DCT
1:18-cv-00963
Applied Predictive Tech Inc v. Marketdial Inc
Key Events
Complaint
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Applied Predictive Technologies, Inc. (Delaware)
- Defendant: MarketDial, Inc. (Delaware) and John M. Stoddard
- Plaintiff’s Counsel: Morris, Nichols, Arsht & Tunnell LLP; Dentons US LLP
- Case Identification: 1:18-cv-00963, D. Del., 06/28/2018
- Venue Allegations: Venue is asserted based on Defendant MarketDial, Inc. being a Delaware corporation and therefore residing in the district.
- Core Dispute: Plaintiff alleges that Defendant’s MarketDial System, a software platform for A/B testing, infringes a patent related to methods for determining optimal parameter settings for business initiative testing models.
- Technical Context: The technology lies in the field of data-driven business analytics, where software is used to test the efficacy of corporate initiatives (e.g., promotions, operational changes) to guide strategic decisions.
- Key Procedural History: The complaint alleges that Defendant's co-founder, John Stoddard, previously had access to Plaintiff's confidential information through a third-party contractual relationship. Plaintiff states it sent a letter to Defendants on December 18, 2017, identifying the patent-in-suit and raising infringement concerns, which did not lead to a resolution. The patent-in-suit is identified as a continuation-in-part of an earlier application.
Case Timeline
| Date | Event |
|---|---|
| 2004-01-30 | '916 Patent Priority Date |
| 2013-10-29 | U.S. Patent No. 8,571,916 Issues |
| 2015-02-XX | MarketDial is allegedly founded |
| 2015-04-02 | MarketDial registers as a foreign corporation in Utah |
| 2017-12-18 | Plaintiff sends pre-suit notice letter to Defendants |
| 2018-06-28 | Complaint Filing Date |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 8,571,916 - "Methods, Systems, and Articles of Manufacture for Determining Optimal Parameter Settings for Business Initiative Testing Models," issued October 29, 2013
The Invention Explained
- Problem Addressed: The patent's background describes the challenge businesses face in accurately measuring the impact of new initiatives (like marketing campaigns or price changes) tested in a small number of locations. The way analysts set up these tests—for instance, by selecting which "control" locations to compare against—is often based on intuition, which can introduce inconsistencies ("noise") into the data and obscure the true results. (’916 Patent, col. 1:41-67).
- The Patented Solution: The invention is a method to "test the test" itself. Before running a real-world business experiment, the system performs "virtual tests" using historical data. It simulates various test configurations (e.g., using different parameter settings for how to handle outlier data or select control groups) on "virtual test sites" where no actual initiative was ever implemented. The system then calculates a "noise value" for each simulated configuration. The configuration that produces the minimum noise is identified as the optimal set of parameters, which is then used to configure the actual business initiative test, thereby increasing its accuracy. (’916 Patent, Abstract; col. 2:15-34).
- Technical Importance: This technology allows a business to move from subjective, anecdote-based test design to an empirical, data-driven methodology for optimizing the design of the test itself, aiming to improve the reliability of A/B testing outcomes. (Compl. ¶22; ’916 Patent, col. 2:5-11).
Key Claims at a Glance
- The complaint asserts independent claim 1. (Compl. ¶95, 98).
- The essential elements of independent claim 1 include:
- identifying a business initiative testing model with parameter settings;
- selecting a parameter setting set to perform a virtual test;
- performing the virtual test on virtual test sites where no actual test was implemented;
- determining actual performance data for the virtual test sites;
- determining actual performance data for a set of control group sites;
- determining a "noise value" reflecting inconsistency between the virtual test site and control group site performance data;
- determining a set of optimal parameter settings that best minimize the noise from the virtual test; and
- configuring the testing model with the optimal parameter settings to test a business initiative.
- The complaint reserves the right to assert additional claims after discovery. (Compl. ¶109).
III. The Accused Instrumentality
Product Identification
- The "MarketDial System" or "Accused System," a software-as-a-service platform for business analytics. (Compl. ¶¶6, 66).
Functionality and Market Context
- The Accused System is marketed as a tool to "construct and analyze a perfect A/B test." (Compl. p. 17). It includes an "EASY STEP-BY-STEP TEST BUILDER" that guides users through creating a test, allegedly by helping them "select the quantity stores sampled, which are the optimal stores to include in the sample, ideal test length, and more." (Compl. p. 18). The system is promoted as having been "[b]uilt by ex-McKinsey and -BCG consultants." (Compl. ¶30). The complaint alleges that the Accused System performs functionalities "substantially similar—if not virtually identical, in certain aspects—to APT's products and services." (Compl. ¶54).
IV. Analysis of Infringement Allegations
'916 Patent Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| identifying, by a computer, a business initiative testing model having a set of parameter settings | The Accused System is alleged to be a business initiative testing model with parameter settings, such as those affecting the use of outlier data and the amount of historical performance data to be analyzed. | ¶99 | col. 25:62-65 |
| selecting a first parameter setting set for performing a virtual test... | The Accused System is alleged to select parameter setting options, such as how to treat outlier data (e.g., consider as-is, disregard, or weigh) and how much historical data to use (e.g., two vs. four years). | ¶¶100-101 | col. 25:66-26:3 |
| performing, by a computer, the virtual test on a set of virtual test sites...wherein each virtual test is a simulated business initiative test...where no actual initiative test has been implemented... | On information and belief, the Accused System performs a "virtual test" using historical data from sites where no actual test was implemented to simulate an initiative and thereby optimize test parameters. | ¶103 | col. 26:4-13 |
| determining, by a computer, actual performance data associated with the set of virtual test sites | The Accused System allegedly obtains and uses historical customer data, such as at least two years of sales data, for the virtual test sites. A screenshot shows an "AUTOMATED DATA UPLOAD" feature. | ¶104; p. 31 | col. 26:14-16 |
| determining, by a computer, actual performance data associated with a set of control group sites... | The Accused System allegedly obtains and uses historical customer data for the control group sites. | ¶105 | col. 26:17-21 |
| determining a noise value for the first parameter setting set, the noise value reflecting an inconsistency between performance data... | The Accused System is alleged to use a virtual test to compare performance data between virtual test sites and control sites, thereby determining a "noise value" associated with each parameter setting. | ¶106 | col. 26:22-29 |
| determining, by a computer, a set of optimal parameter settings...whereby the optimal parameter settings best minimize noise from the results | The Accused System allegedly optimizes the test by choosing parameter settings that ensure the performance data of the virtual test and control groups are "best matched to each other," which is alleged to be the method for minimizing noise. | ¶107 | col. 26:30-35 |
| configuring, by a computer, the business initiative testing model using the optimal parameter settings to test a business initiative for application in the business network | Once optimal settings are found, the Accused System allegedly uses them to configure the testing model. A screenshot of the "Create Test" user interface is provided as evidence of this functionality. | ¶108; p. 33 | col. 26:36-40 |
- Identified Points of Contention:
- Scope Questions: A central question for the court will be whether the accused system’s function of helping a user "build an accurate test" (Compl. p. 18) by selecting "optimal stores" is the same as the patent's claimed method of performing an iterative "virtual test" to minimize "noise." The infringement case may turn on whether the accused system actually performs the claimed simulation loop to optimize test parameters, or if it performs a more conventional analysis to select statistically similar test and control stores.
- Technical Questions: The complaint's infringement allegations are made on "information and belief," pending discovery of the system's source code (Compl. ¶67). A key technical question is what evidence will show that the accused system's underlying algorithm performs the specific function of calculating a "noise value" reflecting an "inconsistency" between a simulated test group and a control group, as required by the claim, rather than a more generic statistical matching function.
V. Key Claim Terms for Construction
- The Term: "virtual test"
- Context and Importance: This term is the central mechanism of the invention. The definition of what constitutes a "virtual test" will be critical to the infringement analysis, as Plaintiff must prove the accused system performs this specific process.
- Intrinsic Evidence for a Broader Interpretation: The specification refers to the process as a "mock initiative test" and a "simulation environment," which could support a construction covering a wide range of simulations using historical data to model a test's properties. (’916 Patent, col. 18:15-23).
- Intrinsic Evidence for a Narrower Interpretation: Claim 1 defines a "virtual test" as being performed on sites "where no actual initiative test has been implemented" and for the purpose of determining "optimal parameter settings" that "best minimize noise." (’916 Patent, col. 26:8-10, 33-35). This suggests a more limited scope tied to a specific process: simulating a test on non-test sites by varying parameters to find the configuration with the least statistical noise.
- The Term: "noise value"
- Context and Importance: This term defines the metric that the "virtual test" seeks to minimize. Its construction will determine the specific calculation the plaintiff must prove the accused system performs. Practitioners may focus on this term because the allegation of "minimizing noise" is the core of the claimed optimization.
- Intrinsic Evidence for a Broader Interpretation: The specification provides a general definition of "noise" as a "quantified measurement of inconsistent performance data," which could be argued to cover any metric of statistical variance or error. (’916 Patent, col. 17:39-41).
- Intrinsic Evidence for a Narrower Interpretation: Claim 1 explicitly defines the "noise value" as "reflecting an inconsistency between performance data associated, with the set of virtual test sites and performance data associated with the set of control group sites." (’916 Patent, col. 26:24-27). An example in the specification quantifies noise as the difference in performance lift trends between the two groups (e.g., 3% lift vs. 2.5% lift), suggesting it is a specific calculation of differential trends, not just a general variance measure. (’916 Patent, col. 21:42-53).
VI. Other Allegations
- Indirect Infringement: The complaint does not contain a count for indirect infringement.
- Willful Infringement: The complaint alleges that Defendant MarketDial's infringement is willful. This allegation is based on MarketDial having received a letter from Plaintiff dated December 18, 2017, which allegedly provided notice of the ’916 Patent and the infringement concerns. (Compl. ¶¶111-112).
VII. Analyst’s Conclusion: Key Questions for the Case
- A core issue will be one of technical operation versus claim scope: Does the accused MarketDial System's functionality for helping users build an "accurate test" by selecting optimal stores perform the specific, iterative "virtual test" method claimed in the ’916 Patent? The case will likely depend on whether MarketDial’s software merely finds statistically similar control stores, or if it executes the claimed process of simulating tests with varying parameters on non-test sites to find a configuration that minimizes a calculated "noise value."
- A key evidentiary question will be what the accused system's source code reveals. The complaint's infringement theory is based on publicly available materials and "information and belief." The outcome of the patent claim will hinge on whether discovery provides evidence that the internal logic of the Accused System performs the specific, multi-step optimization method required by Claim 1, or if there is a fundamental mismatch in technical operation.