Marketing-driven forecast of agricultural production

Application Track:

Theme Driven




Proposed by:


Entity Logo:

Summary of the entity:

The University of Deusto, recently recognized as an International Excellence Campus, was founded in 1886 and comprises 6 Faculties: Psychology and Education, Human and Social Sciences, Engineering, Law, Business and Economic Sciences and Theology. The MORElab research group has worked in several projects dealing with Linked Data, Big Data and Data Analytics. In the Spanish research project THOFU, the group undertook activity recognition over the PlaceLab dataset and opinion mining regarding hotels. In the IES Cities project, the group worked on the creation of a set of urban apps that make use of data published by councils and that leverage the information generated and published by users (crowdsourcing). In WeLive H2020 project, our Open Data Stack has been extended with a repository of micro-services consuming open data and a wizard-like tool that enables citizens to co-create and co-define public services, giving place to an Open Government solution. In EDI – European Data Incubator, DEUSTO is in charge of the project coordination and has an important role in WP3-Big Data Infrastructure and Support, where we share with ENG the duty of providing a Big Data experimentation infrastructure for start-ups (sub-grantees) which tackle through Big Data large corporations’ data challenges.

Summary of the challenge:

The main goal is the development of a predictive service that, from the estimated degree of success of a digital marketing campaign, predicts how its impact will become into real purchase orders to the producers.


The profitability and impact of a digital marketing campaigns strongly depends on the quality of the selected keywords. However, when generating new keyword list that will form the campaign structure, there is no information about their quality in terms of their potential impact, measured as the probability of impression and click through rate (clicks over impressions).

The prediction of the success of a digital marketing campaign is a hard problem by itself, but in this challenge, we want to go one step further and try to estimate its cross-sectorial impact: we want to explore whether having an estimate of the campaign’s impact can be used as a predictor to help producers of the advertised products to forecast and optimise their production needs.

In particular, the challenge focuses in the agriculture sector where many products are fresh and perishable, and good planning can help reduce waste.


REACH Data Provider

AN GROUP: databases with historical orders of agricultural products (quantity, product, price, sales, company)

JOT Internet Media: Keyword list (mobile), and Keyword statistics in mobile campaigns. Two datasets that, when filtered appropriately, reflect the impact of marketing campaigns in the Tourism sector with relevance to the Valencia region and specifically, their inland cities or rural regions (e.g. historic villages in Castellon, cultural heritage in Valencia, nature experiences in Alicante, …). These datasets show the main interests and trends of people aiming to travel or visit Valencia region in the coming period, therefore they may be used to support decisions making by predicting people displacement or means of transport management.

External Data Provider: Provided by applicants

Open source:
Energy price:

Raw material price:


Expected outcomes:

To optimize production of agricultural products based on digital marketing data.

How do we apply?

Read the Guidelines for Applicants

Doubts or questions? Read more about REACH on the About Us page,

have a look at our FAQ section or drop us an email at