April 2023 Release Notes (v22.0)
| Release Name | Dataset Version | Publish Date |
|---|---|---|
| April 2023 | v22.0 | 04/04/2023 |
This data version was was released on 4/4/2023.
Welcome to our April 2023 release notes! We’re rolling out some exciting updates with this release.
Here are some of the key highlights:
- Keep your data up to 3x fresher with PDL’s new monthly data license delivery service.
- Automatically keep records up-to-date with our Person Retrieve Webhook, now available for our Person Data API and Data License customers.
- We’re making it easier than ever to contact leads with our
recommended_personal_emailfield, now available in our Base Person Bundle for API customers and will be available for our Data License customers after May 2023. - Huge increases in parent & subsidiary company information.
- Added 4.8 million profiles to the company dataset (23% increase).
Excited yet? Read on to learn more, or jump to a specific section using the table of contents below.
- Recommended Personal Email in Base Person Bundle
- Parent & Subsidiary Company Tagging Improvements
- Person ID Length Upcoming Changes
- Monthly Data License Delivery Options
- Retrieve Webhook for Data License Customers
- Person Enrichment & Preview API Support for PDL ID Input
We launched the recommended_personal_email field to our Person Dataset in v20. This field is the statistically most likely deliverable personal email for a profile. When we ran deliverability tests on this field, we saw a 37% improvement in deliverability.
Previously, this field had an additional cost on top of our base packages. We are pleased to announce that the recommended_personal_email field is now part of the Base Person Bundle for API customers and will be available for our Data License customers after May 2023! There are no changes made to the pricing of the Base Person Bundle, and we will continue to list additional emails in the emails field.
The recommended email is chosen using a proprietary process that probabilistically correlates to higher deliverability, but is not guaranteed to be deliverable. We are continuing to improve our overall deliverability of this field and our other email fields by investing in our data sourcing efforts.
In this release, we’ve increased our company parent & subsidiary coverage to expand the connections between companies that already exist as profiles in our company data. This change will increase the accuracy and coverage of the parent/subsidiary fields within our company records, which makes it more reliable for users to understand key relationships between matched profiles and easier for users to roll up the headcount or employee movement summaries of companies operating within larger corporate conglomerates.
This change affects the fill rates in the all_subsidiaries, direct_subsidiaries, immediate_parent, and ultimate_parent fields.
Here’s a quick overview of our increased linkages:
| Old # | New # | Increase % | |
|---|---|---|---|
| Companies with 1+ Parent | 32,193 | 81,353 | - 152.7% |
| Companies with 1+ Child | 9,121 | 29,362 | - 221.9% |
Upcoming Person ID Maximum Length Change in October 2023
We will be adjusting the maximum character length of PDL Persistent Person IDs. We will be launching this change in October 2023 to allow for ample time to adjust any of your processes impacted by this change.
Currently, IDs are 27 characters long. We recommend allowing for a 64 character maximum for our IDs and will not have any IDs beyond this length, but in practice we expect the changes to result in IDs closer to 32 characters in length. Existing IDs will not change or change length. Only newly created IDs will have the new length.
Currently, PDL delivers Data Licenses on a quarterly cadence. We've had a number of customers request a more frequent Data License delivery cadence and we hear you! With this release, we are implementing monthly delivery options for Data License customers.
Monthly deliveries will be directly tied to our monthly data build timelines, meaning we will be providing the freshest data we can to our customers.
Monthly delivery options will be available at an additional cost for all of our global datasets, including Person and Company data! If you’d like to upgrade your Data License delivery, please reach out to us.
Last quarter (V21), we released the webhook functionality for the Person Retrieve API. As a quick refresher, the webhook functionality enables customers to receive updated profiles automatically instead of periodically having to make hundreds of thousands of Retrieve API calls to keep their records up-to-date.
With this release, we are expanding the webhook functionality to Data License customers who want a delta of IDs that have changed on a monthly basis.
Functionally, the only difference in the webhook between API and Data License customers is which IDs will be pulled. For API customers, the webhook checks for updates to all person profiles accessed by calling the Person Enrichment, Person Search, and Person Identify APIs. For Data License customers, the webhook will check for updates to all person profiles included in the customer’s data pull query.
If you’d like access, please reach out to us.
The Person Enrichment API will now support PDL Persistent Person ID as an input parameter. Additionally, customers who use our Search or Enrichment Preview APIs can now choose to enrich their new records on ID as opposed to currently supported PII parameters.
If you enrich on ID and anything else, only ID is used and the other inputs for matching are ignored.
pdl_id
pdl_id| Type | Description | Example |
|---|---|---|
String | The PDL Persistent ID for a record in our Person Dataset. | qEnOZ5Oh0poWnQ1luFBfVw_0000 |
This quarter, we updated millions of jobs and locations in our Global Resume Dataset. See below for details:
| Dataset | Geography | Field | Records Updated |
|---|---|---|---|
| Resume | Global | experience | 149,059,103 |
| Resume | Global | location | 96,426,401 |
| Resume | United States | experience | 49,852,494 |
| Resume | United States | location | 40,190,936 |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
personal_emails | 683,676,725 | 708,085,315 | 3.57% |
street_addresses | 784,304,951 | 803,791,775 | 2.48% |
phones | 1,140,318,222 | 1,150,823,707 | 0.92% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
total_records | 732,702,622 | 741,406,022 | 1.19% |
phones | 44,881,015 | 54,048,415 | 20.43% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
phones | 292,736,432 | 310,472,440 | 6.06% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
work_email | 21,710,227 | 28,850,007 | 32.89% |
personal_emails | 222,133,824 | 234,147,898 | 5.41% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
work_email | 3,341,191 | 3,810,241 | 14.04% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
work_email | 241,862 | 288,563 | 19.31% |
personal_emails | 338,597 | 374,996 | 10.75% |
| Linkage | Coverage in v21 | Coverage in v22 | Increase (%) |
|---|---|---|---|
total_records | 22,876,457 | 28,212,381 | 23.33% |
all_subsidiaries | 9,121 | 29,362 | 221.92% |
direct_subsidiaries | 9,121 | 29,362 | 221.92% |
immediate_parent | 32,193 | 81,353 | 152.70% |
ultimate_parent | 32,193 | 81,353 | 152.70% |
- Significant increases in contact information linkages across all person datasets.
work_emailandpersonal_emailslinkages increased across datasets, most notablywork_emailincreased by 33% in the phone dataset.phoneslinkages in the resume dataset increased by 20%.
- Added 9.7 million profiles to the resume dataset.
- Added 4.8 million profiles to the company dataset (23% increase).
- Added Company HQ by country stats to our build process & docs.
- Previously, if a company did not have a value in an insights field, that field would be missing from the response. Now, all fields will populate with any blank values shown as
None.
- Resolved instances where end date is before start date in the experience object.
- Continued reduction of LinkedIn URL duplication across profiles by improving our resolving logic for nicknames (Richard → Rick).
- Fixed non-canonical values appearing in the
sizefield for certain companies in our company dataset. - Resolved instances where certain signifiers would incorrectly be parsed as first or last names. For example, fixed "CSPO" appearing as a last name.