Posted on by Vishnu Devarajan
Subscribe to our Telegram channel for the latest updates in the marketing and advertising scene
The Rise of Smart Machines
11 Jul
by Indhran Indhraseghar
The world of AI has changed entirely since I was CMO at a Houston based AI solutions provider – half a decade before ChatGPT exploded on the scene. Back in those days, Fortune 1000 corporations were the only ones who could afford` us.
Today the potential of AI is available to every business on the planet. Tools like ChatGPT are reshaping how the we write, create, and code.
AI offers the promise of productivity and performance at scale allowing SME (for the first time ever) to dramatically increase the quality and volume of their output. It also opens up levels of prediction and personalization that were previously impossible.
How did this suddenly come about?
Intelligence. By orders of magnitude.
Demis Hassabis, CEO of DeepMind defines AI as “the science of making machines smart”.
If you think about it, traditional software is dumb.
For a typical campaign, you have to create all the marketing assets, specify lists, workflows, and logic, and determine the strategies used to achieve marketing objectives. At no point does the software know how to make dynamic changes based on campaign performance. At no point does it automatically adjust direction in response to data. And it can’t make any real strategic recommendations on how to improve your results.
AI is different. According to the Marketing AI Institute:
AI can determine its own rules and instructions without human oversight in order to achieve your goals.
AI can plot its own pathways to achieve the goals that you give it—pathways that humans may not be capable of seeing.
And AI can learn and improve on its own, improving performance without a human explicitly telling it how to do so.
Let’s say you are sending an email to 10,000 subscribers and you want to make sure it has the highest open rate possible by optimizing the send time.
With traditional software, you have to:
Evaluate past emails in your system to determine which send times performed best when sending similar emails.
Research best practices on the best times to send emails based on available research and commentary online.
Look at your audience breakdown to determine how different time zones might impact send time—or use a feature in your email software to send your email at the same time in different time zones.
When all that’s done, you make an educated guess on the best time to send, then schedule the email.
This approach is unscientific and exhausting.
With AI, your process might look something like this:
Use an AI tool that automatically analyses subscriber email open behaviour and sends your emails to each of your 10,000 subscribers at the time they’re most likely to open.
This is just one of many examples of using AI at scale to achieve goals more efficiently and effectively in ways that humans can’t.
