Seeing Like an Algorithm — Remains of the Day: Why Good Advertising Works (Even When You Think It Doesn't)

Seeing Like an Algorithm — Remains of the Day: Why Good Advertising Works (Even When You Think It Doesn't)

In today's digital age, algorithms play a significant role in shaping our online experiences. Whether it's the recommendations we receive on platforms like TikTok or the targeted ads that follow us around the internet, algorithms have become an integral part of our daily lives. Understanding how these algorithms work and achieve their accuracy is crucial, not just for those interested in TikTok or the short video space, but for anyone in any industry who may come up against a competitor with a machine learning algorithm advantage.

When it comes to algorithms, many experts believe that TikTok hasn't made any groundbreaking advances in machine learning recommendations. Instead, the effectiveness of a machine learning algorithm depends on its training on a specific dataset. This is where the magic of TikTok lies - it creates a closed loop of feedback that inspires and enables the creation and viewing of videos, which in turn helps train its algorithm.

Designing an app that is algorithm-friendly is essential if you want to serve your users best. Every action and interaction within the app provides signals to the algorithm about the user's sentiment towards a particular video. Even before a video is sent to your phone, a human on TikTok's operations team has already watched it and added relevant tags or labels. The app's camera filters are designed to track human faces, hands, or gestures, invoking vision AI at the point of creation.

One of the common design elements in social networks today is the infinite vertically scrolling feed. Instead of showing users one story at a time, these apps display multiple items on the screen. However, this design choice poses a challenge for the algorithm to accurately judge the user's sentiment. By prioritizing lower friction scanning for users, social networks like Facebook, Twitter, and Instagram trade off a more accurate read on negative signals. This can lead to content that drifts away from a user's true interests due to the mismatch between their own preferences and those of their social graph.

Algorithm-friendly design doesn't have to be user-hostile. It simply requires a different approach to serve the user's interests effectively. The goal of any design should be to help the user achieve their desired outcome. While reducing friction is often aligned with this goal, it's not always the only factor to consider. Jobs-to-be-done framework can be a useful tool in understanding the user's needs and designing accordingly.

In today's software era, true competitive advantages are becoming increasingly illusory. Features and UI designs can be easily copied overnight by competitors. What truly sets companies apart is how every element of their design and processes connect with each other to create a dataset with which the algorithm trains itself to peak performance. This requires a deep understanding of how the flywheel works and a commitment to align every element and process with a single purpose and goal.

Now, let's shift our focus to the world of advertising. Many people believe that advertising doesn't "work" because they expect immediate action or persuasion. However, successful advertising rarely relies on argument or calls to action. Instead, it aims to create positive memories and feelings that influence our behavior over time, leading us to make a purchase at a later date.

The best advertisements are adept at leaving lasting impressions. They use images, jingles, and stories to capture our attention and focus it on the brand. These advertisements understand that they have little control over how audiences receive their message. Instead, they strive to create ideas, impressions, and positive feelings about the brand. Any memory that predisposes us to view the brand in a more positive light than its alternatives is considered a win.

It's important to recognize that advertising can influence our behavior, even if we don't realize it. Marketers understand how to engage and create memorable ads that slip past our defenses and leave a lasting impact on our memories. These memories then shape our future behavior, making us more likely to choose the advertised brand over its competitors.

In conclusion, both algorithm-friendly design and effective advertising share a common thread - the power of understanding human behavior and leveraging it to achieve desired outcomes. To apply these insights to your own endeavors, here are three actionable pieces of advice:

• 1. Prioritize training your algorithm: If you're developing an app or platform that relies on algorithms, ensure that you have a closed loop of feedback, enabling the algorithm to continuously learn and improve its recommendations.

• 2. Design for user interests: Create a user-centric design that takes into account the user's true preferences and aligns every element to serve their needs effectively. Remember, reducing friction is important, but it should not be the sole focus of your design.

• 3. Create memorable experiences: If you're involved in advertising, focus on creating engaging and memorable ads that leave lasting impressions on your audience. Use images, stories, and jingles that capture attention and evoke positive emotions, influencing behavior over time.

By incorporating these advice into your strategies, you can harness the power of algorithms and advertising to achieve your goals effectively. Remember, understanding how algorithms see and how advertising influences is the key to success in today's digital landscape.