Masterclassing Connect

2019 Award Winners: Voice Assistant App, Anti-fraud Solution and Use of Video categories

David Murphy

Following our recent Effective Mobile Marketing Awards Ceremony, we continue our round-up of the winning campaign in each category with a look at the winners of the Most Effective Voice Assistant App; Most Effective Anti-fraud Solution; and Most Effective Use of Video categories.

Most Effective Voice Assistant App
BBC Good Food and Hi Mum! Said Dad – BBC Good Food Alexa Skill
BBC Good Food is the UK’s number 1 food media brand across print, digital and live. However, with smart speakers becoming the fastest-growing consumer technology, the evidence suggested that for BBC Good Food to remain the UK’s number 1, voice was imperative. Google reported in 2016 that 20 per cent of all searches on mobile were done by voice and ComScore has predicted that voice will account for 50 per cent of all search queries by 2020. Research also suggest that around 50 per cent of smart speakers are based in the kitchen.

BBC Good Food and Hi Mum! Said Dad faced two challenges: to be the go-to VUI (Voice User Interface) cooking experience; and to come up with new ways to monetise the content. given that the ads the BBC Good Food serves on its website can’t be served on Echo devices.

The Skill’s main USP is that it enables hands-free cooking, so key to its success was to look at how people actually cook and how the Skill could alleviate any pain points and get the user to the one recipe they wanted from the 11,000 recipes in the BBC Good Food store cupboard – all without visual cues.

To do this, they had to consider the most popular search intents for users, such as available ingredients, dishes, diet types, time (e.g. ‘quick’), difficulty (e.g. ‘easy’), cuisine, course and chef, and design questions to ensure that they extracted the precise information we needed in a predictable, unambiguous format that Alexa could query against the database effectively.

The solution was to enable a range of conversational commands with filtering criteria baked in - for instance: “Alexa, Ask BBC Good Food for a quick and easy meal for four, using chicken and tomatoes”. With that single sentence, five filters are applied effortlessly - 2x ingredients, time to cook, difficulty level and serving size.

To optimise the listening experience, Hi Mum! Said Dad restructured the content, breaking instructions down into smaller lines, considering pacing and including functionality for enhanced control, such as pausing, repeating and bookmarking so that users can pick up right where they left off.

The agency also reviewed typical pain points with cookbooks and screen displays and found that the most common annoyance is the separation of recipe method and ingredients. The instruction “add the sugar” is typically met with the question, “how much sugar?" and so they enabled the link between these traditionally separated elements for a seamless walk through. Users could simply ask "how much?" and the Skill would confirm this instantly.

The combined effect of the enhanced search flow, voice tailored content, pacing controls and the bridging of method steps and ingredient quantities make for a genuinely hands-free cooking experience.

To monetize the app, Hi Mum! Said Dad integrated with online retailers to monetize the purchase of ingredients via an affiliate arrangement, and produced unlockable premium content.

Since its launch, Amazon has recognised the Skill as one of the best recipe Skills in the world, to the point where Alexa now directs all recipes searches in the UK and Ireland to BBC Good Food by default. User numbers were shared with the judges in confidence and are impressive.

Judges’ View: “This campaign faced new challenges, and overcame them through smart innovation and carefully crafted design. Voice User Interfaces are still at the development stage – this campaign contributes to the growth of consumer adoption of the technology.”

Most Effective Anti-fraud solution
Scalarr – Scalarr
Machine learning-powered mobile ad fraud detection solution Scalarr was started in 2016 with the idea to develop an ‘efficient and highly’ accurate tool to detect all mobile app install ad fraud. It set out to do this because it found that existing anti-fraud solutions were only able to see ‘primitive’ fraud and unable to detect new patterns and modifications.

The Scalarr team is made up specialists in data science, machine learning, and mobile advertising. These specialists work to ensure that customers do not waste marketing budget on fraudulent non-human traffic and get ‘real’ LTV (lifetime value) and ROI (return on investment) metrics.

The company first introduced machine learning to its fraud detection in June 2017 and by September 2017 it had been joined by its first clients including Aviasales, Pixonic, and Goodgame Studios. A year later, it introduced a neural network model to uncover smart types of fraud.

In 2018, it investigated 170m installs and analysed 22bn events, uncovering 35.7m fraudulent installs. In this time, it also found 18 new fraud patterns.

Judges’ View: “Great results delivered using machine learning to detect new patterns in fraud – truly innovative.”

Most Effective Use of Video
The Home Office, Manning Gottlieb OMD and Venatus Media – #KnifeFree
In November 2018, Venatus supported the Home Office’s #knifefree campaign to help tackle one of the Home Office’s key priorities – reducing knife crime in the UK. Young males aged between 15 and 18 were identified as the group most at risk of carrying or considering carrying knives, and this audience was found to have a passion for gaming, spending more than eight hours playing games each week.

The Home Office worked with Venatus to reach this target audience in a trusted environment, both whilst they were at home on consoles and on the move playing apps. On mobile, via EA apps, there was also the opportunity to cut-through to the target audience and increase engagement as the viewer chooses to watch the advert to receive an in-game reward.

The campaign used three different video creatives, each telling the stories of real people who have turned their lives around. The videos linked to the #knifefree website, providing access to information about the consequences of knife crime and inspiration on how to move away from it.

The targeted gaming campaign ran alongside targeted out of home, paid search, digital audio, and use of the video in shorter teaser formats on Snapchat and Instagram with a ‘swipe up’ message encouraging people to engage with the longer form assets.

The PlayStation video part of the campaign delivered 422,656 impressions with 267,065 completed views, generating a view-through rate of 62.2 per cent compared to the industry benchmark of 50 per cent.

The EA reward video part of the campaign delivered 267,796 views and 5796 clicks, exceeding benchmarks and generating a CTR of 2.2 per cent. The EA reward video received the highest view-through rate of all campaign channels at 95.9 per cent, exceeding the already high industry benchmark of 90 per cent.

Judges’ View: “A highly impactful, carefully targeted campaign which showed great insight into the target audience, where they could be found, and the creative required to engage them.”