10 Practical Tips for the Successful Adoption of Your Machine Learning Products
Hands-on tips for companies to build Machine Learning Products that are being adopted by their users and customers.
The biggest difficulty for products based on machine learning (ML) will be user or customer adoption.
How did I come to this conclusion?
A top executive of one of the biggest European insurance companies told me: “We have the money and technical talent to build sophisticated ML-products, but we do not know how to make users adopt those products. We spent millions of dollars on an ML-based app but only got around 300 users. We do not understand why people do not want to use our app.”
Similar views were echoed by one of the top executives of an Indian software house with multi-billion dollar revenues, who told me: “We are having problems with getting our internet of things (IoT)/ML products adopted by users and customers.”
Why are these companies not successful?
From hype to correction phase
Over the past few years, ML and Artificial Intelligence (AI) has received plenty of hype as can be seen in the Gartner’s Hype Cycle below.
Many startups were able to launch products to the early adopter market where users are more willing to try out products, even though they might be overhyped.
What is happening now, is that ML/AI is moving from the hype phase to the correction phase. In this phase, the market moves from early adopters to the majority and the majority adopts the technology only if they see a clear tangible value. Now, the majority segment is also where the customers and users of corporations belong.
The question is how can ML products be adopted by the majority?
Typically, corporations think top-down. They build products and expect their customers to use them. Unfortunately, disruptive technology adoption cannot work that way. In addition, due to heightened competition and tech giants like Google and Amazon entering various kinds of business segments, corporations need to rethink how they build and launch disruptive products.
10 tips for corporations to build ML-based products for adoption
1. Provide incentives for users to share their data.
In the past, many corporations used fine prints in user agreements to collect user data. The new General Data Protection Regulation (GDPR) does not allow corporations to do that anymore, at least not in Europe. Corporations need to create incentives for people to share their data by providing them with a better experience or a way to save their money.
Gamification to incentivize users to share their data
When I was part of the team that built Road Skippers, we gamified the system with incentives like free movie tickets and coffee and people were very happy to share their data. This was in contrast to the approach taken by many insurance companies, who did not provide enough information and basically asked users to put black boxes in their cars. Obviously, that did not work.
2. Show the generated value to the user for their data.
In 2010, I was working on a project for CEZ around smart meter adoption.
We were trying to answer the key question of how to make users adopt smart meters. Through studying various smart metering projects around the world and speaking with different stakeholders, we came to the conclusion that providing an interface showing real-time feedback results in higher energy savings, making the value creation for the user very clear.
3. Let the user be in control
The machine should be seen as an aid to help humans and not as a mean to replace them. As human beings, we want to feel that we are always in control. With trust being a big issue in the current phase of AI products, users want to feel that they are in control. Therefore, build tools that help users to make decisions easier or take away tedious tasks.
A bot that helps with decision making
When working with an AI bot startup we made sure that the communication flow of the bot is designed in a way that the questions help the user to make decisions. In this way, the user always feels in control as the decision maker.
4. Do not try to change behaviors.
If you are building a tool that requires users to fundamentally change their behavior to use it, chances of failure are high.
I have learned this the hard way when building a startup that required people to change the way they do certain things around travel and this did not work!
People tend to do things the way they are used to doing them. Changing behavior is a long and difficult process, which requires persistence.
Corporations do not have that kind of mentality, but startups do. Do not go into the business of changing behavior — leave that for the startups!
5. Do not overcomplicate the architecture when it comes to product engineering
So many times have I seen corporations and startups overcomplicate their product architecture. There is a tendency to choose the ‘fancy’ technology, even if a simpler technology might do the job.
When talking to an executive at an IoT company in the UK that is building an ML-based prediction engine, they were thinking of using a data warehouse. Now, a data warehouse is a system for sophisticated data analytics and business intelligence. Though, in most cases, tools like Mixplanel or Tableau are enough to meet a company’s data analytics needs. I could see that they did not need a data warehouse, but the person in charge was very eager to use his superior skills in the data warehouse space, so he was pushing for that.
Below is a simplified version of the architecture I devised for an IoT predictive analytics company:
A complicated architecture makes the product development and iteration cycles way longer and will affect the adoption cycle and this is a problem as often during the adoption phase, you need a fast turnaround.
6. Be sure to choose the right database or databases
The database has a big effect on every aspect of your product, including the user experience. ML-based products use ‘big data’ and the natural tendency is to select a No-SQL database. However, there are problems with No-SQL databases. Most of our non-machine generated data, as well as legacy data, is relational in nature and best suited for a SQL database.
When we were building Road Skippers, we ended up using two databases — one SQL and another No SQL.
A non-efficient database structure can make the query time long and thus creates a bad user experience. Because of ML-products (and IoT), there has been a sudden increase in time-series databases (see below). A time-series database is a database optimally suited to store data that is generated over a period of time and many ML-products use such time-series data. Last year, I was co-operating with a SQL-based time-series database, and I found this solved the problem of multiple databases in a very neat way.
7. Try to use the simplest best-fit ML-algorithm
Machine learning models can be quite complicated. The training can take a lot of time and if the data is not clean or good, efforts will go waste. It is better to start with the simplest best-fit model. In most cases, I have seen a simpler Neural Network like Word2vec does the job quite well. Neural networks doing classification or clustering can be good enough to solve many problems. There are a few cases I have come across where more sophisticated networks like long short-term memory (LTSM) will be needed. The figure below shows an LSTM cell. One of the cases for LSTM is here.
And here is a simple classification neural network, which can be used to solve many real-world problems:
8. Educate your customers or users during your product launch
Recently I started reading a book named The Challenger Sale. The authors studied over 6,000 salespeople around the world and classified them into five categories. They saw that the salespeople they referred to as ‘challenger salespeople’ outperformed every other group. This category of people frequently challenges the norm, is more knowledgeable and educates their customers. Subsequently, the customers trust them and ultimately buy from them.
9. Ensure consistent metrics across marketing, sales, and products
I have seen companies using a different metric for user acquisition channels, marketing, and sales. Sometimes those metrics can conflict and your product development team may be optimizing a metric, which does not reflect an increase in adoption.
10. Be sure to identify enthusiasts of your products
Getting back to the technology-adoption life cycle from the beginning, reaching the critical mass starts with early majority users. Identify them well in advance and communicate with them for example through starting a Facebook group.
One more tip: Do not be complacent
Last year, during a panel discussion one of the bank executives claimed: “We are not worried about AI. We will be able to easily adapt.” My reply was: “That is the exact reason why big corporations will be disrupted by tech giants.”
If you want to receive updates on our AI Challenges, get expert interviews, and practical tips to boost your AI skills, subscribe to our monthly newsletter.
We are also on Facebook, LinkedIn, and Twitter.