Customization beyond metadata

Customization beyond metadata

Solutions for linear and non-linear content

Businesses that rely on the internet are always on the lookout for better ways of promoting their products and services to users, viewers and buyers. Media Press Group is developing a system for generating solutions and recommendations for businesses that are under added pressures of tailoring services to customers with diverse and specific requirements. There is little to be gained in presenting a consumer with an advertisement for a product or service that either he/she has already purchased, or has little or no interest in acquiring. For the supplier, it is a waste of time, effort and money. For the consumer, it is an unwelcome distraction. At a time when we can all see that there is an insatiable demand for customisation in the global internet advertising marketplace, it is frustrating for investors and businesses to be aware that there is nothing that seems to fit the bill. However, music and video content owners and operators are able to select purchase solutions for subscribers, users and viewers, by applying a series of recommendations in order to narrow down content – on demand. Customisation means businesses can raise their game by 30%. There are common methods of generating product/service recommendations, notably collaborative filtering.

Troubles with historic data or GDPR?

One type of collaborative filtering solution generates a list of recommended items, based on perceived consumer choice. It works by concentrating on predetermined target groups, akin to communities of users, and generates a set of recommendations for services and products based on historic preferences. But what happens when an entirely new and innovative product or service comes along? However detailed and extensive the market research, no historic statistical data is at hand to identify and target potential customers.

The alternative method turns the process on its head. This item-based filtering process analyses consumer preferences in the opposite way. It matches items to certain target groups. However, there are legal obstacles in the way of constructing databases profiling individual consumer’s purchasing habits. For either filtering method to work effectively, the marketer needs to create a specific user profile. And the GDPR is in a powerful and sentinel position, governing what can and cannot be done with personal data for marketing purposes.

Content-based filtering

Instead of focussing on building detailed and problematic profiles of  individual customers in order to sell the latest version of a product, Media Press Group is posing a rather different appraoch. Where the product comprises linear TV programmes, what is the best method for tailoring content to individual subscribers – in a way that is not intrusive or challenging? The biggest challenge that TV content providers face nowadays is not a lack of content, but how to link content to individual viewers. The third filtering method, known as content-based filtering, also focuses on items.

It does not recommend items according to the perceived or historic preferences of any target group, but instead takes account of the way in which the item shares similarities with other current trending items. It is a reliable, predictive method, based on up-to-date consumer approval responses. Media Press Group’s product recommendation solution owes much to its substantial asset – namely TV metadata – which profiles both linear and non-linear TV content in high quality detail.

With a content-to-content recommendation solution, there is no ‘cold start’ problem in terms of generating vast amounts of data profiling individual consumers. What is more, we are able to make recommendations to viewers according to bespoke tastes. The system is adept at recommending new and untested items (in terms of established viewer popularity), according to different genres of content. The Media Press Group recommendation solution package not only encompasses films and TV series, but also entertainment output, documentaries, shows, lifestyle programmes, sports coverage and kids serials.

Uniform structure

The major drawback that has previously held such a solution at bay is that TV content is stored in a variety of formats by a multitude of competing operators. It has made the job of merging, matching and scoring content across the TV sector all the more problematic. But Media Press Group’s unique selling proposition is that it has a vast store of TV content metadata at its disposal, recorded in a uniform structure, replete with national and local content. This uniform and universal approach makes it child’s play to score programmes and filter recommendations to viewers – appropriately and comprehensively. Our ‘one stop shop’ database records content from start date to cast and crew, and from genre to theme. The results are updated continually and monitored closely by our editorial teams, and the results are integrated into Media Press’ standard EPG or OTT data feed.

Find the best solution on #IBC2018

The options for exploitation are exciting. Media Press’ IT team can design and create the best option for your business.

If you would like to discuss further details about our recommendation solution, then contact us today and set up a meeting at the IBC between 14th and 17th September, at hall 14, stand no. A12.

Media Press Group Sales Team:

Miłosz Cechnicki – Chief Operating Officer milosz.cechnicki@media-press.tv

Christian Töpper – Head of Business Developement christian.toepper@media-press.tv

Gyöngyvér Annus – Regional Director Hungary, Slovakia, Czech Republic, Romania gyongyver.annus@media-press.tv

Tamás Salusinszky – Regional Director Adria tamas.salusinszky@media-press.tv

Dubravko Bilobrk – Director of Adria Content Production dubravko.bilobrk@media-press.tv

No Comments

Post A Comment