Prediction of temporal patterns of keywords and categories behaviour

Application Track:

Code:

Domain:

Proposed by:

Entity Logo:

Summary of the entity:

JOT Internet Media is part of Cube Ventures, one of the leading digital groups in Europe. Based in Madrid and with offices in Germany, Mexico, Brazil and Italy and more than 280 employees, the group is focused on lead generation, services monetization, digital marketing, media and investments.

Summary of the challenge:

Analysis of the keyword and categories behaviour to generate temporal models at day of the week, week of the month and month of the year. These prediction models will allow marketing managers to adapt the marketing campaign management to the seasonality of the user interests. 

Description:

Analysis and prediction of temporal patterns of the behaviour of keywords and categories according to different seasonality criteria

a. By day of the week. Which keywords and/or categories have more traffic and conversions each day of the week. The objective will be to look for patterns/trends that relate some keyword, or some category, with some day of the week and/or weekend

b. By week of the month or beginning/end of the month. Since the traffic varies a lot depending on whether it is the beginning or the end of the month, the objective will be to look for patterns/trends that relate some keyword or category with the first days of the month or the last 7/10 days of the month.

c. By month of the year. Clearly oriented to search keywords or categories that may have seasonality related to the month or season of the year.

The development of prediction models will allow us to anticipate or adapt marketing campaigns according to these criteria of seasonality vs. user searches or categories of interest.

The reporting service should be done at the country level. Initially, the countries to be analysed will be GER, GBR and USA.

Data:

• Marketing Statistics
• Campaign indentifiers, names and categories

Expected outcomes:

At least one of the following should be addressed:

• Outcome 1: Classification and prediction of the behaviour of the keywords and categories based on the day of the week (minimum working day and week end);
• Outcome 2: Classification and prediction of the behaviour of the keywords and categories based on the week of the month;
• Outcome 3: Classification and prediction of the behaviour of the keywords and categories based on the month of the year.