How machine learning frees creativity and strategy for vendors

We are excited to bring Transform 2022 back in person July 19th and approximately July 20th-28th. Join AI and data leaders for clear conversations and exciting networking opportunities. Register today!

Artificial intelligence (AI) and machine learning (ML) have become widespread over the years. These days, it seems that every company is an AI / ML company – and the reality is that American researcher, scientist and futurist Roy Amara stated, “We tend to overestimate the impact of technology in the short term. assess and underestimate its impact in the future. long-term. “

When new technology is developed or deployed, people often talk about the fact that everything will suddenly change in the next two years. However, we also tend to completely ignore its impact, especially if it is a type of technology that can radically change the way sellers solve problems and communicate with customers. If we take full advantage of AI and ML, it is important to first understand the technology and distinguish between the facts and myths of today how it works. Only then can we understand what is real, how this technology can be flexible, and how machine learning and AI can liberate creativity and strategic thinking for marketers.

Mechanical engineering begins with data

Without the ability to analyze data, identify patterns, and use them, data is effectively useless. Machines are brutal optimizers that can organize data to a level where duplication of people is impossible. However, this also does the opposite, because machines today cannot replicate the thinking and creative strategies that people can generate and act on. The information, optimized through machine learning with machine learning, gives vendors a powerful ability to make smart decisions and then have a creative strategy to achieve the desired result.

Car training for vendors: Asking the right questions

What matters to companies and individuals are decisions and actions. When I consulted with large companies that spend millions or tens of millions on “data strategies” or undefined areas, I was often advised to worry about the data they needed to collect beforehand. what decisions they should start with. and the actions they should take as a business. From this perspective, entrepreneurs may ask themselves: What decisions do you want to make more wisely and quickly? Are you structured as an organization to make these decisions? Once they have been identified, you can ask questions such as what information do I need to make these decisions quickly and wisely? And which of these decisions can be automated?

So where does machine learning come from? In what category of problems can it help us? To answer these questions, it is useful to first understand the limitations of this technology. ML does not replicate the generality and adaptability of the human mind – instead (and continuously with other technologies) it enhances the human mind and solves a specific set of problems with extraordinary human ability. To determine if ML applies to a problem, the following set of questions is helpful:

  • Can a person solve a specific task in less than 2 seconds? (This is an estimate; we have not yet reached the point of resolving more complex issues.)
  • Is it worth solving this problem on a scale (e.g., billions of times incredibly fast)?
  • Is it worth doing this task repeatedly, steadily, and consistently?
  • Can we measure “success” by numbers?

If you can answer “yes” to these questions, then you have a problem that is very relevant to the application of mechanical engineering. (Interestingly, these are also some of the tasks that people are terrified of doing because we get bored, distracted, and tired!) This may seem very limited, but many problems fit into a “yes” bucket. , such as spam message detection, fraud detection, pricing optimization and language sensitivity.

Solve vendors ’problems with car training

When it comes to marketing and advertising, there is a whole category of problems that they also fit into the “yes” bucket. Determining audience content and behavior over time, predicting whether an ad will lead a potential customer to visit my site, based on the content of their article, and setting thousands of parameters for efficient and effective budgeting are all such marketing. the problem.

There are also problems that do not fit into this classification, for example: how do I convey my complex message in a way that cuts the volume? How can I effectively connect with an audience that I don’t resonate with right now? How do I balance long-term and short-term goals?

Mechanical engineering is not magic: it can give vendors incredible abilities to deepen our understanding, optimize transmission against well-defined goals, respond quickly and rationally to change, and make our ideas more unpredictable, slower, and more audible. .

Communicate with customers in real time

For marketing, a lot of useful information and examples relate to customer behavior. Digital campaigns are significantly less effective when they are currently unable to respond to changing conditions. For example, if you sell coffee beans, you want to appeal to people who are still interested in buying coffee, rather than those who searched the internet last week and bought it yesterday. Everyone has experienced an online purchase for a product that they get, and once they have each device and platform, they will spam them with the same product next week. Although this can be useful for products that customers typically buy (laundry detergent, toiletries, etc.), most people only need one coffee maker.

Not only does real-time data ensure that campaigns reach the right people, but it also allows sellers to respond to changing market conditions. By combining machine learning with real-time data, vendors can see results directly, instead of waiting for results at the end of the campaign. This means that brands can detect and invest in things like the popular show, the recently released Netflix or anything that tends to be on Twitter, or even address the rapidly changing dynamics in the supply chain. If brands have learned anything in the last two years, it is that global events can have an impact on behavior and shopping patterns at the same time.

While cars can take care of analyzing data around demographics, web browsing behavior and previous purchases, having the right creative vendor – who can link current trends to campaign goals and ask the right questions from cars – is what drives cars. well differentiated. campaign than a great campaign. To get another great quote, this time from Alan Kay, “Simple things should be simple, complex things should be possible”. In addition to helping us gain a deeper understanding and understanding of audience behavior, high technology should also make it easier for marketers to respond to this information by gaining new creative ideas in minutes, not months.

Can ML predict the future?

Predicting the future is impossible. But machine learning technology, combined with real-time data, can allow vendors to understand emerging trends and changes in behavior when they occur and respond to these changes by providing automated campaigns in minutes and see if they last longer. hours and days work or not. . Real progress is about learning and about testing strategies and ideas.

The small impact that ML will have on the advertising technology industry over the next decade does not depend on the ideas generated by the AI ​​or the reduction in dollars spent on the operations that are carried out; The big impact is of shortening the gaps between marketing strategy, understanding, idea and implementation and allowing us to understand deeper and faster, be more creative and test ideas more reliably and easily and measure impact more effectively. This technology, like all other technologies, is not meant to replace people, but to free us from repetition and boredom and give us the power to be more human.

Peter Day is the CTO Quantcast


Welcome to the VentureBeat community!

DataDecisionMakers is where experts, including technical people who work with data, can share insights and innovations related to data.

If you would like to learn about cutting-edge and up-to-date ideas, best practices and future data and data technologies, join us at DataDecisionMakers.

You might even consider submitting your own article!

Read more from DataDecisionMakers

Leave a Comment