In her recent article for Harvard Business Review, “Why AI Customer Journeys Need More Friction,” MIT Sloan Senior Lecturer Renée Richardson Gosline pushes back on the conventional wisdom from marketers that a company should reduce or altogether remove friction (synonymous with “pain points”) to yield a competitive advantage. At many firms, artificial intelligence (AI) has become the go-to tool for creating frictionless experiences and removing impediments that slow down efficient customer journeys. Gosline argues that companies need to incorporate more “good friction” into their customer journeys, warning that frictionless strategies can lead to harm.
What is good friction?
Good friction is human-first. It engages users in decision-making during the conversion process and ultimately can lead to brand endorsement. When a company makes plain how a user’s data will be used, it establishes trust between the user and the company.
With good friction, users are given more opportunities to make decisions and understand the implication of those decisions. Creating communities where customers can connect with each other beyond the transactional touch point is another type of good friction that can offer companies more insight into the customer service experience, insight that standard metrics don’t necessarily capture.
What is bad friction?
Bad friction takes away a user’s autonomy and traps them in an impenetrable “black box” system, where users cannot decipher how their data is being used. A major signifier of bad friction in AI systems is when a user is given easy entrance into the conversion process, but many barriers to exit. As an example, Gosline cites WhatsApp’s Terms of Service fiasco, in which users who did not agree to the terms saw such an increase in friction on the app that it became all but impossible to use.
What you can do to add good friction
There are some steps companies can take to offset the risk of bad friction.
Practice acts of inconvenience: To preserve affirmative consent, companies should audit their models and ask whether AI should be used at all in the customer journey. Deliberately adding friction or kinks in the process, AKA “good friction” can increase transparency and promote human-first thinking.
Experiment to prevent auto-pilot applications of machine learning: Gosline argues that a culture of experimentation should be company-wide and not just reserved for data scientists. Executives should provide regular opportunities to test good friction and remove bad friction during the customer journey.
Look out for “dark patterns”: Companies should remove any part of the customer experience that resembles a “black box.” If it is easy for a customer to enter a contract or experience but hard to exit, they are being manipulated. Contracts that auto-renew and data sharing “agreements” that conceal privacy violations are examples of dark patterns. Removing them supports customer agency and is a critical part of gaining the customer’s trust and loyalty.
It’s an ethical imperative for companies to empower their customers with human-first technology. With a potential AI Bill of Rights looming, and companies like Google eradicating third-party cookies, now is the time to modify AI practices. Friction, when used thoughtfully, can enhance customers experience and secure a company’s reputation as being human-first.
You can read the complete HBR article here.
Renée Richardson Gosline is a senior lecturer and research scientist at the MIT Sloan School of Management. She teaches in the highly rated Executive Education course Breakthrough Customer Experience (CX) Strategy.