Peer to Peer Magazine

P E E R T O P E E R : I L T A ' S Q U A R T E R L Y M A G A Z I N E | S P R I N G 2 0 1 9 13 managing the future of an entire practice, they have full and consistent perspectives and detail. Authoritative and Complete Search As we designed our data warehouse, it became apparent that no one system would be able to ingest and search the quantity and complexity of data to which we would have access. Every existing business system had advanced search capabilities, but only for the data in its own limited universe. Every time we added a chat bot, AI assistant, document analysis system, or other so-called innovation, we had to contend with yet another unique embedded search at their core. It was unreasonable to expect our team to adequately support and tune so many search engines and, worse, we marketing intelligence. To support these process flows, we also subscribe to dozens of premier information sources. As a result, we have great business systems, but also a high class problem of "great silos" of data. While we leverage APIs to trade data across systems, no system has more than a small percentage of the overall information. So, for years, we've been laying the groundwork for aggregating our data across these silos, in anticipation of assembling a big data warehouse. We've been doing the boring but critical early work of standardizing granular matter tagging, enabling detailed matter opening and conflicts data collection, instituting phase and task tracking for every matter, and so on. Implementing our new business systems helped improve the quality and availability of our internal data, but it didn't address the challenge of providing the added context that external data sources had to offer. Requests to data and legal analytics vendors to download and blend their data with our own were previously met with incredulous stares. Now, increasingly, vendors are coming to recognize that sharing data creates both a revenue stream and an opportunity to establish mutually beneficial partnerships with customers. Master Firm Data Collection Data now forms the core of our business systems. Our data warehouse pulls together data from our internal systems and dozens of external sources, matching entities (people, companies, matters) across sources in a carefully designed structure. We maintain information on matter parties, third-party experts, court records, relevant lawyers, company governance, and the correlations between them all. Today, we have over 500,000 entities and 250M data points, with an expectation of aggressive growth from this foundational level. Having a comprehensive core that any system can access means that whether a person is pricing a matter, helping a client, marketing, hiring, improving processes, or saw a future where searchers would get incomplete and different answers from any one of those solutions. We wanted a single, authoritative source that would provide the official Winston answers, whether a person is chatting with a bot on their phone or searching our Winston Intelligence platform. Although our master data plan didn't include undertaking a leading edge search project, it was clear we needed to create our own search capabilities to make the most of our rich data collection. As we considered our options, we noticed that most of our vendors (and most vendors marketing AI-based solutions) are using Elasticsearch, or its brother, Solr, at their core. While the prevalence of Elasticsearch didn't ensure it was the right option for our enterprise use, we knew that SharePoint search was becoming

• Cover