DCT
1:25-cv-00324
Zoominfo Tech LLC v. ZENLEADS Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: ZoomInfo Technologies LLC (Delaware)
- Defendant: Zenleads Inc. (d/b/a Apollo.io) (Delaware)
- Plaintiff’s Counsel: Duane Morris LLP
- Case Identification: 1:25-cv-00324, D. Del., 05/21/2025
- Venue Allegations: Venue is alleged to be proper in the District of Delaware because the Defendant is a Delaware corporation and therefore resides in the district.
- Core Dispute: Plaintiff alleges that Defendant’s go-to-market and sales intelligence platform infringes patents related to AI-driven, predictive web crawling and analysis for generating sales leads.
- Technical Context: The lawsuit concerns the field of AI-powered business-to-business (B2B) intelligence platforms, which automate the process of identifying and prioritizing potential customers.
- Key Procedural History: The complaint notes that during the prosecution of the parent patent, the U.S. Patent Examiner found the claims patent-eligible because they recite an "inventive concept" that improves the technical field of web crawling by performing the crawl based on a prior categorization step, which goes beyond routine and conventional data gathering. The ’964 Patent is a continuation of the ’609 Patent.
Case Timeline
| Date | Event |
|---|---|
| 2015-02-10 | Priority Date for '609 and '964 Patents |
| 2015-01-01 | Apollo Enters Business Intelligence Market (approx.) |
| 2019-08-13 | '609 Patent Issued |
| 2022-07-19 | '964 Patent Issued |
| 2025-05-21 | Complaint Filed |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 10,380,609, "Web Crawling For Use In Providing Leads Generation And Engagement Recommendations," issued August 13, 2019
The Invention Explained
- Problem Addressed: The patent describes the challenge of efficiently filtering a large pool of potential sales leads to identify promising customers (’609 Patent, col. 1:26-38). It notes that traditional web crawlers, which recursively follow hyperlinks and save pages, are not optimized for this specific data acquisition and analysis task (Compl. ¶36; ’609 Patent, col. 6:21-25).
- The Patented Solution: The invention is an automated system that intelligently crawls websites to generate and score sales leads. Unlike a traditional crawler, this system first categorizes web pages into types (e.g., product pages, contact pages) using a trained classifier. The results of this categorization are then used to guide the subsequent crawling process itself, making the data collection more targeted and efficient (’609 Patent, Fig. 2; col. 6:45-54). The system then generates a "feature matrix" to compare prospective clients with a company's existing successful clients based on "fitness, engagement, and intent characteristics" to recommend new leads (’609 Patent, Abstract).
- Technical Importance: This approach claims to provide a specific technological improvement to web crawling itself by creating a feedback loop, which is alleged to reduce the time and storage required for a crawl while increasing the precision of the data collected (Compl. ¶36; ’609 Patent, col. 6:49-58).
Key Claims at a Glance
- The complaint asserts independent Claim 1 (Compl. ¶41).
- The essential elements of Claim 1 include:
- Determining similarities between the "fitness, engagement, and intent characteristics" of target clients and an entity's existing clients.
- Generating recommendations for engagement based on these determined similarities.
- Categorizing web pages located on the Internet using a trained classifier that analyzes content and code.
- Crawling the web pages based on the categorization to collect unstructured text information.
- Generating a feature matrix for the target client and comparing it to a feature matrix for existing clients to generate the recommendations.
- The complaint does not explicitly reserve the right to assert dependent claims.
U.S. Patent No. 11,392,964, "Predictive Analytics For Leads Generation And Engagement Recommendations," issued July 19, 2022
The Invention Explained
- Problem Addressed: As a continuation of the '609 Patent, the '964 Patent addresses the same problem of improving the efficiency and effectiveness of automated sales lead generation (’964 Patent, col. 1:29-40).
- The Patented Solution: The '964 Patent describes the same technical solution as its parent patent: an intelligent system that uses page categorization to guide a web crawl, and then compares prospects to existing clients to generate lead recommendations (’964 Patent, col. 2:41-51, Fig. 2).
- Technical Importance: The technology claims to advance the art by creating a more intelligent and purpose-built web crawler for the specific technical problem of B2B lead generation, rather than using generic data gathering techniques (Compl. ¶¶ 37-38).
Key Claims at a Glance
- The complaint asserts independent Claim 1 (Compl. ¶54).
- The essential elements of Claim 1 include:
- Generating recommendations for engagement based on determined similarities between target and existing clients' "fitness, engagement, and intent characteristics."
- Categorizing web pages located on the Internet using a trained classifier.
- Crawling the web pages based on the categorization to collect unstructured text.
- Generating a feature matrix for the target client and comparing it to one for existing clients to generate the recommendations.
- The complaint does not explicitly reserve the right to assert dependent claims.
III. The Accused Instrumentality
- Product Identification: The "Apollo Accused System," which includes the “Apollo Platform,” “Apollo AI,” “Apollo Sales Automation,” and “Apollo Labs” (Compl. ¶29).
- Functionality and Market Context: The complaint describes the Accused System as a go-to-market and sales engagement platform that uses proprietary algorithms to crawl the web, parse websites, and build a "living data network" of company and contact information (Compl. ¶¶ 29, 45). The platform allows users to filter for prospective clients using criteria such as company industry, employee numbers, and "buying intent" to find companies matching a user's "Ideal Customer Profile" (Compl. ¶43). It then provides an "AI auto-generated scoring model" and a "Recommendations" tab to surface "high-value prospects" (Compl. ¶¶ 44, 47). The complaint alleges Apollo markets itself as a direct "ZoomInfo replacement" (Compl. ¶30).
IV. Analysis of Infringement Allegations
'609 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| [a] determining, using a computer, similarities between (a) fitness, engagement, and intent characteristics of a plurality of target clients and (b) fitness, engagement, and intent characteristics of an entity's existing clients; | The Apollo system allows users to filter by parameters like company industry, employee count, buying intent, and website visits to find and compare companies matching a user's Ideal Customer Profile (ICP). | ¶43 | col. 17:15-19 |
| [b] generating recommendations for engagement with the plurality of target clients, wherein components of the recommendations for engagement are based on determined similarities... | The system prioritizes companies for engagement using an "AI auto-generated scoring model" and presents "high-value prospects" in a "Recommendations" tab. | ¶44 | col. 17:20-29 |
| [c] categorizing a plurality of web pages, hyperlinks, and link structures located over the Internet using a trained classifier that uses features from content and code on web pages; | The system allegedly uses "proprietary algorithms that regularly crawl the web at scale, parse millions of public-facing websites, and build a web-wide index of people and company data." | ¶45 | col. 17:29-34 |
| [d] crawling, by a computer, the plurality of web pages, hyperlinks, and link structures based on the categorization of the plurality of web pages, hyperlinks, and link structures to collect third party unstructured text information; and | The system allegedly uses algorithms that are "continuously learning and improving" to "dig deep on leads" and gain insights like "intent data [and] employee trends." | ¶46 | col. 17:35-41 |
| [e] generating a feature matrix for the target client and comparing one or more value of the feature matrix of the target client with one or more values of a feature matrix for the entity's existing clients to generate the recommendations of engagement. | The "Recommendations" tab allegedly "appears to invoke an underlying matrix for the target client" which is used to make matrix-based comparisons with user-defined criteria. | ¶47 | col. 17:42-49 |
'964 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| [a] generating recommendations for engagement with the plurality of target clients, wherein components of the recommendations... are based on determined similarities between... a fitness, engagement, and intent characteristics... | The Apollo system prioritizes companies for engagement by assessing criteria, assigning weights, and using an "AI auto-generated scoring model" to identify high-potential prospects. | ¶56 | col. 18:28-39 |
| [b] categorizing a plurality of web pages, hyperlinks, and link structures located over the Internet using a trained classifier that uses features from content and code on web pages; | The system is described as having "proprietary algorithms that regularly crawl the web at scale, parse millions of public-facing websites, and build a web-wide index." | ¶57 | col. 18:40-45 |
| [c] crawling, by a computer, the plurality of web pages, hyperlinks, and link structures based on the categorization... to collect third party unstructured text information; and | The system is described as "continually optimizing" and "operating," using algorithms to "dig deep on leads" to gain insights such as intent data. | ¶58 | col. 18:46-52 |
| [d] generating a feature matrix for the target client and comparing one or more value... of the feature matrix of the target client with one or more values of a feature matrix for the entity's existing clients to generate the recommendations... | The "Recommendations" tab is alleged to invoke an "underlying matrix for the target client" to make "matrix-based comparisons with user-defined features and other criteria of existing clients." | ¶59 | col. 18:53-59 |
- Identified Points of Contention:
- A screenshot of a Glassdoor review alleges that Apollo's CEO "frequently instructs us to simply imitate our competitors without any original thinking." (Compl. p. 8). While not strictly technical, this visual evidence supports the complaint's narrative of copying, which may become relevant to questions of willfulness.
- Technical Question: A central point of contention may be whether the accused system performs "crawling... based on the categorization" as required by the claims (’609 Patent, cl. 1[d]; ’964 Patent, cl. 1[c]). The complaint alleges the system crawls and that its algorithms are "continuously learning" (Compl. ¶46), but the defense could argue this is distinct from the specific feedback loop described in the patent, where categorization results for some pages are used to "control or guide the crawl procedure" for other pages (’609 Patent, col. 6:50-52).
- Scope Question: The infringement case rests on mapping Apollo's functionalities to the claimed "fitness, engagement, and intent characteristics." A dispute may arise over whether the data points used by Apollo (e.g., "buying intent," "website visits") are coextensive with the scope of these terms as defined and used within the patent specification.
- Evidentiary Question: For the "feature matrix" element, the complaint alleges the system "appears to invoke an underlying matrix" (Compl. ¶¶ 47, 59). This phrasing suggests the allegation may be based on inference from the user interface's output rather than direct evidence of the system's internal architecture, which could be a focus during discovery.
V. Key Claim Terms for Construction
The Term: "fitness, engagement, and intent characteristics"
Context and Importance: This triad of characteristics forms the basis of the similarity comparison that drives the patented method. The outcome of the case may depend heavily on whether the data collected and analyzed by the Apollo platform (e.g., company size, ICP matching, buying signals) falls within the court's construction of these terms.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification provides non-limiting examples, suggesting the terms are illustrative categories. "Fitness characteristics" may include "average number of employees... type of business" (’609 Patent, col. 2:23-26). "Engagement characteristics" may include how clients are "requesting services, requesting help, providing feedback" (’609 Patent, col. 2:32-35). "Intent characteristics" can be "short term user intent as indicated by clicks on content, browsing of content, comments, etc." (’609 Patent, col. 2:40-43).
- Evidence for a Narrower Interpretation: A defendant could argue that the terms must be construed in light of the specific examples provided for "Company Alpha," a mobile device repair company, potentially limiting their scope. It could also be argued that all three distinct types of characteristics must be analyzed by an accused system to meet the limitation.
The Term: "crawling... based on the categorization"
Context and Importance: This phrase describes the core inventive concept highlighted by the patent and the Patent Examiner: an intelligent crawling process guided by a classification step. Practitioners may focus on this term because it distinguishes the claimed invention from a more conventional two-stage process of (1) performing a generic web crawl and (2) subsequently categorizing the collected data.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: Plaintiff may point to language suggesting a general guiding relationship, such as how categorization can "control or guide the crawl procedure" to "reduce the number of irrelevant pages crawled" (’609 Patent, col. 6:50-52), arguing that any influence of categorization on crawling behavior meets the claim.
- Evidence for a Narrower Interpretation: The defense may highlight Figure 2, which explicitly depicts a feedback loop where "Adjust crawling procedure/parameters" (220) influences the initial crawling steps (’609 Patent, Fig. 2). This could support an argument that the claim requires a specific, dynamic adjustment of crawling parameters in response to real-time categorization results, as described in the specification (’609 Patent, col. 6:45-49).
VI. Other Allegations
- Indirect Infringement: The complaint makes a passing reference to indirect infringement (Compl. ¶9), but the formal counts allege only direct infringement under 35 U.S.C. § 271(a) (Compl. ¶¶ 41, 54). The complaint does not plead specific facts to support the elements of inducement or contributory infringement, such as knowledge of the patents and intent to cause infringement by third parties.
- Willful Infringement: The term "willful" is not explicitly used in the complaint. However, the pleading lays a substantial foundation for such a claim by repeatedly alleging that Apollo engaged in a "deliberate internal policy of imitation" (Compl. ¶32), "copying competitors" (Compl. ¶8), and "cloning competitors' products" (Compl. ¶9). These allegations are supported by purported excerpts from internal employee reviews (Compl. ¶¶ 32, p. 8-10). The complaint alleges a strategy of copying ZoomInfo's technology, though it does not allege pre-suit knowledge of the specific patents-in-suit.
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of claim construction: Can the tripartite definition of "fitness, engagement, and intent characteristics" be construed to read on the specific types of data points and filters (e.g., ICP filters, buying intent signals) that the Apollo platform is alleged to use?
- A key evidentiary question will be one of technical operation: Can the plaintiff produce evidence that Apollo's system performs the specific feedback loop required by the claims—that its web crawling is dynamically guided by the results of its categorization step—or will discovery show two separate, sequential processes?
- A central theme of the complaint is deliberate copying: While not yet a formal willfulness count, the extensive allegations that Apollo built its business by imitating ZoomInfo, supported by purported internal reviews, raises a significant question of conduct that will likely influence the tenor of the litigation and could heavily impact damages if infringement is found.