Seven Steps to Actionable Insight from Real-Time Multi-Channel Feedback Sources


Share on LinkedIn

While market research in the form of sample-based solicited feedback is still a crucial tool for product testing, making market-entry decisions, and in overall strategy formulation, there is strong trend towards more spontaneous, real-time feedback. This type of feedback is often unstructured and from multiple channels, bringing with it its own set of challenges.

This post will present a high-level outline of the most important steps to gaining truly actionable insight from real-time, multi-channel feedback sources.

1. Ensure that you are working with as rich data as possible.

Background variables like customer identity, related demographics, and key purchase behaviors can be used as rules for effective distribution of insights to people in different organizational units and roles.

2. Make sure the structure of all feedback data sets (i.e., feedback data from one source) is as uniform as possible.

Many predictive analysis projects and real-time feedback management processes fail simply because each data set has a unique set of fields, which makes it difficult or impossible to compare the insights from multiple feedback sources. A uniform data structure can be pretty easily achieved using a few simple data scripts and macros and importing user, demographics and key purchase behaviors from other systems such as CRM.

3. Automate the flow of data from feedback channels to the aggregation and analysis service.

Because most operational insights lose value with time and uploading data manually requires discipline and resources, all data sources should be connected in real-time using, for example, REST or SOAP protocols.

4. Choose an aggregation and analysis service that is capable of automatically analyzing free-form text feedback.

Since most of the feedback data will be free-form text-based, an automatic, often multi-language text analysis functionality must be included. And customized ontologies need to be created to capture the differences between various industries and companies.

5. Create a single “master” data set with all the feedback.

All data that has a comparable structure should be aggregated to a master data set for top management, PR, marketing, and risk management purposes. In the master data set, the data source should be used as a new background variable.

6. Create a dashboard user-interface in which people have access only to the exact data that is relevant for performing their specific tasks.

People from different organization units, and functions have different data needs and don’t have time to sift through irrelevant data. An effective dashboard must allow rule-setting based on background variables or text analysis topics or sentiment.

7. Enable more advanced, big-data analysis using visualization and management dashboard solutions.

Make the analysis results available to third-party CRMs (Salesforce, MS Dynamics) as well as visualization (e.g. Tableau) and management dashboard solutions (e.g. QlikView) in the right format and structure for further correlation analysis by big-data experts.

Matti Airas
My main mission during the last three years have been trying to figure out how free-form text feedback (social media, forums, NPS, transaction queries etc.) analysis can improve overall customer satisfaction. Specialities: Customer Experience Management, Text Analytics, Internet applications and services, mobile applications, customer development and sales, management and startups.


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here