DiscoverText

Overview

DiscoverText is a cloud service for analyzing multilingual text and its associated metadata. It is best suited to research and document-review teams that need both machine classification and human coding; its strongest case is a workflow that connects those methods with eDiscovery tools.

Key features

Classification with human review

Reusable machine classifiers, called “sifters,” can surface relevant items and sort text into topics, sentiment, or other categories. Pairing those classifiers with annotation, crowdsourcing, adjudication, and iteration gives teams a way to combine automated sorting with human coding and review. That breadth suits projects where classification needs oversight; it may be more than a reader needs for basic text analysis.

DiscoverText is intended for free text and metadata from sources such as surveys, customer feedback, CRM, chats, email, public comments, and RSS feeds. It can no longer collect Twitter data in real time, and it has not been connected to Facebook’s open API since 2014. Researchers can upload non-Twitter social data they already hold in spreadsheets, but should not choose the service expecting it to collect social data from those platforms.

eDiscovery and API

Deduplication and near-duplicate clustering help organize document collections, while redaction can produce PDF collections with Bates stamps and spreadsheet indexes. These are useful capabilities for document-review work, not just text categorization.

API access is for Enterprise license holders who have been granted an API key. Documented operations cover projects, archives, buckets, datasets, deduplication, clustering, and tag clouds; search, word-cloud building, and some reporting are marked as forthcoming. The API therefore supports programmatic workflows, but its access is not part of the Professional plan.

Pricing

Professional costs $99 a month, billed $99 / month. It includes 100,000 Units, 25 Live Data Feeds, one Professional Seat, no Enterprise Seats, unlimited projects, analytics and reporting, adjudication tools, Cloud Explorer, standard support, and Gnip availability. It is the practical starting point for an individual who needs the core analysis and adjudication tools, but the single seat limits collaboration within the plan.

Enterprise costs $3,000 a month, billed $3,000 / month. It raises the allocation to 1,000,000 Units and 100 Live Data Feeds, with one Enterprise Seat and two Professional Seats. It adds Top Meta, Active Learning, interactive histograms, the FOIA Toolkit, and priority support while retaining unlimited projects and Gnip availability. This is for teams that need those advanced tools and larger quotas; the steep price jump makes it a poor fit for small or occasional projects.

Custom Enterprise has custom pricing and a custom account package. DiscoverText also offers a free plan, though no general free-plan quota or feature set is stated. Students and professors can receive free access, training, and project support, and academic group licenses are free. A free trial is promoted as up to 30 days with enterprise-level features, while the signup page calls it a 7-day trial, so readers should confirm the applicable trial length before relying on it.

Platforms

DiscoverText is cloud-deployed, with web access and an API for eligible Enterprise license holders. It is a natural fit for browser-based work; API access depends on the license and an issued key.

Who it's for

Researchers and organizations handling substantial text collections are the clearest fit, especially when a project needs repeatable machine classification alongside human annotation, adjudication, or eDiscovery document handling. Academic users have a particularly strong reason to consider it because free access, training, project support, and free group licenses are available. Readers who need real-time Twitter collection, broad social-platform collection, or API access on the entry plan should look elsewhere.

Pros and cons

• Pro: Reusable sifters and human coding workflows bring machine classification and human review into the same process.
• Pro: Deduplication, near-duplicate clustering, and Bates-stamped redaction support document-review work beyond ordinary text analysis.
• Pro: Free academic access, training, project support, and group licenses make it unusually accessible to academic projects.
• Con: Professional is limited to one seat and 100,000 Units, while Enterprise starts at $3,000 a month.
• Con: Real-time Twitter collection is no longer supported, and Facebook open API access has been absent since 2014.
• Con: API access requires an Enterprise license and an issued key, and some documented API capabilities remain forthcoming.

Alternatives

Dovetail is worth considering for a lighter entry point: its free plan includes one channel, one project, basic chat, and AI summaries. Quirkos is a better match for readers seeking a student-oriented, lower-cost cloud option; its Student plan is $5 a month when paid annually and includes unlimited cloud storage and live collaboration. For a free tool with a desktop platform list, spaCy supports Windows, macOS, and Linux, while CATMA offers a free option across web, Windows, macOS, and Linux. Speakerbox is another free option for Linux, macOS, and Windows; KH Coder is freemium for Windows and macOS. Readers who prefer a paid product with a free trial can consider WordStat for Linux, macOS, and Windows or ATLAS.ti for macOS, web, and Windows.

Browse more options in Text Analytics Software or Social Media Analytics Software.

Verdict

Choose DiscoverText if your work needs machine classification and human annotation in one cloud workflow, especially when academic access or eDiscovery tools matter. Its mix of reusable sifters, review methods, and document handling is compelling, but the high Enterprise price, limited Professional seat count, and reduced social-data collection make it a poor choice for teams centered on social listening or inexpensive multi-user access.