> For the complete documentation index, see [llms.txt](https://support.iafcertsearch.org/verifiers/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://support.iafcertsearch.org/verifiers/release-notes/release-announcements/release-notes-august-10-2026.md).

# \[Release Notes] August 10, 2026

#### **Smarter Company Matching**

Company name matching now recognizes more near-matches when you search, match, or verify a company.

**What's New:**

* **Recognizes more near-matches.** Extra words, like a country, city, or trading name, no longer block a match. Different legal forms, like "Ltd" vs. "Limited", are recognized as the same company automatically. Close matches are labeled **"loosely matched"** with a confidence score.
* **New request option: `enhanced_fuzzy_matching`.** `enhanced_fuzzy_matching` is an optional boolean. The default is **false**.
  * **Results:** when **true**, matches are graded by how closely the name really matches, so results land in the correct band. Weak matches that are not the same company drop to the Very Low band (not matched) instead of coming back as loose matches. Turning it off does not stop loose matching; it keeps the current scoring.
  * **Performance:** this setting trades a little speed for more precise scoring. Exact matches are unaffected, but every name that is not an exact match goes through an additional pass that scores how closely it matches the Certified Entity. The impact depends on your request:
    * **Single requests:** barely noticeable, since only one name is scored.
    * **Bulk requests:** slower, and the more names in the batch that are not exact matches, the longer it takes, because each one is scored individually.

To know more about this feature, go to: [Verification Users API Developer Guide](https://support.iafcertsearch.org/verification-users-api-developer-guide/api-integration/verification-apis)

## Frequently asked questions

<details>

<summary>Will company name matching results change for me?</summary>

Only if you turn it on. Send `enhanced_fuzzy_matching: true` at the root of your request payload to get the improved matching. Leave it out, or set it to `false`, and matching behaves exactly as before.

</details>

<details>

<summary>What kind of name differences does this handle better?</summary>

Extra words, like a country, city, or trading name. Legal forms, like "Ltd" vs "Limited", are handled automatically, regardless of this setting. Non-exact results are labeled "loosely matched" with a confidence score.

</details>

<details>

<summary>Will this slow down my requests?</summary>

Exact matches are not affected. For names that are not exact matches, each one goes through an extra scoring pass. For a single request this is barely noticeable. For bulk requests, the more non-exact-match names in the batch, the longer it takes, since each is scored individually.

</details>