As part of the upcoming IBM Win with AI Summit, I'm tackling some questions about how companies can make AI work for them. In today's episode, we look at the journey to AI using the Trust Insights Enterprise AI Journey Framework:
- Data foundation
- Measurement & analytics
- Insights & research
- Process automation
- Data science
- Machine learning
- Fully deployed-AI
FTC Disclosure: I am compensated by IBM to attend and create content for the summit.
Can't see anything? Watch it on YouTube here.
Listen to the audio here:
- Got a question for You Ask, I'll Answer? Submit it here!
- Subscribe to my weekly newsletter for more useful marketing tips.
- Find older episodes of You Ask, I Answer on my YouTube channel.
- Need help with your company's data and analytics? Let me know!
What follows is an AI-generated transcript. The transcript may contain errors and is not a substitute for watching the video.
In today's episode Hi. I'm part of the upcoming win with AI summit that IBM associate in New York City full disclosure, I am being compensated to attend and talk about the event.
Well, the questions leading up to the event is what two companies need to do in order to win with AI and wanted to share some thoughts about the the enterprise AI journey? How does a company get to using AI? And what are the precursors of the prerequisites in order to do this because it's not something that at a production level and especially at an enterprise level for bigger companies, you can just kind of do you absolutely can dabble into individual practitioners can certainly experiment with different tools and technologies. But how do you roll this out in a a well thought out
way you need to follow the journey, there's a there is a defined path of getting to artificial intelligence is
intelligent use within the company. And it is a seven step journey. And so let's actually bring this up here to here
in the beginning, before you can ever talk about AI, you need to have a solid data Foundation, you need to have all your data you need to find, clean it, prepare it, unify it, make them have the ability to process it in an intelligent way across the enterprise in all the different functions that you eventually want to benefit from AI. Remember that AI delivers the three A's as its benefits, acceleration, faster results, accuracy, better results and alleviation, reduction of stuff that you don't want people doing anymore because it's not a good use of their time. In order to get to those benefits. You first need that data foundation. Where's the data? Is it in good condition? Can we even use it
can we can we plug into it? The second level. The second iteration of this journey is around measurement. And analytics companies need to become data driven. And what that means is you make decisions using data. The analogy I use often is, every time you get in the car and turn on your GPS app, or your Maps app on your smartphone, you are literally becoming data driven. You decide where you want to go, and then the machine helps you get there. And so for companies they need to identify and then measure all the KPIs that they care about and understand what happened be able to explain for any KPI what happened in a in a relevant way. And this applies to every function in the company. This is not just marketing. Obviously, my background is marketing. So I know that best but sales service finance operations, HR across the board if you don't have a measurement function in place, and the ability to do analytics to explain away
What happened in great detail to for any function that's the next step in the journey. The third step in the journey is insights and research. And this is where you develop your qualitative capabilities. And a lot of companies skip over this part. And they really shouldn't, because nothing in the numbers themselves ever really does a great job of telling you why something happened. Yes, website traffic because up 40% yesterday, why
fewer people applied for our open jobs in our New York City Office. Why we don't know that unless you have qualitative capabilities. And you need that insights and research function to help explain your data story to tell your data story better to do market research, to do surveying to do Public Opinion Research and to take all the different data sources that are available and turn them into good explanations that you can then use to inform your strategy and tactics. The fourth step in this journey and this is this is one that can be concurrent once you've got your
Measurement analytics in place is process automation. Once you know the KPIs, it is time to start rolling out improvements. And you don't necessarily need machine learning or AI to be able to just do stuff better to be more efficient. This can be everything from the implementation, for example of Agile in your development function to actual robotic process automation, where you're teaching computers to do very mundane things. Okay, this person at this desk copies and pastes the same spreadsheet every single week. Let's get some code in place to do that. And companies have been doing Process Automation ever since computers entered the workplace. But this is something you won't have as a formal discipline that is part of the foundation of AI. If you think about it, go back a couple episodes to our discussion on the known knowns, the unknown unknowns and so on and so forth. Process Automation is all about taking advantage of the known unknowns, you know your data you know it's in your data and
You know, what's in your KPIs, and you can start making improvements on all of the things that you already know our issues. The fifth step in the journey is building a data science capability. and machine learning and data science are not the same thing. They are related. And one of the predicates, I think of machine learning is good data science capabilities. This is where you start exploring the unknowns where you start building statistical and mathematical capabilities, building models,
trying to explain things in the data and writing code and building engineering and it functions around around your data to help at scale up to help get you to answers better answers, faster answers,
but still very much human informed is still a ton of human judgment in data science to even understand what to experiment with. The data science function is very much about the scientific method. So all the things when if you go back to
The earliest steps in the journey. If you've got a KPI and you're not sure
how to get to metrics that drive the KPI, that's where a data science function will come into health. explore those unknowns and to answer those questions within your data. The six step in the journey is really rolling out your machine learning capabilities. This is supervised learning, unsupervised learning, reinforcement learning, maybe even some deep learning, you're advancing your process automation, you're starting to really deliver on those three A's, you're getting the answers faster, you're getting better answers and you are using machine learning to to drastically improve your process automation and your data science capabilities. The machines are doing more and more of that work, letting the people who are in network direct it instead of do it. And then the last step of course, is rolling out in production AI across the enterprise in every relevant role, every function that can have machines delivering better faster results in
Reducing, mundane, boring,
unexciting tasks that's that's the end state you're looking for is your human talent is focused on the things humans are best at judgment, empathy, relationship, building, human to human contact, grand strategy, multidisciplinary stuff, that, that AI machine learning simply isn't isn't able to do yet. So when it comes to how can a company make the leap to a high
there isn't a leap is a journey. It's a path from data foundation all the way through that enterprise rollout. And you can't really skip steps. You can selectively maybe leap frog little places here. They're at a very tactical level. But at a strategic organizational level. There's there's no shortcuts. There is the implementation of great technology there is building the talents that you have setting up processes adapting your culture as an organization and of course
course all of the the technology that's needed. One of the critical things people do wrong this is believe at AI is a technology implementation. It is only one third technology. It is also all of that that people that process that culture that that organizational stuff that goes with the technology in order to make it successful.
Despite what you see in the movies and the news AI isn't that have taken everyone's job,
certainly not in the in the foreseeable future. The technology is just not there yet, but it absolutely can make your job easier if you deploy it correctly. And if you follow the journey from data foundation through analytics insights, process automation, data science and machine learning to that full AI future. We're going to have more in this series coming up. So stay tuned. As always, please subscribe to the YouTube channel on the newsletter. I'll talk to you soon. Take care.
You might also enjoy:
- You Ask, I Answer: Microsoft Clarity vs. Google Analytics?
- How To Start Your Public Speaking Career
- How To Set Your Consulting Billing Rates and Fees
- Almost Timely: The 2020 Essays
- How to Set Your Public Speaking Fee
Want to read more like this from Christopher Penn? Get updates here:
Get your copy of AI For Marketers