Amber Toro-Keech's Linkedin Analytics

I have optimized over 100 ad accounts for maximum efficiency, developed forecasting models that predict within 1% of the future, and developed SEO analytics solutions that give deep keyword insights before a visitor even hits the website. Developing solutions to complex questions that drive high impact is what I am all about. ⭐Project Highlights⭐ 💻Page Purpose Project - I used Python and Natural Language Processing for content analysis of thousands of web pages to categorize those pages based on the purpose they serve in the customer experience. Not only were we able to solve a problem using NLP that we would not have been feasible using traditional methods, we were then able to use that work to identify areas where the customer journey was flowing as expected, vs. areas that needed improvements. 💻Google Ad Impression Dashboard - I developed a dashboard in Tableau that not only segmented and visualized the time series data that Google provides, I used that data to develop additional features to give us a better insight into demand and market fluctuations relevant to our niche in the space. This helped us to better understand search patterns and demand relevant to our domain, not obscured by looking at larger segments of the market. 💻SEO Keyword Opportunity Integration - I developed a strategy to combine different data sources to help us better understand the impact of individual queries on SEO. This tool was used to help the SEO team understand where to focus their efforts to deliver the biggest impact, not just on traffic but on conversions. 💻Marketing Mix Modeling - I built a time series forecasting model that forecasts conversions based not only the company's actions, but also influences from the market. I used feature engineering on the data available to us to model market factors along with developing an ad stock feature to more accurately represent the impacts of TV. I then used Python to test different machine learning models to forecast future conversions. 💻Founding Member of Women in Machine Learning & Data Science SLC Chapter - I have found that the best way to stay excited and up to date in the field is to be an active member of the community. I have been able to work efficiently across many different types of projects because of my diverse data science toolkit. I believe in using the best tool for the job, and am happy to learn new tools and technology to help drive projects forward. ⭐Programming Tools⭐ Python Apache Spark SQL Java C++ Pandas SciKit-Learn Tensorflow Tableau