There has been tremendous progress in Algorithms, Data Science, Machine Learning (ML) and Artificial Intelligence. When there is enough data and the workflows are simple, these technologies deliver tangible benefit.
However many of the tasks performed by those knowledge workers in Sales, Marketing or Customer Success are Small Data problems, not Big Data problems, and the workflows are not that simple.
If an enormous quantity of homogenous data is not available, or if the related workflow is complex or ‘textured” (e.g. consulting or design services, any form of enterprise business development, or many of the other of the ‘thinking tasks’ carried out by knowledge workers) the machine on its own is unlikely to succeed. In these scenarios the right approach is a combination of human+machine, and Augmented Intelligence is a better answer than Artificial Intelligence.
When Google anticipates what you might be searching for, corrects the spelling of your search term, or suggests other questions you might want to ask, that’s a big data problem being solved and we are seeing machine learning at work. The machine has learned from what millions of other users before you have asked.
Let’s deconstruct this example: “How old is Stephen Tyler?” For the sake of the example I spelled Steven Tyler’s name incorrectly, using ‘Stephen’ instead of ‘Steven.’ What happened was:
- Google figured that out, because most people type ‘Steven Tyler,’ not ‘Stephen Tyler.’
- Then Google also decided that if I was interested in knowing his age, then I might be interested in the age of Mick Jagger and Axl Rose also because it has learned this is a common search path.
- Google also figured out that Liv Tyler is Steven Tyler’s daughter, so it displayed her age as well, just in case I cared.
- It also prompted me with other questions that ‘People also ask.’
There’s a lot going on here—Google has a generic ML algorithm that is answering the question: “When someone types something into the search box, what do they really mean, and what else might they be interested in?”
Like the other Internet giants, Amazon and Facebook, Google has one very significant advantage over the rest of us. It is dealing with enormous quantities of data. Approximately 5 billion searches every day. That’s a lot of data. Also the workflows involved in Google search, or Amazon’s recommendation engine are simple, though sophisticated in their application.
Artificial Intelligence is very good at driving efficiencies but less so in driving effectiveness in scenarios where the workflows are complex. That’s where Augmented Intelligence takes over. In my book Tomorrow | Today: How AI Impact How We Work, Live and Think, I map Competitive Advantage and Scope of Impact against Application of Knowledge in the context of the application of AI.
As Tomasz Tunguz from Redpoint said in a recent blog post entitled When Machine Learning Just Isn’t Enough:
The challenge is always the same. How can software improve a current workflow to such an extent that a user is willing to stop their current workflow and learn a new one? …… It’s the workflow that keeps the users coming back. Tomasz Tunguz, Redpoint
This is particularly true in the context of tasks that knowledge workers, like those in sales, marketing and customer success teams in enterprise B2B face everyday. The success of a new technology is its ability to co-opt human behavior. As humans in business are trained by their experiences in consumer-land – instant access to information, always on answers, micro-moment interactions and too much choice – we are becoming increasingly impatient, have no tolerance of poor service, are frustrated and unforgiving of errors made by machine algorithms, and constantly interrupted by emails, texts, tweets, snaps, likes and shares. The threshold for meaningful application of AI, in whichever mode, Artificial or Augmented, is high. I am betting on Augmented Intelligence that has awareness of the context in which the workflow happens and responds accordingly.
For those of us who are focused on providing or using applications that support knowledge workers – like the exciting work going at Altify – that’s the opportunity is to leverage technologies like ML but to do so with an approach that heavily supports the technology with human-infused knowledge to augment the intelligence in the software. Operating in particular business niches, as opposed to consumer domains, we can differentiate through applications that use unique training data specific to those niches, and enable business knowledge experts to pour their knowledge of the domain and the associated workflows into the application. This approach compensates for the absence of vast quantities of training data and enables much more accurate outcomes for knowledge-intensive applications and blends the best of human+machine.
If you want to read more of my blogs please subscribe. Follow me on Twitter or connect on LinkedIn. I want you to agree or disagree with me, but most of all: I want you to bring passion to the conversation.
Donal Daly is Executive Chairman of Altify having founded the company in 2005. He is author of numerous books and ebooks including the Amazon #1 Best-sellers Account Planning in Salesforce and Tomorrow | Today: How AI Impacts How We Work, Live, and Think. Altify is Donal’s fifth global business enterprise.