Enterprises have been successfully leveraging analytics platforms such as business intelligence (BI), predictive analytics based on machine learning (PAML), cognitive search, and others for multiple applications and use cases for decades. That has not been the case with text analytics or natural language processing (NLP) technologies. Forrester tracks over a hundred vendors whose platforms are mostly based on text mining and NLP, but most of them choose to go to market after narrow use cases such as customer or employee experience (CX and EX), social listening, document and file analytics, smart contracts, eDiscovery, and a plethora of other use cases.
Well, “the times they are a-changin’.” In our recent technology landscape — we call them Now Techs — we identified 30 text analytics/NLP vendors whose clients deploy the platforms for multiple use cases and applications. We bucketed them into three major categories:
- Text analytics platforms that primarily support people-focused applications. Vendors in this category typically concentrate on CX, EX, market research, brand management, and other applications for marketers, researchers, CX pros, and HR pros.
- Text analytics platforms that primarily support document-focused use cases. These vendors specialize in document classification and categorization, eDiscovery, contract analytics, and other document-focused use cases.
- General-purpose text analytics platforms that cover multiple use cases. Platforms in this category address a broad set of text analytics use cases — both people- and document-focused. They include functionality to ingest, integrate, cleanse, mine, enrich, and analyze text from the most popular unstructured and semistructured data sources and data types.
In our latest Forrester Wave™ evaluations, we evaluated general-purpose platforms but still separately, since people-focused and document-focused applications and use cases do have some unique requirements. For example:
- People-focused applications often use voice/speech as a source (as in contact-center call recordings), and since customers, prospects, and employees who answer surveys and post on social media typically do not spell-check or grammar-check, these platforms must have linguistic rules and ontologies to correct spelling and grammar errors before analysis.
- Document-focused applications — used mainly to ingest and process thousands or even millions of documents — must have strong document ingestion capabilities such as high speed and accurate optical character recognition (OCR) and forms processing (digitizing text and numbers found in forms like invoices, insurance claims forms, and others).
Our research identified seven vendors with strong capabilities to support both use cases, and we evaluated them in both reports. One vendor was only evaluated in the people-focused Wave and two only in the document-focused Wave, since even though under the covers these three vendors have technology that can address all use cases, the vendors mainly go after specific opportunities.
The key finding in our extensive evaluation research: Text analytics/NLP platforms have grown up. It’s time enterprises consider a single text analytics/NLP platform to address multiple use cases across many domain areas. Feel free to reach out with a client inquiry for further questions.