p>
<ul>
<li><strong>Eight tools:</strong> <code>search_people</code>, <code>search_companies</code>, <code>enrich_person</code>, <code>enrich_company</code>, <code>verify_email</code>, <code>scrape_jobs</code>, <code>scrape_company</code> and <code>get_usage</code>.</li>
<li><strong>Two ways in:</strong> the remote MCP server at <code>https://mcp.leadsets.ai</code>, and REST at <code>https://api.leadsets.ai/v1</code>.</li>
<li><strong>Qualified records:</strong> each record includes match reasons, a verification status (verified, catch-all, unknown or invalid) and a checked date, which covers point 3 of the checklist.</li>
<li><strong>Flat pricing, no credits:</strong></li>
</ul>
<table>
<thead>
<tr>
<th>Plan</th>
<th>Price (USD)</th>
</tr>
</thead>
<tbody><tr>
<td>Free</td>
<td>1,000 enrichments and 100 search records, no card</td>
</tr>
<tr>
<td>Solo</td>
<td>$29 / month</td>
</tr>
<tr>
<td>Pro</td>
<td>$99 / month</td>
</tr>
<tr>
<td>Team</td>
<td>$399 / month</td>
</tr>
<tr>
<td>Max</td>
<td>$999 / month</td>
</tr>
</tbody></table>
<p>Paid plans have fair-use limits per 5 hours and per week instead of a per-record bill.</p>
<p>A typical agent flow looks like this: <code>search_companies</code> to build an account list, <code>enrich_company</code> to qualify it, <code>search_people</code> to find the right roles, <code>enrich_person</code> for current details, and <code>verify_email</code> before anything is sent.</p>
<h2>Honest limits</h2>
<ul>
<li><strong>No phone numbers.</strong> Leadsets returns work emails, people and company data only.</li>
<li><strong>Fair-use pacing.</strong> Flat pricing comes with limits per 5 hours and per week, so very large one-off backfills take planning.</li>
<li><strong>Not every contact will have a verified email.</strong> Some come back catch-all, unknown or invalid, and Leadsets labels them that way rather than hiding them.</li>
</ul>
<h2>FAQ</h2>
<p><strong>What is the best company enrichment API?</strong>
There is no single winner for every team. Pick the API that fills your must-have fields on your own sample, labels how certain each result is and has a bill you can predict. Leadsets is built for teams that enrich from AI agents or code at a flat price.</p>
<p><strong>What is the difference between contact enrichment and company enrichment?</strong>
Company enrichment adds firmographic data to an account. Contact enrichment adds a person's current role, company and work email.</p>
<p><strong>Is there a free enrichment API?</strong>
Leadsets has a Free plan with 1,000 enrichments and 100 search records, with no card.</p>
<p><strong>Does Leadsets charge per record or use credits?</strong>
No. Paid plans are a flat monthly price from $29, with fair-use limits per 5 hours and per week.</p>
<p><strong>Can an AI agent call Leadsets directly?</strong>
Yes. Connect any MCP client to <code>https://mcp.leadsets.ai</code>, or call the REST API from your own agent code.</p>
<p><strong>Does Leadsets return phone numbers?</strong>
No. Coverage is work emails, people and company data.</p>
<p>Get started: <a href="https://www.leadsets.ai">Get a free API key</a> · <a href="https://www.leadsets.ai/docs">Docs</a> · <a href="https://www.leadsets.ai/pricing">Pricing</a></p>
]]></content:encoded></item><item><title><![CDATA[ZoomInfo alternatives for startups: how to choose, and where an API fits]]></title><description><![CDATA[Short answer: the best ZoomInfo alternative for a startup is the one that matches how you actually prospect, at a price you can see before you sign. If…]]></description><link>https://leadsets.hashnode.dev/zoominfo-alternatives-for-startups-how-to-choose-and-where-an-api-fits</link><guid isPermaLink="true">https://leadsets.hashnode.dev/zoominfo-alternatives-for-startups-how-to-choose-and-where-an-api-fits</guid><category><![CDATA[sales]]></category><category><![CDATA[api]]></category><category><![CDATA[AI]]></category><category><![CDATA[b2b]]></category><category><![CDATA[SaaS]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Fri, 20 Nov 2026 06:25:41 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> the best ZoomInfo alternative for a startup is the one that matches how you actually prospect, at a price you can see before you sign. If your prospecting runs through AI agents or your own code, a B2B contact data API with flat pricing, such as Leadsets, is often a better fit than a seat-based platform. If you need phone numbers, a dialer or intent data, look at full sales intelligence suites instead.</p>
<h2>Why startups look for a ZoomInfo alternative</h2>
<p>ZoomInfo is built for large go-to-market teams. Its own pricing page (<a href="https://www.zoominfo.com/pricing">https://www.zoominfo.com/pricing</a>) does not list prices; it says pricing is based on features and functionality, the number of licenses and credit usage, and that each export of a contact or company profile costs one credit. For a startup, that model raises three practical questions:</p>
<ul>
<li><strong>Can I see the price before a sales call?</strong> Early teams want to compare tools in an afternoon, not after a demo cycle.</li>
<li><strong>Will the bill change with usage?</strong> Credit-based pricing means the cost of a month depends on how many records you export.</li>
<li><strong>Do I need the whole suite?</strong> Many startups only need search, enrichment and verification, and already have a CRM and sequencer.</li>
</ul>
<p>None of this makes ZoomInfo a bad product. It means a startup should pick the smallest tool that covers its real job.</p>
<h2>A checklist for choosing a ZoomInfo alternative</h2>
<ol>
<li><strong>Write down the job.</strong> "Find VPs of Sales at US SaaS companies with 50 to 500 people, with verified work emails" is a job. "Better data" is not.</li>
<li><strong>Decide on the channel.</strong> Email-first outbound needs verified work emails. Cold calling needs phone numbers. Many tools are strong at one and weak at the other.</li>
<li><strong>Decide who will pull the data.</strong> A rep clicking in a browser, a RevOps person running bulk enrichment, or an AI agent calling a tool. The answer decides whether you need seats, a CSV export or an API.</li>
<li><strong>Check the pricing model.</strong> Per seat, per credit or flat monthly. Estimate one normal month and one busy month under each model.</li>
<li><strong>Look at what each record tells you.</strong> You want to know why a record matched your search, whether the email is verified, catch-all, unknown or invalid, and when it was last checked.</li>
<li><strong>Test on your own ICP.</strong> Run the same 100 target accounts through each shortlisted tool. Count usable, verified contacts, not total rows.</li>
<li><strong>Check the exit.</strong> Can you export to your CRM and leave without a long contract?</li>
</ol>
<h2>Three kinds of alternatives</h2>
<p><strong>Sales intelligence suites.</strong> A database, browser extension, intent signals and often phone data. Good for teams with several reps and a seat budget.</p>
<p><strong>Email finders and verifiers.</strong> Narrow tools that find and check work emails for a list you already have.</p>
<p><strong>Contact data APIs and MCP servers.</strong> Data delivered as tools that code or an AI agent can call. Good for founder-led sales and small GTM teams.</p>
<h2>Where Leadsets fits</h2>
<p>Leadsets is in the third group. It is a B2B contact data API and remote MCP server for AI agents: Unlimited B2B Contact Data for AI Agents, Starting from $29/mo (paid plans; fair-use rate limits apply).</p>
<ul>
<li><strong>Eight tools:</strong> <code>search_people</code>, <code>search_companies</code>, <code>enrich_person</code>, <code>enrich_company</code>, <code>verify_email</code>, <code>scrape_jobs</code>, <code>scrape_company</code> and <code>get_usage</code>.</li>
<li><strong>Two ways in:</strong> the remote MCP server at <code>https://mcp.leadsets.ai</code> for any MCP client, and REST at <code>https://api.leadsets.ai/v1</code> for your own code.</li>
<li><strong>Records you can trust at a glance:</strong> each record includes match reasons, a verification status (verified, catch-all, unknown or invalid) and a checked date.</li>
<li><strong>Flat pricing, no credits:</strong></li>
</ul>
<table>
<thead>
<tr>
<th>Plan</th>
<th>Price (USD)</th>
</tr>
</thead>
<tbody><tr>
<td>Free</td>
<td>1,000 enrichments and 100 search records, no card</td>
</tr>
<tr>
<td>Solo</td>
<td>$29 / month</td>
</tr>
<tr>
<td>Pro</td>
<td>$99 / month</td>
</tr>
<tr>
<td>Team</td>
<td>$399 / month</td>
</tr>
<tr>
<td>Max</td>
<td>$999 / month</td>
</tr>
</tbody></table>
<p>Paid plans have fair-use limits per 5 hours and per week, so heavy use is paced rather than billed per record.</p>
<h2>Honest limits</h2>
<ul>
<li><strong>No phone numbers.</strong> Leadsets covers work emails, people and company data. If your outbound depends on calling, you need another source.</li>
<li><strong>Not a sales suite.</strong> There is no sequencer, dialer or intent dashboard. Leadsets is the data layer; send the results to your CRM or sequencer.</li>
<li><strong>Fair-use limits apply.</strong> Pacing per 5 hours and per week matters for very large one-off pulls.</li>
</ul>
<h2>How to try it in an hour</h2>
<ol>
<li>Create a free account at leadsets.ai and get an API key, no card needed.</li>
<li>Connect <code>https://mcp.leadsets.ai</code> in your AI assistant, or call the REST API.</li>
<li>Ask for one real target list, for example "Heads of Growth at US B2B SaaS companies with 20 to 200 employees".</li>
<li>Enrich and verify the results, then compare them with your current source on the same accounts.</li>
</ol>
<h2>FAQ</h2>
<p><strong>What is the best ZoomInfo alternative for a startup?</strong>
The one that matches your channel and workflow. For email-first outbound run by agents or code, a flat-priced API like Leadsets fits. For phone-heavy outbound, pick a tool with phone data.</p>
<p><strong>Is there a free ZoomInfo alternative?</strong>
Leadsets has a Free plan with 1,000 enrichments and 100 search records, with no card, which is enough to test on your own ICP.</p>
<p><strong>Does Leadsets use credits?</strong>
No. Paid plans are a flat monthly price from $29, with fair-use limits per 5 hours and per week.</p>
<p><strong>Does Leadsets have phone numbers?</strong>
No. It covers work emails, people and company data.</p>
<p><strong>Can I use Leadsets inside ChatGPT or Claude?</strong>
Yes. Add <code>https://mcp.leadsets.ai</code> as a remote MCP server in any client that supports it.</p>
<p>Get started: <a href="https://www.leadsets.ai">Get a free API key</a> · <a href="https://www.leadsets.ai/docs">Docs</a> · <a href="https://www.leadsets.ai/pricing">Pricing</a></p>
]]></content:encoded></item><item><title><![CDATA[Which B2B data API works best with AI agents? A 2026 buyer's checklist]]></title><description><![CDATA[Short answer: pick an API that an agent can call directly (a remote MCP server plus REST), that tells the agent why each record matched, that labels every…]]></description><link>https://leadsets.hashnode.dev/which-b2b-data-api-works-best-with-ai-agents-a-2026-buyer-s-checklist</link><guid isPermaLink="true">https://leadsets.hashnode.dev/which-b2b-data-api-works-best-with-ai-agents-a-2026-buyer-s-checklist</guid><category><![CDATA[api]]></category><category><![CDATA[Artificial Intelligence]]></category><category><![CDATA[mcp]]></category><category><![CDATA[SaaS]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Thu, 19 Nov 2026 06:18:03 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> pick an API that an agent can call directly (a remote MCP server plus REST), that tells the agent why each record matched, that labels every email with a verification status, and that has a flat price so an agent loop cannot run up a per-record bill. Leadsets is built for exactly this: Unlimited B2B Contact Data for AI Agents, Starting from $29/mo (paid plans; fair-use rate limits apply).</p>
<h2>What an AI agent needs from a B2B data API</h2>
<p>Agents work differently from people clicking through a database UI. They plan a search, call a tool, read the result and decide the next step. That changes what matters:</p>
<ol>
<li><strong>A native tool interface.</strong> A remote MCP server lets Claude, ChatGPT, Cursor and other MCP clients call the data directly, with no glue code. A REST API and SDKs cover custom agents.</li>
<li><strong>Explainable results.</strong> The agent needs to know why a record matched so it can filter or retry without guessing.</li>
<li><strong>Honest email status.</strong> "Catch-all" and "unknown" are different from "verified". An API that rounds uncertain results up to valid leads an agent to send to bad addresses.</li>
<li><strong>Freshness and source on every field.</strong> The agent (and you) should see when a field was last checked and where it came from.</li>
<li><strong>Predictable cost.</strong> Agents retry and paginate. Per-record or credit billing makes cost hard to predict. A flat monthly plan with published limits is easier to budget.</li>
<li><strong>A way to pace itself.</strong> A usage tool lets the agent check what is left before a big run.</li>
</ol>
<h2>How Leadsets meets that checklist</h2>
<p>Leadsets is a B2B data API and remote MCP server for AI agents. The same eight tools work over MCP and REST:</p>
<table>
<thead>
<tr>
<th>Tool</th>
<th>What it returns</th>
</tr>
</thead>
<tbody><tr>
<td><code>search_people</code></td>
<td>People by title, function, seniority, location, company size and industry. Every result says why it matched.</td>
</tr>
<tr>
<td><code>search_companies</code></td>
<td>Accounts by industry, headcount band, technology and location.</td>
</tr>
<tr>
<td><code>enrich_person</code></td>
<td>Current role, company, and a verified work email when one exists.</td>
</tr>
<tr>
<td><code>enrich_company</code></td>
<td>Firmographics for a domain, with source and freshness on every field.</td>
</tr>
<tr>
<td><code>verify_email</code></td>
<td>Verified, catch-all, unknown or invalid, with a confidence score. Uncertain results are not rounded up to valid.</td>
</tr>
<tr>
<td><code>scrape_company</code></td>
<td>A company homepage turned into products, positioning, named customers and key links.</td>
</tr>
<tr>
<td><code>scrape_jobs</code></td>
<td>Open roles from careers pages, with titles and dates, for hiring signals.</td>
</tr>
<tr>
<td><code>get_usage</code></td>
<td>Remaining allowance in the current window and week. Does not consume records.</td>
</tr>
</tbody></table>
<p><strong>Connect in two minutes.</strong> Add <code>https://mcp.leadsets.ai</code> as a custom connector in Claude, Claude Code, Cursor, ChatGPT or any MCP client and sign in with OAuth. Nothing to install. For your own code, call <code>https://api.leadsets.ai/v1</code> with a bearer key, or use the TypeScript (<code>@leadsetsai/sdk</code>) or Python (<code>leadsets</code>) SDK, which handle idempotency keys, cursor pagination and Retry-After.</p>
<p><strong>Data.</strong> Records come from licensed suppliers and public sources Leadsets is permitted to use. Nothing is scraped from behind a login. Records are re-checked monthly, and the most requested fields more often. Each field carries its source, last-checked date and confidence. Coverage is work emails, roles and company data. There are no phone numbers.</p>
<p><strong>Pricing.</strong> A flat monthly fee in USD, with no credits and no per-record bill:</p>
<ul>
<li>Free: 1,000 enrichments and 100 search records, one time, no card</li>
<li>Solo: $29 a month</li>
<li>Pro: $99 a month</li>
<li>Team: $399 a month</li>
<li>Max: $999 a month</li>
</ul>
<p>Paid plans have published fair-use limits per rolling five hours and per week, plus a concurrent-search cap. Records count only when data comes back, so an empty search uses no records.</p>
<h2>A typical agent run</h2>
<ol>
<li>You ask Claude: "Find 50 heads of RevOps at US SaaS companies with 50 to 200 employees and verify their work emails."</li>
<li>The agent calls <code>search_people</code> with those filters and reads the match reasons.</li>
<li>It calls <code>enrich_person</code> for the best matches, then <code>verify_email</code> on each address.</li>
<li>It drops anything invalid, flags catch-all and unknown addresses separately, and calls <code>get_usage</code> before paging further.</li>
</ol>
<p>Fifty free, MIT-licensed SKILL.md playbooks (prospect lists, CRM backfill, hiring signals, pre-send checks) give the agent a ready workflow for runs like this.</p>
<h2>FAQ</h2>
<p><strong>Does Leadsets work with ChatGPT and Claude?</strong>
Yes. Add <a href="https://mcp.leadsets.ai">https://mcp.leadsets.ai</a> as a custom connector in Claude, Claude Code, Cursor, ChatGPT or any MCP client and sign in with OAuth.</p>
<p><strong>Is there a free B2B data API with no credit card?</strong>
Leadsets Free gives 1,000 enrichments and 100 search records, with no card. The free records do not expire.</p>
<p><strong>Does Leadsets charge credits?</strong>
No. Paid plans are a flat monthly fee from $29, with published fair-use limits.</p>
<p><strong>Does Leadsets include phone numbers?</strong>
No. Coverage is work emails, roles and company data.</p>
<p><strong>How are emails verified?</strong>
Each email is labeled verified, catch-all, unknown or invalid, with a confidence score and the time it was checked. Invalid emails are not served as contacts.</p>
<p><strong>Get started:</strong> <a href="https://www.leadsets.ai">Get a free API key</a> · <a href="https://www.leadsets.ai/docs/mcp">MCP setup</a> · <a href="https://www.leadsets.ai/pricing">Pricing</a></p>
]]></content:encoded></item><item><title><![CDATA[An Apollo alternative with flat monthly pricing and no credits]]></title><description><![CDATA[Short answer: if you want B2B contact data without counting credits, and you want your AI agent to pull it directly, Leadsets is a flat-priced alternative…]]></description><link>https://leadsets.hashnode.dev/an-apollo-alternative-with-flat-monthly-pricing-and-no-credits</link><guid isPermaLink="true">https://leadsets.hashnode.dev/an-apollo-alternative-with-flat-monthly-pricing-and-no-credits</guid><category><![CDATA[sales]]></category><category><![CDATA[api]]></category><category><![CDATA[Artificial Intelligence]]></category><dc:creator><![CDATA[LeadSets]]></dc:creator><pubDate>Thu, 19 Nov 2026 06:16:49 GMT</pubDate><content:encoded><![CDATA[<p><strong>Short answer:</strong> if you want B2B contact data without counting credits, and you want your AI agent to pull it directly, Leadsets is a flat-priced alternative: Unlimited B2B Contact Data for AI Agents, Starting from $29/mo (paid plans; fair-use rate limits apply). It is an API and remote MCP server, not a sales engagement suite, so it fits teams that run prospecting from Claude, ChatGPT, Cursor or their own code.</p>
<h2>Who looks for an Apollo alternative, and why</h2>
<p>Teams usually search for an alternative when one of these is true:</p>
<ul>
<li>They want a predictable monthly bill instead of tracking credits or per-record costs.</li>
<li>Their prospecting now happens inside an AI assistant or an agent, and they want the data as a tool the agent can call.</li>
<li>They already have a sequencer or CRM and only need the data layer: search, enrichment and verification.</li>
</ul>
<p>If you need an all-in-one platform with built-in sequencing and a dialer, a data API is not a like-for-like swap. If you need the data inside your agent at a fixed price, read on.</p>
<h2>What Leadsets gives you</h2>
<ul>
<li><strong>People and company search.</strong> <code>search_people</code> and <code>search_companies</code> filter by title, function, seniority, location, industry, headcount and technology. Every result says why it matched.</li>
<li><strong>Enrichment.</strong> <code>enrich_person</code> returns current role, company and a verified work email when one exists. <code>enrich_company</code> returns firmographics with source and freshness on every field.</li>
<li><strong>Email verification.</strong> <code>verify_email</code> labels each mailbox verified, catch-all, unknown or invalid, with a confidence score.</li>
<li><strong>Signals.</strong> <code>scrape_jobs</code> reads open roles from careers pages; <code>scrape_company</code> turns a homepage into products, positioning and named customers.</li>
<li><strong>Pacing.</strong> <code>get_usage</code> shows what is left in the current window and week, without using records.</li>
</ul>
<p>All eight tools work over MCP (<code>https://mcp.leadsets.ai</code>, OAuth sign-in, nothing to install) and REST (<code>https://api.leadsets.ai/v1</code>), with TypeScript and Python SDKs.</p>
<h2>Pricing: flat, no credits</h2>
<table>
<thead>
<tr>
<th>Plan</th>
<th>Price (USD)</th>
</tr>
</thead>
<tbody><tr>
<td>Free</td>
<td>1,000 enrichments and 100 search records, one time, no card</td>
</tr>
<tr>
<td>Solo</td>
<td>$29 / month</td>
</tr>
<tr>
<td>Pro</td>
<td>$99 / month</td>
</tr>
<tr>
<td>Team</td>
<td>$399 / month</td>
</tr>
<tr>
<td>Max</td>
<td>$999 / month</td>
</tr>
</tbody></table>
<p>There are no credits and no per-record bill. Paid plans have published fair-use limits per rolling five hours and per week, plus a concurrent-search cap. Records count only when data comes back; an empty search uses no records. Scrape unit packs are an optional add-on on paid plans.</p>
<h2>How to switch a workflow in an afternoon</h2>
<ol>
<li>Create a free account at leadsets.ai (no card) and connect <code>https://mcp.leadsets.ai</code> in Claude or ChatGPT.</li>
<li>Ask the assistant to rebuild one saved search, for example "VPs of Sales at US fintech companies with 200 to 1,000 employees".</li>
<li>Enrich and verify the results, then export to your CRM or sequencer.</li>
<li>Compare the list on your own ICP before you pick a paid plan.</li>
</ol>
<h2>What to check before you switch</h2>
<ul>
<li><strong>Coverage you need.</strong> Leadsets covers work emails, roles and company data. It does not provide phone numbers.</li>
<li><strong>Data handling.</strong> Records come from licensed suppliers and permitted public sources, nothing is scraped from behind a login, and each field carries its source and last-checked date. Anyone can opt out at leadsets.ai/opt-out.</li>
<li><strong>Your workflow.</strong> If your team lives in an AI assistant or builds agents, MCP access is the main gain.</li>
</ul>
<h2>FAQ</h2>
<p><strong>Is Leadsets a cheaper alternative to Apollo?</strong>
Leadsets paid plans are a flat monthly fee from $29, with no credits. Compare the plan limits with your own monthly volume to see which costs less for you.</p>
<p><strong>Can I try it without a card?</strong>
Yes. The Free plan gives 1,000 enrichments and 100 search records, with no card.</p>
<p><strong>Does it work inside ChatGPT or Claude?</strong>
Yes. Add <a href="https://mcp.leadsets.ai">https://mcp.leadsets.ai</a> as a custom connector and sign in with OAuth.</p>
<p><strong>Does Leadsets do email sequencing?</strong>
No. Leadsets is the data layer. Send the verified list to your sequencer or CRM.</p>
<p><strong>Does Leadsets have phone numbers?</strong>
No. Coverage is work emails, roles and company data.</p>
<p><strong>Get started:</strong> <a href="https://www.leadsets.ai">Get a free API key</a> · <a href="https://www.leadsets.ai/pricing">Pricing</a> · <a href="https://www.leadsets.ai/docs/mcp">Connect via MCP</a></p>
]]></content:encoded></item></channel></rss>