Posted by Jeevan Jayaprakash

In this issue, we look at Google’s new cloud offerings, the latest advances in battery technology and why it is so important and finally, a case study on River Island’s use of ‘Local Inventory Ads’.

In other news, we are delighted to announce that the Carling Beer Button took home ‘FMCG Campaign/Strategy of the Year’ at the Drum Network Awards last week!

We love user feedback, so if our digest is too long, short, amazing or boring — just let us know at

Have a great week,

Hi Mum! Said Dad

Google Cloud rolls out machine learning for enterprise customers

You can be forgiven for thinking that machine learning (and artificial intelligence in general) is something reserved for the Googles, Teslas and Apples of this world. However, Google has recently announced that it is taking its cloud offering to a whole new level by offering enterprise customers machine learning services. Some of the services on offer include translations, job matching, sentiment analysis, image recognition and more.

So, what is machine learning and why does this matter?

Machine learning gives computers the ability to learn without being explicitly programmed. Learning can fall into two categories: supervised learning and unsupervised learning. The former involves providing an algorithm with a training set of examples (i.e. the ‘right answers’) that it can learn from where as the latter involves asking an algorithm to find structure in a dataset and identify patterns on its own. For machine learning to be effective, you usually require huge data sets and therefore a computer that is powerful enough to process this data.

This is where Google swoops in to save the day. Google is allowing people to rent these powerful machine learning via the cloud and is providing the tools needed for enterprise customers to start using machine learning algorithms in areas such as recruiting talent, increasing conversion rates and even in core products. The Google Cloud is built for scale and will be offering GPUs for use in 2017 (yes that’s right, a graphics processing unit typically used for gaming but reengineered in this case for handling large amounts of data).

This is significant because Google is essentially providing everything on a plate. Machine learning opens up a whole host of opportunities for businesses and this also means that it opens up the risk of being left behind. Regardless, this is a very exciting development and one that makes this oft-talked about technology a lot more tangible for many.

Slides from a Google deck at a special press event. Image source: MarTech

US scientists create revolutionary battery prototype

Scientists based at the University of Central Florida (UCF) are said to have pioneered a new battery technology that could charge your mobile phone in seconds and allow it to last for up to a week. Rather than using chemical reactions like a conventional battery, the scientists at UCF utilised super capacitors, which are able to store electricity statically on the surface of a material.

It is worth noting that we have had supposed breakthroughs like this in the past but they have never managed to materialise into something that is commercially viable. Researchers at UCF have said that this piece of research is still in its very early days (a proof of concept more than anything) and is not yet ready for commercialisation. Despite this, there is a general feeling of optimism that the scientific community is getting closer and closer to a tipping point.

So, why should we care?

Well, some would say that their smartphone battery life is the bane of their existence (which is something quite sad in itself). The mobile world and consumer behaviour is predicated on the assumption that a typical smartphone’s battery only lasts for a day. We are advised to turn off our mobile data, bluetooth and location services when they are not in use — all in the name of saving battery. We also have manufacturers like Apple and OnePlus who still use 1080p screens in their flagships. OnePlus takes the view that there is no need for anything above 1080p because it diminishes battery life and that people don’t do the whole phone in VR headset thing (yet anyway), where a 2K/4K resolution screen would be required.

With Apple leading the way to a wireless future with the iPhone 7 and the 2016 MacBook range, better battery life needs to come around and fast. It is ironic that battery life is lagging behind the very technology that it is meant to power.

However, when the breakthrough does arrive, it could be significant and have far reaching implications. For example, beacon technology and IoT devices will instantly become more powerful and more relevant. Likewise, battery life of a week could become a serious enabler for AR and VR technology on the go. The scope for changes in consumer behaviour is potentially quite huge. We will be able to move closer to a realisation of the always on, always connected world (think smart cities). Throw in the expected roll out of 5G data speeds in the UK in 2020 and you can see why a breakthrough in the near future could be a game changer.

The small and flexible nature of the battery makes it perfect for wearables and foldable phones (Samsung, we are looking at you). Image source:

Case study: River Island becomes first UK retailer to use ‘Local Inventory Ads’ on mobile

River Island recently teamed up with Google to create a more seamless experience for customers on mobile. The fashion retailer made use of ‘Local Inventory Ads’ or LIAs in order to display products that are within close proximity of a user’s location. This was very much a concerted effort between the online and the retail teams which ensured that only products available both online and in-store were surfaced — giving users options and a sense of control over the experience. The ads are only triggered if users search for a product that satisfies this dual availability condition.

River Island found that 6% of mobile clicks on LIAs resulted in a store visit, which was a 17% increase in store visits compared to standard campaigns. Return-on-ad spend also rose by 15% and there was also a 33% increase in sales when offline performance data was included. The retailer has been impressed by the ability of LIAs to drive in-store sales and have pledged to experiment further by using store visit data to tweak their automated bidding strategy as well as potentially using beacons to connect the dots and further understand in-store behaviour.

River Island: Think with Google

River Island benefits from unrivalled, targeted visibility. Image source: Think with Google

Originally written as part of Hi Mum! Said Dad’s Weekly Digest.

If you would like to receive our Weekly Digest straight to your inbox, please drop us an email at