As some of you may know, B Squared is hosting the first annual HELLO Conference on March 28, 2019.
We’re super excited about this.
We’re excited not just because we’re hosting our first ever B Squared event, but because we have an incredible line-up with seasoned pros like Christopher Penn of Trust Insights.
Chris was kind enough to sit down for a session of B2TV with us ahead of the HELLO Conference where he explained Artificial Intelligence (AI) and Machine Learning (ML) in a way we can all understand. Yes, it’s possible!
Are the robots coming to take our jobs? Can AI really help us scale?
Chris, the go-to guy for machine learning and marketing analytics, answered these questions for us and gave us a sneak peek at what he’ll be discussing at the conference.
If you’re interested in learning more about AI, but are overwhelmed by the concept, check out our interview with Chris.
B Squared: We’re here with Christopher Penn, co-founder of Trust Insights, a data and analytics consulting company. They help people like you and me – marketers – understand their data analytics better as well as get better results through that data.
We’re going to start off with some softball-sized questions. Then we’ll try to get into some of the scarier stuff as we go along.
First and foremost, some people, even marketers, when they hear “Artificial Intelligence” or “AI” they still don’t know what it means. Can you explain in layman’s terms what you describe as AI?
Chris Penn: The formal definition of artificial intelligence is getting the software, getting computers, to be able to perform tasks which typically require human intelligence. So if you can understand the sounds coming out of my mouth, the words that I’m saying, that’s called natural language processing.
We have to teach machines explicitly how to process language and vision. It’s getting machines to emulate human intelligence capabilities. AI is a super broad term and it can mean anything that fits into this category of trying to get a machine to do what a human can do.
Everything from robotics to process management to speech to composing music – all of these things would fit under the umbrella of AI.
Pay Attention To Machine Learning
Chris Penn: Where marketers are, and should, be paying attention is in a subset of AI. This subset is called Machine Learning (ML) which is when we teach machines to learn.
With traditional software like your word processor, or an app on your phone, humans write the code for these things and the machine spits out data. It could be silly awards in Candy Crush, it could be word processing documents, it could be a video from your phone. All of these things are data that the app produces.
ML is the opposite.
You feed a whole bunch of data to a machine learning software and it writes its own software. It designs software for itself that creates a certain type of outcome.
Very broadly speaking, there are two categories of Machine Learning, there’s supervised and unsupervised.
Supervised means “I want to know something, I want you to explain a result better.” For example, let’s say you’re doing social media analytics. You have the customer’s revenue numbers and you have all your social data. You’re going to use supervised learning to say, “Okay, I want to know what data points, what factors, what dimensions lead to or have a relationship to that revenue number.” You would use supervised learning techniques to find that.
The second type is unsupervised learning where you don’t know the outcome, you don’t know what’s happening.
A really good example of this is social conversation monitoring. When you choose a hashtag and want to know what people are saying around it, you need to download tens of hundreds of thousands of conversations. The machine kind of arranges all of the results for you. It says, “Hey, here’s a group that seems to cluster together, and here’s another group that seems to cluster together.” This group is all talking about one thing, and this other group is talking about something else. In this case, you didn’t know that was going to be the outcome, but you needed the machine to sort it out for you.
So, very broadly, those are the two big categories – supervised and unsupervised.
How AI Helps Marketers
Chris Penn: When we’re talking about AI for marketers, we’re talking about marketing trying to do three things:
- Faster: AI should help us do our work faster. There’s no way any of us could read 1,000 tweets a second. Acceleration.
- Better: Accuracy. Better quality data. Humans are prone to mistakes and if you’re living your life inside of spreadsheets, there’s a good chance that you or someone in your chain of command is introducing errors along the way. We can improve the accuracy by taking that away from humans and giving it to machines to do the same repetitive stuff. This will give us better accuracy.
- Cheaper: Or more efficient. This means getting stuff that is repetitive and of low value away from humans and over to machine so that you the human can do more valuable things. The person will be happier. The work will be better.
These are the three things that marketers should be looking for when embarking on using machine learning to make life better.
B2: There are so many articles that we’ve been seeing that say, “AI’s going to take your job” or “the robots are coming.” It’s been made into a scary thing. In the three examples you just gave, it’s not completely removing the human. We’re not removing the human from the intelligence tools, we’re helping the human being make better decisions faster and taking out some of those low-level tasks to make things cheaper and more efficient.
CP: Maybe. And I say maybe because there was a fascinating piece from a reporter who was at Davos last week. They were privy to and talking to a number of CEOs and major movers and shakers, and the behind the scenes conversation for corporations is “We want AI to remove as many jobs as possible because humans are expensive”.
This one representative from a major bank said, “Our goal is to take thousands of jobs and turn them into dozens because we will have much better results and our margins and our quarterly numbers for the stock market will be so much better.” A lot of CEOs at that event see AI as the “golden ticket” to hitting their quarterly numbers and shedding a lot of headcount.
One of the things that’s important for all of us to understand is, there are a number of things you should be doing in your career to make your job more resistant to being taken by a machine.
The more repetitive your job is, the more repetitive every task is, the easier it is to automate away. The more creative or collaborative or broad in scope your job is, the more difficult it is to automate. You could automate pieces of it away, but you can’t automate the whole thing away. Machines are really bad at things like general life experience. They’re really bad at empathy, they’re really bad at broad judgement. Things that make you a great human are things that you will need to double down on. Machine algorithms have a very difficult time crossing disciplines.
If you’re a Twitter ads manager, that’s a super easy task to automate. It’s almost automated as it is. But if you’re a Twitter ads manager who is also good at data visualization and project management, you’re super hard to replace because you have multiple things you bring to the table. So the more we, as people, can build up our skill base across disciplines and network within our companies, that makes it harder for you to be replaced.
B2: So it doesn’t have to be scary if you’re smart about it. This is just my viewpoint, but as someone who owns a company, I love that answer because it means the people who should be on the team will be making those smart decisions on how they can remain on the team and the people who don’t want to go the extra mile are probably going to weed themselves out.
CP: Right, exactly. And at the very least what you want to do is you want to be able to, even if you never change the amount of headcount you have, you want to use machine learning to scale.
For example, the amount of time people spend on curating content can be hours per week. At Trust Insights, one of the things that we do is we built our own software to gather data, score it, analyze it and then, pre-curate it. Then it just goes through one final, passive, set of human eyes to go “okay even though that was a popular article, its on topic, it’s still not something we feel comfortable publishing”, but instead of spending 10 or 15 hours a week curating content for our individual social accounts and for the company account, it’s five to 10 minutes a week. It’s a great time saver, delivers quality, and frees us up to be able to do more valuable work.
B2: Yes. And that is the case in point why social media marketers should be looking at AI. There are a lot of things that we have to touch but there are a lot of tasks we do that could be replaced by machines – not removing the human completely because we still want a human, creative spin, but yes, I can see it already.
Now without giving away the kitchen sink can you dive into a little bit about what you’re going to be talking about at the HELLO Conference?
The Four Practical Applications Of AI
CP: We’re going to look at practical applications of AI specific to social & digital marketers.
- Mining
- Clustering
- Forecasting
- Driving
For the full interview with Chris and to learn more about his practical applications of AI for marketers, check out our B2TV session!
Have you dipped your feet into the AI pool yet? Let us know in the comments section below!
Latest posts by Brooke B. Sellas (see all)
- When to Let Go: Ensuring You Are Focusing On the Best Customers - November 29, 2023
- Optimizing Your ABM Strategy with Social Listening for Enhanced Results - November 15, 2023
- How To (Better) Appreciate Your Customer Experience Professionals - November 8, 2023
