Tag Archives: machine learning

The Only Blog Post You Truly Need to Read About Tech Trends in 2018

7c050ba1-0833-44cb-acff-27f93f7eb28b 2

JWT Mirum APAC SXSW Delegate

As many of you might know I was 1,5 weeks in Austin, Texas in SXSW. I attended probably too much of talks and been quite amped up even after. During the seminar I was part of JWT Mirum APAC delegate and I blogged all the way through from the conference (including some videos). Below I have compiled majority of lessons I got from the seminar.

Besides this I saw my idols Arnold Schwarzenegger, U-God (from Wu-Tang Clan), Bushwick Bill (from Geto Boys) and saw the reunion of one of the most progressive hiphop acts of all time Dr. Octagon. Great trip indeed. Below you can read all my observations and insights from my trip:

Summary

Three Main Lessons from SXSW

All About The Data

Data is The New Oil, but also the Oil Spill


How to make your Data actionable?

How to Lie with Data?

From Big Data to Smart Data: How Blockchain is Enabling Both Convenience and Security of Your Own Data

Don´t Take It for Face Value: Facial Recognition as the New Data Point

AI Requires us to Be More Human

“We have not even reach the limit of human intelligence”

From e-Commerce to V-Commerce

Disruptive Innovation

Cannabis is Leapfrogging Technological Innovation

How to Innovate in Atom-based Industries?

Necrotech: What will Be Your Digital Legacy When You Die?

New Model of Content Creation

How to Live More Fulfilling Life?

How I Found The Happiness in SXSW?

Scientific Secrets of Perfect Timing

Weird Stuff

img_2724-e1521775599282.jpg

Dr. Octagon (Kool Keith, Dan The Automator, Q-Bert) from the Year 3000

Cheetos Vision: The Future of AI? Rapping robot

Deejaying without touch

When in SXSW, prepare to queue

Why Living in Singapore Makes You Perfectly Prepared for SXSW?

I hope that I am able to participate next year as well. It was truly great experience.

Tagged , , , , , , , , ,

Is Digital Targeting Just a Hoax?

Before I went to holiday, there was lots of chatter about the ”failed” Facebook targeting experiment of P&G. This naturally gave fuel to the fire to those denouncing digital advertising (namely Ad Contrarian). Essentially P&G run targeted Facebook for Febreze (pet owners and large families for example), but they got better results when they were just targeting broader audience of just over 18 year olds.

If you have been doing marketing at professional level for a while the results were not surprising at all. However, you should not use this as a proof point that targeting does not work, because of the following reasons:

1. FMCG is a different kind of beast, you can just blast your audience with bazooka

“The bigger your brand, the more you need broad reach and less targeted media,”

– Brian Weiser, Pivotal Research Analyst

Majority of P&G brands (including Febreze) are unique brands because they are truly for everyone. Majority of FMCG is mass reach, so it is not surprising that when you have broad targeting you have better results than when just focusing on few sub-segments. Actually in most of the markets you should not even bother with Facebook. If you have money running TV ads, they would still probably be more effective than doing anything on Facebook. And that is essentially what P&G has done. They have increased their TV spending. FMCG is first-and-foremost about top-of-mind and visibility on shelf. To achieve that you opt for the channel getting you maximum awareness.

Pretty much all the rest of the brands cannot work with such a broad sweep. Not all of the products live and die through the mass awareness. If you need to get 1000 quality leads, targeting the whole population is not most likely be more cost-effective than smart targeting. The main benefits of digital advertising come when you are selling in eCommerce, because you can then truly track your results and optimize. Then shooting with bazooka is not the right tactic.

2. Targeting without personalization is not targeting

Apparently they run the same creative to all the different segments. This is akin to running nighttime ad at 11AM. It is like narrowing the list of girls you want to go out to date with, but addressing them all with the same name. If content is king, context is truly the king kong. As you have narrowed your audience, you should also narrow your message to be as relevant as possible to your target audience.

3. Targeting based on intuition is not targeting

In the articles it was not said how the different target groups (pet owners and large families) were selected, but I would assume that they were based on human intuition. The beauty of digital advertising is that you let machines to try out different target groups, different messages and let them automatically favor what truly works. Humans are incapable of handling that many tasks and they are more biased than smart algorithm.

So the failure of Febreze seems obvious in hindsight. You started narrowing although your audience is as broad as it gets. You did not narrow your message to your narrow audience. Lastly you based your targeting on human intuition instead of testing potential audiences with machine learning.

The more we let algorithms handle our marketing, the more effective it will become. P&G experiment shows more human fault than failure of highly-targeted, highly automated algorithm-driven approach.

Tagged , , , , , , ,

Machines Will Eventually Beat Humans in Everything

“I don´t think it will be a close match. I believe it will be 5–0, or maybe 4–1. So the critical point for me will be to not lose one match.”
Lee Se-Dol (Korean Go champion before his matches against Alphago)

Lee-Se Dol was able to predict the future; it was just the opposite he was envisioning. Alphago (computer Go program done by Google subsidiary Deepmind) slaughtered him in six games.

alphago

Machines beating humans in a game is nothing new. In chess the gap between machines and human is already tremendous. Best chess machines are even able to win joint teams of human and computers. What makes AlphaGo´s victory intriguing is that Go is much more complicated game than chess. The first move of Go can involve 361 positions (chess has only 81) and Go game generally lasts more turns than chess.

Simple heuristics get most of what you need. For example, in chess and checkers the value of material dominates other pieces of knowledge — if I have a rook more than you in chess, then I am almost always winning. 
Go has no dominant heuristics. From the human’s point of view, the knowledge is pattern-based, complex, and hard to program. Until Alphago, no one had been able to build an effective evaluation function.”
-Jonathan Schaeffer (Creator of Chinook, first program to beat humans in Checkers)

The machine victory in Go happened decade earlier than experts predicted.

AlphaGo is based on deep learning and neural networks. So while Deep Blue beat Kasparov with sheer computing strength, Alphago has more artificial intelligence behind it. Firstly neural networks were trained on 30 million moves from games played by human experts. That resulted to ability to predict human move 57 percent of the time. But that gets you to the same level as human players not necessarily able to beat them. So secondly, AlphaGo played thousands of games between its neural networks, and adjusting connections using trial-and-error process through reinforcement learning.

How many humans are even able to comprehend what above means (lest train themselves in even somewhat similar manner)?

Machines can already replace humans in more fields than we are willing to admit. And more importantly, they are playing better job as well. Machines can crunch data to obtain experience, which is impossible for humans during their lifetime. We have to start embracing machine learning and collaborating with machines more if we want to survive. Advertising industry has been especially almost hostile to any technological improvement. That will be a road to sure destruction. Beating a Go champion is much harder task than to do a subpar brand campaign. If we don´t take more proactive and positive approach to data and artificial intelligence, we will make ourselves redundant.

Machines can either be our allies our friends. I would opt for the latter choice.

Tagged , , , , ,
%d bloggers like this: