DCT
1:25-cv-01112
Onepass Data Technology LLC v. Verizon Communications Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: OnePass Data Technology LLC (Ohio)
- Defendant: Verizon Communications, Inc. and Cellco Partnership d/b/a Verizon Wireless (Delaware)
- Plaintiff’s Counsel: Wilks Law LLC; Leichtman Law PLLC
- Case Identification: 1:25-cv-01112, D. Del., 09/05/2025
- Venue Allegations: Venue is alleged to be proper in the District of Delaware because both Verizon defendants are incorporated in and reside in Delaware.
- Core Dispute: Plaintiff alleges that Defendant’s systems and methods for cleansing mailing list addresses, used to qualify for discounted U.S. Postal Service postage rates, infringe patents related to high-efficiency, single-pass data processing architectures.
- Technical Context: The technology concerns software methods for efficiently standardizing and correcting large databases of mailing addresses by minimizing data transfers between a computer's main memory and remote storage, a critical task for large-scale mailers.
- Key Procedural History: The complaint notes that U.S. Patent No. 7,376,680, one of the two patents-in-suit, successfully underwent an Ex Parte Reexamination at the U.S. Patent and Trademark Office, which confirmed the eligibility of its claims. U.S. Patent No. 8,682,866 is a continuation of the application that led to the '680 patent, indicating a shared technical disclosure.
Case Timeline
| Date | Event |
|---|---|
| 2003-04-07 | '680 and '866 Patents - Earliest Priority Date |
| 2008-05-20 | '680 Patent - Issue Date |
| 2011-08-01 | Start of USPS CASS Cycle N Period Mentioned in Complaint |
| 2014-03-25 | '866 Patent - Issue Date |
| 2017-04-26 | '680 Patent - Ex Parte Reexamination Certificate Issued |
| 2023-08-01 | Start of USPS CASS Cycle O Period Mentioned in Complaint |
| 2025-09-05 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 7,376,680 - "System And Method For Cleansing, Linking And Appending Data Records Of A Database"
- Patent Identification: U.S. Patent No. 7,376,680, issued May 20, 2008 (the “’680 Patent”). (Compl. ¶9).
The Invention Explained
- Problem Addressed: The patent describes conventional data cleansing systems as inefficient because they process a database against multiple reference files sequentially. This requires reading each data record from remote storage (e.g., a hard disk) and writing it back for each reference file processed, which is described as "exceedingly costly in both time and computer processor resources." (’680 Patent, col. 1:57-2:5).
- The Patented Solution: The invention proposes a "single pass" method where a data record is read from remote storage only once, processed against multiple different reference files while being held in the computer’s main memory (RAM), and then written to an output file only once. (’680 Patent, Abstract; col. 3:15-22). This architecture, illustrated in Figure 2 of the patent, is designed to eliminate the redundant data read/write operations that slowed down prior art systems.
- Technical Importance: The single-pass approach was aimed at delivering significant performance improvements for processing extremely large databases, a key operational challenge in the direct mail and data management industries. (’680 Patent, col. 1:47-56).
Key Claims at a Glance
- The complaint asserts independent claim 1 and dependent claims 2-7. (Compl. ¶34).
- Essential elements of independent claim 1 include:
- A method for creating a cleansed output file from a "single pass" through an input file.
- Selecting an input file and a reference file.
- Computing a search key.
- For each data record, performing a series of steps including retrieving the record, searching the reference file using a matcher process, determining a matching record, creating a new cleansed record, and writing it to an output file.
- A final limitation requires that these steps "are performed in a single pass," which is defined to mean "each data record of said input file is read from a remote storage location only once...and each said new cleansed data record...is written to a remote storage location only once."
U.S. Patent No. 8,682,866 - "System And Method For Cleansing, Linking And Appending Data Records Of A Database"
- Patent Identification: U.S. Patent No. 8,682,866, issued March 25, 2014 (the “’866 Patent”). (Compl. ¶11).
The Invention Explained
- Problem Addressed: As a continuation of the '680 Patent's application, the ’866 Patent addresses the same problem of inefficiency in conventional multi-pass database processing. (’866 Patent, col. 2:1-5).
- The Patented Solution: The invention describes a method and system for "data hygiene" where an input data record is retrieved from "remote storage" a single time, stored in "local memory," and then processed through a "series of updating processes" against different reference files while remaining in that local memory. After all processing is complete, the final cleansed record is written from local memory back to remote storage. (’866 Patent, Abstract; claim 1). This preserves the core efficiency concept of minimizing I/O operations with remote storage.
- Technical Importance: This architecture provides a framework for performing complex, multi-stage data cleansing operations efficiently on large datasets by centralizing the processing in a computer's faster local memory. (’866 Patent, col. 4:41-46).
Key Claims at a Glance
- The complaint asserts independent method claim 1, independent system claim 16, and dependent claims 2-3 and 9. (Compl. ¶56, ¶67).
- Essential elements of independent claim 1 (method) include:
- Obtaining an input data record from "remote storage."
- Storing the record in a "local memory."
- Processing the record through a "series of updating processes" against a plurality of different reference files.
- Outputting the cleansed record from local memory to remote storage, with the input record having been retrieved from remote storage only a "single time."
- Essential elements of independent claim 16 (system) include:
- A "local memory."
- A plurality of reference files.
- A "matcher" that updates an input data record while it "remains in said local memory."
- A "secondary matcher" that performs another type of matching while the record "remains in said local memory."
- "means for writing" the record to an output file, where the record is "read and written only once."
III. The Accused Instrumentality
Product Identification
The complaint identifies the accused instrumentalities as Verizon's "Accused Methods" and "Accused Systems" for cleansing mailing list address records. (Compl. ¶24).
Functionality and Market Context
- The accused systems are alleged to use CASS-certified software to standardize, correct, and verify Verizon's customer mailing lists to meet U.S. Postal Service requirements for automated mail processing under "Cycle N" or "Cycle O" standards. (Compl. ¶27-¶28).
- The purpose of this functionality is to qualify for reduced "automation prices" on postage for Verizon's First-Class Mail® and Standard Mail®, which represents a significant cost savings. (Compl. ¶26, ¶36). The complaint provides an image of a Verizon mailing that displays a "Presorted First-Class Mail U.S. Postage Paid" permit imprint as evidence of Verizon's participation in such programs. (Compl. p. 7).
- The complaint alleges these systems comprise "non-volatile memory" for storing input and reference files (e.g., USPS ZIP + 4®, DPV®, LACSLink®) and "volatile memory" for processing the records. (Compl. ¶27).
IV. Analysis of Infringement Allegations
’680 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| A computer-implemented method for creating a cleansed output file... from a single pass through an input file. | Verizon uses its Accused Methods to create a cleansed customer mailing list to qualify for automation prices. | ¶38 | col. 26:15-19 |
| retrieving said data record from said input file on remote storage. | Verizon's methods retrieve an address record from an input customer mailing list maintained in non-volatile memory. | ¶42 | col. 26:29-31 |
| searching said reference file with a matcher process... generating a candidate data record list. | Verizon's methods search USPS reference files using a matching process to generate a candidate address record list. | ¶43 | col. 26:32-38 |
| cleansing said data record of said input file according to said matching data record, thereby generating verified information. | Verizon's methods cleanse the address record according to a matching record from a USPS reference file. | ¶46 | col. 26:42-44 |
| wherein said steps... are performed in a single pass... such that each data record of said input file is read from a remote storage location only once... and each said new cleansed data record... is written to a remote storage location only once. | Verizon's methods are alleged to perform the steps such that each input mailing record is read from non-volatile memory only once, and each cleansed record is written to non-volatile memory only once. | ¶49 | col. 26:45-56 |
’866 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| obtaining an input data record of an input file maintained in remote storage. | Verizon's methods obtain an address record from a customer mailing list maintained on non-volatile memory. | ¶61 | col. 25:1-3 |
| storing said input data record in a local memory. | Verizon's methods store the obtained address record in volatile memory. | ¶62 | col. 25:4-5 |
| processing said input data record, through a series of updating processes... wherein a different reference file is used for each of said updating processes... | Verizon's methods process the address record through a series of cleansing steps against different USPS reference files (e.g., postal coding, address verification, etc.). | ¶63 | col. 25:6-18 |
| outputting said cleansed input data record from said local memory to said remote storage, wherein said input data record is retrieved from said remote storage a single time... | Verizon's methods write the cleansed address record from volatile memory to non-volatile memory. | ¶66 | col. 25:40-44 |
Identified Points of Contention
- Scope Questions: A primary issue may be whether the architectural concepts described in the patents map cleanly onto Verizon's accused systems. For both patents, the analysis will question whether the accused system's use of "volatile memory" for processing and "non-volatile memory" for storage (Compl. ¶27) satisfies the claimed distinction between "local memory" and "remote storage."
- Technical Questions: The complaint alleges on "information and belief" that Verizon's process meets the critical "read only once" and "write only once" limitations. (Compl. ¶49, ¶79). A key question for discovery will be what factual evidence exists to demonstrate that Verizon's real-world, large-scale systems adhere to this specific single-I/O-operation constraint. For the system claims of the ’866 Patent, which invoke "means-plus-function" language, a further question is whether the structure of Verizon's "matcher software module" is structurally equivalent to the corresponding algorithms disclosed in the patent specification.
V. Key Claim Terms for Construction
The Term: "single pass" (’680 Patent, claim 1)
- Context and Importance: This term is central to the asserted method claim of the ’680 Patent. The infringement analysis depends on whether Verizon's multi-step cleansing process can be characterized as a "single pass."
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The claim itself provides a functional definition, stating that a "single pass" is one in which "each data record of said input file is read from a remote storage location only once...and each said new cleansed data record...is written to a remote storage location only once." (’680 Patent, col. 26:48-56). This language may support an interpretation where any process meeting this functional I/O requirement qualifies, regardless of the number of internal processing stages.
- Evidence for a Narrower Interpretation: The specification contrasts the invention with prior art systems that "perform one operation at a time," resulting in multiple file reads and writes. (’680 Patent, col. 1:57-66). This context could support a narrower definition that requires a more holistic or monolithic process, not merely a series of discrete processes executed sequentially in memory before a final write.
The Term: "local memory" and "remote storage" (’866 Patent, claims 1 and 16)
- Context and Importance: The architectural distinction between these two types of memory is foundational to the claims of the ’866 Patent. The plaintiff's infringement theory maps "volatile memory" to "local memory" and "non-volatile memory" to "remote storage." (Compl. ¶61-62, ¶72, ¶79).
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification explicitly identifies "main memory 106, preferably random access memory (RAM)" and separately identifies "secondary memory 108" which includes a "hard disk drive 110 and/or a removable storage drive 112." (’866 Patent, col. 7:29-33). This provides direct support for equating "local memory" with RAM (volatile) and "remote storage" with disk-based storage (non-volatile).
- Evidence for a Narrower Interpretation: The patent was filed based on a 2003 priority date. A defense may argue that in modern, complex computing environments with sophisticated caching, network-attached storage, and tiered memory systems, the simple binary distinction between RAM and a hard disk as described in the specification does not accurately capture the architecture of the accused systems, raising questions about the applicability of these terms.
VI. Other Allegations
Indirect Infringement
The complaint includes allegations that if any steps were performed by a third party, Verizon "directed and/or controlled each such step." (Compl. ¶36, ¶58). These allegations appear directed at satisfying the standard for divided infringement under 35 U.S.C. § 271(a) rather than pleading separate counts for induced or contributory infringement under § 271(b) or (c).
Willful Infringement
The complaint does not allege pre-suit knowledge of the patents but includes a request in the prayer for relief for enhanced damages if willfulness is later established through discovery. (Compl. p. 20).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of architectural mapping: does the alleged operation of Verizon's address cleansing systems, which separates tasks between "volatile" and "non-volatile" memory, align with the patents' claimed distinction between "local memory" and "remote storage" and the "single pass" processing framework? The resolution will depend heavily on claim construction and how these 2003-era terms apply to modern computing architectures.
- A second key issue will be one of evidentiary proof: the complaint's central factual premise—that Verizon's systems read each input record from remote storage only once and write each cleansed record only once—is alleged "on information and belief." The case may turn on whether discovery uncovers technical evidence to substantiate this highly specific operational constraint within Verizon's complex, real-world systems.
- A third question will be the viability of the means-plus-function claims in the '866 Patent. To prove infringement of system claim 16, the plaintiff will face the heightened evidentiary burden of demonstrating that the accused "matcher software module" is structurally equivalent to the specific algorithms disclosed in the patent's specification, a frequent point of failure for such claims.