Apple, Intel, & the Elephant in the Room

  2 m, 30 s

Our friends over at Bottlenose built their own sentiment analysis engine before switching to a Salience integration. Their case got the Lexalytics team talking about a question many companies face when adding text analytics to their products: Build or Buy? In light of this discussion, I’d like to share a story that sheds some light on the topic.

The other day I read an article about Apple’s switch from PowerPC computer processors to Intel’s x86 processing architecture. The short version goes like this: In 2005 (yes I know, I’m late to the table), Steve Jobs announced that although Apple could continue to use PowerPC and make great products, they could make even better products by relying on Intel processors. Many analysts took this to mean that Apple was concerned about PowerPC’s ability to deliver competitive processors for ultrathin laptops, due to heat issues and power requirements.

A decade later, we all know how that decision has turned out for Apple (quite well). But there’s a question here that rarely gets asked or answered: 

Why doesn’t Apple just build their own processors?

The answer is simple. Apple is very good at designing sexy computers, operating systems, and applications. But Apple is not so good at building the processor chips that power those computers. Creating their own chips would require an enormous investment of time, money, and labor, with no guarantee of returning the results they wanted. And so Apple turns to outside companies to provide the CPUs for their laptops, desktops, and other devices. Through this partnership, Apple is free to focus on what they do best: designing those gorgeous computers and operating systems.

Likewise, your organization is very good at building powerful applications that create value for your customers. But let’s directly address the elephant in the room: you’re not so skilled at developing text analytics engines. When you add text analytics to your products, you’ll have to answer that classic “Build vs Buy” question.

For example, let’s say you provide an attractive dashboard for Customer Experience Management (CEM), and you want to let your customers analyze tweets and other social content. Will you:

  • spend thousands of dollars and man-hours developing your own buggy, minimum-viable text analytics system?
  • use a professional text analytics engine as a base, then build your awesome dashboard on top of us, knowing that we’ve already taken care of the technical bit?

Researching and developing a text analytics engine is a costly endeavor. Lexalytics has spent over a decade refining and expanding our systems to provide fast, reliable insights into textual content of all types, in more than 20 languages.

When it’s your turn to Build or Buy, think about how the most profitable company in the world chose to focus on what they do best. What will you choose?

Hungry for more information? Check out our in-depth technical white paper, “Build vs Buy for Text Mining,” then contact us to chat with a Lexalytics wizard.

Categories: Insights