Keyword impact estimation for high-performance marketing campaigns

by | Nov 10, 2020 | 0 comments

Application Track:

Ready Made

Code:

REACH-2020-READYMADE-JOT_2

Proposed by:

JOT INTERNET MEDIA (JOT)

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:

The main goal is the development of a predictive service enabling the selection of the best keywords to implement high impact digital marketing campaigns. This selection will allow the generation of brand awareness and market engagement thanks to the keyword performance estimation (probability of impression and click through rate indicator)

Description of the global challenge:

JOT implements a massive and performance driven strategy to generate high quality web traffic. The profitability and impact of the marketing campaigns strongly depends on the quality of these keywords, so the selection of the most adequate terms is critical to reach our clients demands.

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). In this way, marketing managers do not have proper insights supporting the definition of the most adequate keyword list to reach their marketing goals.

In this scenario, there are many technical challenges to be solved to offer this prediction as a Service model, such as keyword data analysis, semantic classification, data volume, data variability and so on.

In this challenge data set will be related to laptop campaigns in all categories for Spanish and English campaigns where the campaigns performance is described by 20 variables.

In conclusion, this global challenge aims to generate a service enabling the selection of the most suitable keywords ensuring expected marketing campaigns impact. This selection should be based on two main factors: probability of impression (linked with brand awareness) and click through rate (related to audience engagement)

Sub-challenges composing this experiment:

This challenge is composed of 2 sub-challenges:

  • Estimation the probability of impression at keyword level (REACH-2020-READYMADE-JOT_2.1)
  • Prediction of click through rate (REACH-2020- READYMADE -JOT_2.2)

Expected global results:

  1. Achieve that 50% of the keywords of a campaign impress
  2. Generate campaigns with >15% of ROI
  3. Processing of 500,000 keywords in batch mode
  4. Robust predictions supporting decision making (Error metric >0,75)

ESTIMATION THE PROBABILITY OF IMPRESSION AT KEYWORD LEVEL

Code:

REACH-2020-READYMADE-JOT_2.1

Summary of the sub-challenge:

The goal is to get a system capable of filtering the large list of keywords in marketing structures and select those with higher potential according to each specific campaign.

Description of the challenge:

The generation of keyword list forming the marketing structure involves processing time and costs not always returned due to low quality/performance keywords. This aspect has also associated financial cost that limit the gross margin of the marketing campaign. For that reason, it is needed a system filtering the large list of keywords and selecting those with higher potential impact according to the campaign definition and goals.

Data to be used:

  • Keyword list (mobile)

Expected outcomes:

  • High impact keyword list (+50% of impression)
  • Analysis of 500,000 keywords in batch mode
  • Multi language (English and Spanish as a must)

PREDICTION OF CLICK THROUGH RATE

Code:

REACH-2020-READYMADE-JOT_2.2

Summary of the sub-challenge:

The aim is to calculate the expected Click Through Rate (CTR) at keyword level to select only the most profitable keywords depending on each campaign specific goals.

Description of the challenge:

In performance driven campaigns, it is needed that a significant number of user’s clicks on the ads and visit the publisher landing page. Although the configuration of the campaigns may ensure high probability of appear in top position of the search platform ranking, the account managers need to know the expected number of clicks, as it generates cost and its relation to the number of impressions (which generate brand awareness). This rate is called click through rate (CTR) and determines the performance of the campaign in terms of profits and market engagement.

For that reason, it is required a predictor calculating the expected CTR at keyword level to be able to select only the most profitable keywords depending on the campaign goals.

Data to be used:

  • Keyword list (mobile)

Expected outcomes:

  • Implementation of campaigns with ROI > 15%
  • Analysis of 500,000 keywords in batch mode
  • Multi language (English and Spanish as a must)

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 opencall@reach-incubator.eu.