Skip to content

// use case · Trust and safety engineers, risk analysts and product teams at marketplaces, communities and two-sided platforms.

Screen new marketplace accounts before they list, buy or message

Scammers and spammers on marketplaces cycle through throwaway accounts, and the email is the first thing you see. Look up each new account, hold back listing or messaging rights when verdict.is_spam is true, and store the signals so your trust model can learn from them. Seller emails on company domains can also get a company profile for business verification.

The problem

  • Scam sellers come back the day after we ban them, with a new throwaway address.
  • New accounts start messaging buyers with phishing links within minutes.
  • We want to verify business sellers without asking for documents from everyone.
  • Our moderators spend their day on accounts that were obviously fake at signup.

How pingutil solves it

  1. 1

    Check at signup

    Call /v1/lookup with skip_enrichment: true for every new account and store verdict and email_signals with the user.

  2. 2

    Gate risky capabilities

    Limit listing, messaging or payouts for is_spam accounts until they verify another way. Let low_trust accounts in with tighter limits.

  3. 3

    Profile business sellers

    For sellers with a company domain, a full lookup returns company.name and company.website to compare against what they told you.

  4. 4

    Feed your model

    Use randomness_score and the boolean signals as features alongside your behavioural data.

Fields you will use

Response fieldWhat to do with it
verdict.is_spamHold back listing and messaging rights
verdict.low_trustApply tighter limits to new accounts
email_signals.randomness_scoreUse as a model feature
company.nameCross-check a business seller’s claimed name
company.websiteCompare against the seller’s listed site
curl -s -X POST https://pingutil.com/v1/lookup \
  -H "Authorization: Bearer $PINGUTIL_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "input": "seller@example-shop.com", "skip_enrichment": true }' \
  | jq '{category, verdict, email_signals}'

Compared with the alternatives

Trust and safety teams typically combine identity verification vendors, device and IP intelligence, and in-house behavioural models. ID checks are strong but add friction and cost per user, so they are usually reserved for sellers or payouts. Email signals are cheap enough to run on every account and help decide who needs the heavier checks.

See provider comparisons →

Limits to plan for

  • A scammer with a real webmail account passes every email check; behavioural signals still matter.
  • Company profiles describe the domain, not the seller. A person can use a real company domain without working there.
  • Privacy relay addresses can be flagged as randomized; allow known relay domains if your users rely on them.

Common questions

Should I ban accounts with low_trust?
No. low_trust covers role and sales-style addresses that are often legitimate. Use it to apply tighter default limits, not to block.
Can it verify that a seller works at a company?
Not directly. It confirms the domain belongs to a real company. Pair it with an email confirmation link sent to that address to show they control it.
Does it check IP address or device?
No. pingutil only looks at the email address and its domain.

Read next