DCT
2:19-cv-00496
Applied Predictive Tech Inc v. Marketdial Inc
Key Events
Amended 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: Dentons US LLP
- Case Identification: 2:19-cv-00496, D. Utah, 08/29/2019
- Venue Allegations: Venue is alleged to be proper in the District of Utah because Defendant MarketDial has a regular and established place of business in the district.
- Core Dispute: Plaintiff alleges that Defendant’s business analytics software platform infringes a patent related to methods for optimizing the parameters of business initiative testing models.
- Technical Context: The technology operates in the "big data" analytics sector, addressing the challenge retailers face in accurately measuring the financial impact of business initiatives like sales promotions or operational changes.
- Key Procedural History: The complaint alleges that Plaintiff provided Defendant with actual notice of the asserted patent via a letter dated December 18, 2017, a fact that may be relevant to the claim for willful infringement. The complaint also contains extensive allegations of a pre-existing relationship between the companies' principals, which form the basis for parallel trade secret misappropriation claims.
Case Timeline
| Date | Event |
|---|---|
| 2004-01-30 | Earliest Priority Date for ’916 Patent |
| 2013-10-29 | U.S. Patent No. 8,571,916 Issues |
| 2015-02-01 | MarketDial Co-Founded |
| 2017-12-18 | Plaintiff sends letter to Defendant identifying ’916 Patent |
| 2019-08-29 | Second Amended Complaint Filed |
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"
- Patent Identification: 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 identifies a problem for businesses, particularly retailers, in accurately measuring the true impact of new initiatives. The analysis is often clouded by "inconsistent data" or statistical "noise," making it difficult to determine if a change was genuinely effective and should be rolled out more broadly (Compl. ¶29; ’916 Patent, col. 2:1-10).
- The Patented Solution: The invention provides a computer-implemented method to empirically determine the optimal parameters for a testing model before it is used to analyze a real-world initiative. It does this by running "virtual tests"—simulations on historical data for locations where no actual test occurred—and iteratively trying different analytical parameter settings (e.g., how to handle outlier data). The system identifies the parameter settings that best "filter out noise," defined as the inconsistency between the virtual test group and a control group. These optimized parameters are then used to configure the model for analyzing actual business initiatives, thereby increasing the accuracy of the results (Compl. ¶¶31, 33; ’916 Patent, col. 17:33-49).
- Technical Importance: The claimed method aims to replace intuitive or anecdotal approaches to test design with an automated, data-driven process for refining the analytical model itself, addressing a "big data" challenge for businesses (Compl. ¶¶29, 37).
Key Claims at a Glance
- The complaint asserts independent claim 1 of the ’916 Patent (Compl. ¶114).
- The essential elements of independent claim 1 include:
- Identifying a business initiative testing model having a set of parameter settings.
- Selecting a set of parameter options for performing a "virtual test."
- Performing the virtual test on a set of "virtual test sites" (locations where no actual test was implemented) using varied parameter settings.
- Determining actual performance data for the virtual test sites and a set of control group sites.
- Determining a "noise value" that reflects the inconsistency between the performance data of the two groups.
- Determining a set of optimal parameter settings based on the settings that best minimize the noise from the virtual test.
- Configuring the testing model with the optimal parameter settings for use in testing a real business initiative.
- The complaint reserves the right to assert additional claims of the ’916 Patent after discovery (Compl. ¶128).
III. The Accused Instrumentality
Product Identification
- The "MarketDial System," an A/B testing software platform (Compl. ¶¶7, 84).
Functionality and Market Context
- The Accused System is a software tool that allows users to "construct and analyze a perfect A/B test" for business initiatives such as promotions, pricing changes, and store remodels (Compl. ¶¶84-85).
- Functionality described on the defendant's website includes an "EASY STEP-BY-STEP TEST BUILDER" that helps users select optimal stores and test length, and features for automated data upload and store attribute analysis (Compl. ¶¶84, 118, 127). A screenshot from the defendant's website shows an interface for creating a test and calculating an "Estimated Confidence" level based on the number of stores and testing weeks (Compl. ¶¶127, 145).
- The complaint alleges the Accused System was developed using Plaintiff's trade secrets and is marketed as being "[b]uilt by ex-McKinsey and -BCG consultants," referencing the past employment of Defendant's founders (Compl. ¶¶51, 95).
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 identifies a testing model with parameter settings, including parameters affecting the use of outlier data and the amount of historical performance data to be analyzed. | ¶118 | col. 27:61-63 |
| selecting a first parameter setting set for performing a virtual test... | The Accused System selects parameter setting options, such as whether to consider, disregard, or weigh outlier data, and how many years of historical data to use. | ¶¶119-120 | col. 27:64-28:2 |
| performing, by a computer, the virtual test on a set of virtual test sites... wherein each virtual test is a simulated business initiative test performed on test sites where no actual initiative test has been implemented... | The Accused System allegedly performs a virtual test on sites using their actual historical performance data from a period prior to any initiative implementation. A screenshot from MarketDial's website depicts a "Create Test" interface used to set up these tests. | ¶122, ¶127 | col. 17:40-42 |
| determining, by a computer, actual performance data associated with the set of virtual test sites; | The Accused System obtains and uses actual historical sales data provided by its customers for the virtual test sites. A marketing image from Defendant's website describes "AUTOMATED DATA UPLOAD" functionality. | ¶123, ¶43 | col. 28:6-8 |
| determining, by a computer, actual performance data associated with a set of control group sites... | The Accused System obtains historical sales data for control group sites from the pre-period before any business initiative. | ¶124 | col. 28:9-13 |
| determining a noise value for the first parameter setting set, the noise value reflecting an inconsistency between performance data associated, with the set of virtual test sites and... control group sites... | The Accused System allegedly uses a virtual test to determine optimal parameters by comparing performance data between the virtual test sites and control group sites, thereby determining a noise value. | ¶125 | col. 17:36-38 |
| determining, by a computer, a set of optimal parameter settings for the business initiative testing model based on results from the virtual test whereby the optimal parameter settings best minimize noise from the results; | The Accused System allegedly chooses parameter setting options that optimize the test by best matching the performance data of the virtual test and control groups, thereby minimizing noise. | ¶126 | col. 28:22-26 |
| 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 the optimal parameter settings are determined, the Accused System allegedly uses them to configure the business initiative testing model for analyzing an actual initiative. | ¶127 | col. 28:27-30 |
- Identified Points of Contention:
- Technical Questions: The complaint alleges on "information and belief" that the Accused System performs the specific, iterative virtual test process to optimize model parameters. A key question for the court will be what technical evidence, presumably from source code and internal documents, supports this allegation. The marketing materials provided in the complaint describe setting up and analyzing A/B tests but do not explicitly detail a process of running simulations to optimize the test parameters themselves by minimizing a "noise value."
- Scope Questions: The infringement case may turn on the scope of the term "virtual test". The question for the court will be whether the term is limited to the patent's specific, iterative process of testing multiple parameter combinations to find one that minimizes noise, or if it could be construed more broadly to cover any pre-test simulation that uses historical data from test and control locations to inform the setup of a single, real-world A/B test.
V. Key Claim Terms for Construction
The Term: "virtual test"
Context and Importance: This term is central to the claimed invention. Infringement will depend on whether the Accused System's functionality meets the definition of a "virtual test". Practitioners may focus on this term because the dispute will likely center on whether the defendant's system performs the specific, iterative, parameter-optimizing simulation required by the claims, or a more conventional pre-test analysis.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification describes the process as creating a "simulation environment" and performing a "mock initiative test," language that could potentially support a broader reading covering various forms of pre-test simulation (Compl. ¶31, citing ’916 Patent, 17:33-49).
- Evidence for a Narrower Interpretation: The claim links the "virtual test" to a specific outcome: finding "optimal parameter settings" that "best minimize noise." An exemplary embodiment describes the virtual test being "repeated for each combination of parameter settings," suggesting a specific, iterative process rather than a one-off simulation (Compl. ¶39, citing ’916 Patent, 20:2-8).
The Term: "noise value"
Context and Importance: The objective of the claimed "virtual test" is to find parameters that minimize this "noise value". The construction of this term is therefore critical to defining the required functionality.
Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The parties could dispute whether this term covers any generic measure of statistical variance or error that an A/B testing platform might seek to reduce.
- Evidence for a Narrower Interpretation: The claim itself provides a definition, stating the term reflects "an inconsistency between performance data associated, with the set of virtual test sites and performance data associated with the set of control group sites" (Compl. ¶44). The specification further describes noise as a "quantified measurement of inconsistent performance data for sites used in the analysis performed by the model," tying the term directly to the comparison of the two simulated groups within the patented method (Compl. ¶31, citing ’916 Patent, 17:36-38).
VI. Other Allegations
- Willful Infringement: The complaint alleges that Defendant's infringement has been willful. The basis for this allegation is a letter dated December 18, 2017, through which Plaintiff allegedly provided Defendant with actual notice of the ’916 Patent. The complaint alleges that Defendant's infringement continued after receiving this notice (Compl. ¶¶130-131).
VII. Analyst’s Conclusion: Key Questions for the Case
- A key evidentiary question will be one of technical proof: what evidence will emerge from discovery to show that the MarketDial System's internal architecture performs the specific iterative "virtual test" recited in claim 1? The complaint's allegations, made on "information and belief," will require technical evidence to substantiate the claim that the accused software runs simulations for the express purpose of identifying optimal analytical parameters by minimizing a calculated "noise value," as opposed to conducting a more conventional analysis to set up a test.
- A central legal issue will be one of claim scope: can the term "virtual test", as defined and used in the patent, be construed to cover the A/B test setup and analysis features shown in Defendant's marketing materials? The resolution of the case will likely depend on whether the court adopts a narrow construction requiring a multi-iteration, noise-minimizing feedback loop to find optimal model parameters, or a broader construction that could read on other forms of pre-test simulation.