Real-Time API — Low Latency Verification

Median (p50) Response

48 ms — From edge nodes within the same continent as the target mail server. Typical for US-to-US or EU-to-EU validations.

95th Percentile (p95)

142 ms — Captures cross-continental lookups and mail servers with slower SMTP responses. Still well within our sub-200ms commitment.

99th Percentile (p99)

198 ms — Worst-case scenarios including retry logic, DNS cache misses, and high-concurrency bursts during peak hours.

Throughput Capacity

1,200 req/s per API key at default rate limits. Enterprise plans scale to 8,000 req/s with dedicated edge pools and no queueing.

Error Rate

0.03% — Measured over 47 million requests in the past month. The vast majority are upstream DNS timeouts, not VeriMail infrastructure failures.

Cache Hit Ratio

61.4% — Domain-level results are cached for 4 hours by default. Frequently validated domains like gmail.com and outlook.com respond in under 12 ms.

Python (requests)

import requests

api_key = "YOUR_API_KEY"
endpoint = "https://api.verimail.io/v1/verify"

payload = {
    "email": "sarah.chen@example.com",
    "checks": ["syntax", "mx", "smtp", "disposable", "role"]
}

response = requests.post(
    endpoint,
    json=payload,
    headers={"Authorization": f"Bearer {api_key}"},
    timeout=2
)

result = response.json()
print(f"Status: {result['status']}")
print(f"Confidence: {result['confidence_score']}%")
print(f"Latency: {result['meta']['response_time_ms']} ms")

Node.js (fetch)

const apiKey = "YOUR_API_KEY";
const endpoint = "https://api.verimail.io/v1/verify";

async function verifyEmail(email) {
  const res = await fetch(endpoint, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "Authorization": `Bearer ${apiKey}`
    },
    body: JSON.stringify({
      email,
      checks: ["syntax", "mx", "smtp", "disposable", "role"]
    })
  });

  const data = await res.json();
  console.log(`Status: ${data.status}`);
  console.log(`Confidence: ${data.confidence_score}%`);
  console.log(`Latency: ${data.meta.response_time_ms} ms`);
  return data;
}

verifyEmail("jordan.patel@acme-corp.io");

cURL

curl -X POST https://api.verimail.io/v1/verify \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "email": "admin@shopify-store.net",
    "checks": ["syntax", "mx", "smtp", "disposable", "role", "typo"]
  }'

# Example response (142 ms):
# {
#   "email": "admin@shopify-store.net",
#   "status": "valid",
#   "confidence_score": 94,
#   "checks": {
#     "syntax": "pass",
#     "mx": "pass",
#     "smtp": "pass",
#     "disposable": false,
#     "role": true,
#     "typo": null
#   },
#   "meta": {
#     "response_time_ms": 142,
#     "cache_hit": false,
#     "server": "edge-eu-west-1"
#   }
# }

Layer 1 — Domain Result Cache

When we validate mike@fastmail.com, the full result — MX records, SMTP banner, disposable status — is cached for the entire fastmail.com domain for 4 hours. The next email to that domain skips DNS and SMTP entirely, returning in under 15 ms. This layer alone accounts for 61.4% of all cache hits.

Layer 2 — MX Record Cache

MX lookups are cached independently for 24 hours across all edge nodes via our Redis cluster. Even if a domain's SMTP status changes, we already know where its mail servers live, shaving 40–80 ms off the first SMTP probe. Invalid MX records are cached for only 30 minutes to handle rapid infrastructure changes.

Layer 3 — Edge Syntax & Pattern Cache

Syntax validation, disposable-domain matching, and role-address detection happen on the edge server itself. These checks load from a pre-warmed in-memory hash map that updates every 6 hours from our master list of 14,200+ known disposable providers and 3,800 role-based patterns. Zero network calls required.

Smart Retry Logic

If an SMTP handshake times out after 120 ms, VeriMail retries once against a secondary MX server. If both fail within 180 ms total, the result is marked unknown with a confidence score, rather than blocking your request. You get a response every time — never a timeout error.

Batch Endpoint Optimization

The /v1/verify/batch endpoint accepts up to 100 emails per request and processes them in parallel across available edge capacity. A batch of 100 emails typically completes in 180–350 ms total — not per-email — because domain-level checks are deduplicated automatically. Ideal for CRM imports and list hygiene jobs.

Cache Invalidation Controls

Enterprise customers can trigger manual cache purges via the dashboard or the POST /v1/cache/purge endpoint. When a domain changes its mail infrastructure — for example, migrating from Google Workspace to Microsoft 365 — you can force a fresh verification on the next request. Purge propagation takes under 2 seconds across all edge nodes.