Part One: Tomorrow’s Product Managers Need to Create the Conditions for Success With AI


When people talk about product management of the future, the first thing that comes to mind is artificial intelligence (AI). AI is changing the fundamental structure of every industry. We’re interacting with technology in new ways, from giving voice commands to virtual assistants to having Smart Reply suggest quick responses to our messages.

Does that mean it’s time to embrace the mathematical techniques that enable the building of intelligent software applications using the family of techniques known as deep learning? Absolutely! There’s no doubt that delegating parts of the software development process to machine learning models is going to be mandatory.

As we’ve become accustomed to having customer research and User Experience (UX) lead product vision, a new problem is stopping eager product managers from putting the process they were using, and applying them to AI problems. The reality is that we will need to evolve by finding ways to have a solid data set understanding.

In this two-part article series, we’ll dive into the importance of having solid data, model and problem understanding as a product manager. We’ll also look at how a product manager can create the conditions for success with AI through collaboration as well as considerations to keep in mind when building and shipping AI and machine learning projects.

The Current Product Management Process


Not all products are created equal. Where there is no deep learning, the process becomes sequential. Product managers gather real-world feedback and make decisions on what features to build. They lay out very clearly what they have learned about the actual customers, through qualitative or quantitative knowledge.

They state the additional problems they look forward to solving and propose solutions to grow the customer value while tying them to business results. They make a pitch and explain why it’s worthwhile for the company to invest in the solution. They then step back and trust the team to develop the prioritized features.

ML Systems are Trained on Existing Data Sets.


Machine learning (ML) is a complex space where results are achieved through frequent iterations. Through a large dataset of similar cases, the computers discover patterns and relationships in data instead of being manually programmed.

Where there is deep learning, the process becomes ‘everything is just a probability’. When faced with no exact numbers and definite results, product managers get a little clueless about what they should focus on.

Algorithmization (dependency on deep learning) requires strategic data acquisition from the onset. Most issues with machine learning are solved by understanding and preparing data better, and that’s the first responsibility of the product manager.

AI engineering teams benefit from interpretable and understandable solutions — so the product manager must assemble a small, diverse team to tackle the issue. They become involved or include data engineering and data science to extract meaning from and interpret the unique composition of the data set.

Deep learning forces product managers to focus more on the data to train the software. They relinquish much of their dependence on user experience and real-world feedback, adopting a trial-and-error iterative process meant to prevent skewed outputs.

For example, the Gmail platform predicts quick responses according to the message received. Focusing on conducting user research and tinkering with wireframes won’t help the team understand where the biggest improvements need to be made.

Creating the Conditions for Success With AI

It’s a product manager’s job to help an organization succeed with AI. In short, AI is a lifecycle that requires the integration of data, machine learning models, and the software around it. It covers everything from scoping and designing to building and testing all the way through to deployment — and eventually requires frequent monitoring.

With AI projects often becoming complex, it’s critical that everyone understands their roles. While it’s important that a product manager remains realistic about expectations, they also need to create the conditions for success and not over-promise. Engineers and data scientists, on the other hand, need to define and build the infrastructure that will allow an organization to learn how to use data effectively so as not to under-deliver. It’s essential that everyone on the team is able to collaborate cross-functionally — that means data engineers, data owners, data analysts, BI specialists, DevOps etc.

Here’s what it looks like if we break it down further:

Data scientist: It is the data scientist’s job to build accurate ML models. This is done laboratory-style where they might not know the requirements for the real-world usage and operation.

Engineer: It is the machine learning engineer’s job to scale AI processes to real-world use cases as well as iteratively improve the ML model and manage its post-deployment performance to ensure continuity of operations.

Product manager: The product manager must understand the consequences of putting an ML model out into the world. Not only do they need to have an intimate understanding of the data and be aware of any data quality issues and insufficiencies, but they also need to define the fall-back solution in the case of less-than-certain results.

With AI, you often don’t know what’s going to happen until you try it. Building and iterating an ML model to achieve a level of accuracy the business expects can be a daunting task — which is why it’s so important to set those expectations to see whether it makes sense to build an AI or ML project. Because even with good data training and clear objective metrics, it can be difficult to reach accuracy levels that are sufficient enough to satisfy end-users.

Building Your AI/ML Project


Feeling a little intimidated by the task? Don’t worry, your first AI/ML project doesn’t need to change the world. Product managers can start implementing AI by adopting a bottom-up approach.
Start small, choose a pain point to solve and collect data that represents the problem space. Next, collaborate with data scientists to understand the abstraction of that data set. And finally, ensure you have the right team (cross-functionally) and processes to solve the problem as well as a robust governance model to ensure the AI/ML models are monitored effectively. But remember, it’s all about collaboration.
Read on for more in the second part of this two-part article series, ‘Your Guide to Executing AI and Machine Learning Projects’.



Bain Public offers consistent roadmap planning processes & tools for business leaders and product managers organized around what motivates, inspires and improves growth. We offer a variety of articles, e-books and approaches designed to help organizations understand their digital strategies, introduce elements of roadmapping and establish product-led change amongst the senior leaders and managers. Our approach, product, expert advice and coaching help entangle complex technology, people and roadmap dynamics. For more information, visit bainpublic.com or call 514–442–8487.

Thanks to Loren O’Brien-Egesborg for contributing to this article as well as reading drafts and overseeing aspects of its publication. Also, if you have any feedback or criticism about this article, shoot us an email at info@bainpublic.com.