This very publication was founded on the age old adage:

“Half the money I spend on advertising is wasted; the trouble is I don’t know which half”

A new business is trying to answer that question, which 50% is working?

Former advertising executives Matt Farrugia and Henry Innis are the founders of Mutiny, a media analytics platform which claims to translate media data into clear, quantitative insights designed to improve decision making around spend and measure marketing ROI.

AdNews reports the start up has recently attracted the interest of seasoned advertising industry leaders, with Nick Garrett and Andrew Baxter having invested in the firm, though the size of the investment is unknown at this point.

With the advertising spend in Australia expected to exceed $17.5 billion in 2021, it’s a worthy market to explore.

Pathmatics provides insights into competitive digital ad spend.

Mutiny wouldn’t be the first to try and crack this lucrative market. Firms have tried in the past, with mixed success when trying to provide measurement to what was previously deemed unmeasurable.

Last month, digital marketing intelligence platform Pathmatics announced it’s expansion into Australia, attempting to lift the lid on brands’ ad effectiveness by offering brands and marketers insights into not just their own ad spend and creatives but also their competitors, across digital, video, display and mobile.

The Dark Arts

While ad spend itself may be easier to determine, others have tried to value more intangible metrics like brand and social media impact.

NYU Stern Business School marketing professor turned media identity Scott Galloway launched L2 Inc in 2010, a subscription research and business intelligence firm that benchmarks the digital competence of consumer brands using a range of available touch points. L2 was eventually acquired by Gartner for US$134m.

In an attempt to crack the code on the value of social media efforts, using social media analytics to rate its users according to online social influence and assigning a “Klout Score”, a numerical value between 1 and 100.

Klout wasn’t without its skeptics. Several objections to Klout’s methodology were raised regarding both the process by which scores were generated, and the overall societal effect. It was argued Klout scores were not representative of the influence a person really has, with examples like Barack Obama, then President of the United States, having a lower influence score than a number of bloggers.

Mutiny founder Henry Innis responded on LinkedIn, “We’re definitely not Klout.”

“The short answer is econometrics, and instead of using Data Scientists to find model fit use ML (Machine Learning) techniques with priors.”

Klout was later acquired by Lithium Technologies for around US$200m, which intended to harvest the tool’s artificial intelligence and machine-learning capabilities for use in its other products.

How To Train Your AI

AI and Machine Learning may be the key to future technologies, but its not a silver bullet in and of itself to solve real world problems. The challenge exists, as it does everywhere with AI and Machine Learning, on how we train the machines.

Machine Learning is rocket fuel to improve upon optimised systems – so if we don’t know what success looks like, how can we teach our machines?

Mutiny and their WarChest platform are attempting to crack that code and if they get it right, the rewards could be exponential.

Previous post

Cryptocurrency Volumes Plunge

Next post

French Regulator Fines Google 500 Million Euros