The Blend360 Data Science Practice

August 3, 2021

Feng Jia, VP of Data Science Solutions


Patrick Hennessy, Tim Berry and Ozgur Dogan started the Data Science Practice at BLEND360 in 2017 and since that time, we have witnessed tremendous growth. Today, our data science solutions group brings expertise from their prior companies-firms such as Merkle, Amazon, Epsilon, Wunderman, Harte Hanks, Groupon, IBM, RR Donnelley, and more. We also have over 100 practitioners, experts who are leading data science projects daily, alongside and for our clients in diverse industries.

Associated with business growth, we have expanded our data science practice into many new areas. Today, we consider ourselves a strong player in the marketplace in the five key areas:

Audience Insight & Targeting

Audience insight and targeting involves audience curation, segmentation/profiling, lifetime value and loyalty modeling. Our solutions can be applied to both prospecting and customer management. Additionally, our wheelhouse includes marketing channel/touch point optimization, campaign management and reporting. And clients across industries seek our expertise in this area.

Marketing Measurement

Marketing measurement refers to the disciplines that use Design of Experiment to set up A/B tests or MVT(multi-variate test) for marketing campaigns to determine the optimal combination of product, promotion and pricing. It also refers to the use of Media Mix Modeling and Multi-Touch Attribution solutions to measure the effectiveness of different marketing and promotion tactics. Recognizing how important it is for marketers to understand which marketing tactic is working and which is not, we brought in experts, with experience from Merkle, Neustar and AC Nielsen, to enrich BLEND360’s knowledge in the MMM and MTA area. Now clients regularly rely on us to deliver top-down aggregated-level marketing mix modeling, and bottom-up individual-level multi-touch attrition and journey analysis for our clients.

3rd Party Data Solutions

3rd Party data solutions include our data optimization lab – a pre-built solution BLEND360 developed to enable our clients to refine their 3rd party data sources on evaluation and selection, leveraging 3rd party data for prospect targeting, customer append, digital channel onboarding, and matching. Our expertise on 3rd party data has been consistently recognized by our top clients. It also gives us competitive edge when developing customer/prospect insights & targeting solutions.  

The above-mentioned capabilities are inherited DNA from our Addressable Marketing Agency roots. On this solid foundation, we have expanded into the following new data science practice areas over the last several years.


Simulation and Optimization

We have witnessed advancements in data science (where Python became the dominant programming language) and technology (such as Machine Learning, Deep Learning and AI types of DS tools) amplify the business value and impact made by data science solutions. As depicted in the chart below, data science has moved away from providing only descriptive and diagnostic values to supplying predictive and prescriptive values. In our practices, BLEND360 uses data science to model clients' business processes, run simulation and “what-if” scenarios, and leverage linear and non-linear programming to optimize business outcomes. To enable optimization towards future goals, we often bundle predictive modeling and timeseries forecasting into the optimization solution. Related business use cases can be any type of personalization connected to pricing, product, and promotions.

Emerging Areas

Through projects with our clients, we have also embraced many new data science areas. One example is Natural Language Processing (NLP). NLP is a branch of data science that allows computers to understand and interpret human language. From written customer reviews to voice commands to virtual assistants, NLP extracts insight from human language to make decisions based on that information. Recently at BLEND360, we have leveraged NLP techniques to automate client call centers’ note reading processes and ran through clients’ social media listening platforms to extract useful insights.

Another example is fraud detection area. In addition to existing analytics and machine learning tools, graph data science tools have drawn attention in the fraud detection area. We are still in the early stage of exploring this approach.  

 

BLEND360

We live in the era of digitalization. The amount of data in the world was estimated to be 44 zettabytes (one zettabyte has 21 zeros) at the dawn of 2020. At BLEND360, we are working to continuously develop and enhance our data science practice and solutions to wrangle all that data, empowering businesses and driving better value and outcomes for our clients. This is our passion and our mission!

RELATED POSTS