Hot Topics
- Home
- /
- Campaign Management Solution Overview
- /
- Building a campaign data visualization platform
Figure 1 – Illustrates the data requirement for a campaign to launch new cell phone. Campaign Operation teams and business users need data to be presented in manner relevant to the business context. However, the data in campaign data mart is stored in a manner that aids performance and scalability. Traditional solutions adopted by campaign teams and business users result in reconciliation issues, higher consumption of processing resources and longer campaign cycle times.
Building a campaign data visualization platform
Introduction
Strategies for campaigns constantly change to stay in tune with market trends and customer requirements. Implementing such changing strategies to create effective campaigns is possible only if campaign infrastructure provides the data required to implement these strategies in form that is understood by business. However, there is lot of scope for improvement in the way this need is implemented.
In most cases, campaign implementations do not follow a standard design. This leads non-uniform implementation of campaign processing steps across different sub-systems.
Changing visualization requirements creates complications
To be effective in campaign operations, it is essential to keep pace with constantly changing user requirement of how they would like to work with data. There is a difference between how users like to visualize data when designing a campaign and how the data is actually stored in the repository. While users would like to see data in familiar business terms the data is however stored in a manner that cannot be readily understood because it is optimized for execution from a technology perspective. The challenge arises in the way this difference is resolved by most campaign teams.
Measuring effectiveness of campaigns
Campaign teams adopt ad-hoc approach to monitor status and measure campaign effectiveness. Standardizing the process helps to eliminate manual collation of status information.
Consider the example of a campaign to launch a new cell phone. Such campaign requires data to be presented in user understandable terminologies as shown in Callout A in Fig 1 above. The campaign data store however stores this information in a very different table structure with separate tables as shown in Callout B. This design improves performance and scalability but constrains users from readily using the data.
Therefore users form data sets, which are closer to the way they want to visualize the data, through user created tables as shown in Approach A. This approach has drawbacks. It creates redundant data, which consumes system resources and leads to poor performance. The data in such tables have reconciliation issues with base campaign data tables, due to difference in its frequency of data refresh with base campaign data store. Tables created for specific campaigns need to be refreshed with data that is already available in the campaign data store thereby increasing cycle time and processing overhead in the data center. Further, tables created in Approach A become redundant very quickly as it is unlikely to meet the requirements of subsequent campaigns without changes or reconsolidation.
The above challenges are particularly acute in enterprises with large transaction volumes and an aggressive campaign marketing operation.
Data readiness crucial to execute campaign
Predictable data consolidation is required every time, else be prepared to lose business. However, campaign marketing organizations pay a heavy price by not paying attention to supporting data infrastructure.
Implementing a semantic layer for data visualization
In order to handle fast changing data visualization requirements, a semantic layer should be built above the campaign data store. This is shown in Approach B, Fig 1. Semantic layer maps complex data into familiar business terms such as product, customer, or revenue thereby enabling users to use the data autonomously.
This layer allows faster access of data with provision of implementing business rules with minimal effort and time. This layer consists of views along with an efficient data aggregation, filtration, consolidation and refresh mechanism.
An effective semantic layer also allows campaign operation teams to intelligently define what is the dynamic data to be extracted from underlying data, how much data is to be pulled, frequency of data retrieval and also privileges for data access in terms of business groups and individual users.
Building agile and efficient marketing teams
Ad-hoc approaches in implementing online campaigns are impacting business. It is taking longer than ever to implement new campaigns and operation costs are on an increase.
Implementation of an effective semantic layer starts with the documentation of all business needs in terms of measures, metrics and all related business terminologies associated with data. This information is used to design components of semantic layer in term of base data, data governed by business rules and final data to be used by campaigns. These elements of semantic layer further drills down to tables, column names, filter condition, group and sort criteria for the data, etc.
These elements should be grouped in a hierarchical fashion to the extent possible. The completed element list, definitions and underlying tables should be discussed with the entire operations team in order to identify similarities and differences in data terminologies that create confusions in semantics.
Using this consolidated information an efficient semantic layer is built. With the implementation of a semantic layer, users get data based on the dynamic requirements of campaigns without redundancy, performance and data reconciliation issues.
Improving marketing campaigns
EXILANT helped one of the largest consumer electronics product company to establish and enhance their campaign system. The solution enabled the client to manage and improve sales and services of their products.
Summary
A semantic layer provides a good platform for users to work with a business view of data. This will help speed up campaign cycle time, improve team productivity, and reduce data center costs while still providing data in a way that users need to execute campaigns effectively. With our expertise in implementing the campaign management systems, we can assist clients in their initiatives to improve performance of their campaign data infrastructure.
Recent Links :
Measuring effectiveness of campaigns
Campaign teams adopt ad-hoc approach to monitor status and measure campaign effectiveness. Standardizing the process helps to eliminate manual collation of status information.
Data readiness crucial to execute campaign
Predictable data consolidation is required every time, else be prepared to lose business. However, campaign marketing organizations pay a heavy price by not paying attention to supporting data infrastructure.
Managing marketing campaigns
Marketers face challenges in operating their campaign smoothly despite having multiple systems. They need "ready to execute" infrastructure and analytical approach to take charge of the campaigns.
