About Prasanna

Hello there! Thank you for visiting my blog. This blog is the go-to-place for any content created by me. So please do come back often to check what is new.

Disclaimer - As such, all the views here are my own, and does not represent anything else I am associated with.

I am a marketing data analyst with more than 8 years of experience in the field. I have worked with various companies in the past, and have helped them in understanding their customers better. I have also helped them in making data driven decisions.

You can email me on - prasanna.k158@gmail.com

Me and Marketing Data Analytics

I use KPI framework to define KPIs for website and marketing campaigns. The build data pipelines so that most up to date data is available for analysis. I use SQL, Python, R, Tableau, Power BI, Google Analytics, Adobe Analytics, Excel, Google Sheets, and many other tools to get the job done.

I follow best practices for data analysis and visualization.

My toolset includes -

  1. Anamoly detection (Isolation Forest, One Class SVM)
  2. Marketing Mix Modeling (Markov Chains)
  3. Lead scoring (Random Forest, Gradient Boosting)
  4. Revenue Forecasting (Time Series Analysis)

Also -

  1. Customer Segmentation
  2. Customer Lifetime Value
  3. Customer Churn Prediction

Work Samples

1. QTD Target vs Actual Dashboard (Burnup chart)

I have set up these charts to track performace against target on a weekly basis while giving visibiltiy against target and last quarter performance. Week by week attaining target is a key metric for any marketing team. This can be used to identify the segments which are underperforming and take corrective actions. The method can be extended for other metrics like leads, MQLs, Pipeline as well as geographies. These charts when embedded in slides are readable and useful for weekly reviews.

Below chart is cummulative opps created for each week in a quarter to show pacing against the quarterly goal

QTD Target vs Actual Dashboard

2. Anamoly Detection

Anamoly detection catches things to look at on a proactive basis. It reduces the chances that huge changes go unnoticed for a few days.

Anamoly Detection

3. SEO Analysis Python Library

Created a Python library to analyze HTML and extract SEO related information. Link - https://github.com/prasannakulkarni333/python-webcitizen-audit/blob/master/src/webcitizen/webpage.py

The documentation is available at - https://webcitizen.prasannakulkarni.com/