community
cancel
Showing results for 
Search instead for 
Did you mean: 

Past Analytics Excellence Awards

Suggest an idea

PATH logo.png

Author: Jeff Bernson, Senior Director, Results, Management, Measurement and Learning 

Company: PATH 

Team Members: Mike Hainsworth, Allan Walker, Doug Morris, Anya A’Hearn, Philip Riggs, Chris DeMartini, Jonathan Drummey, Jeff Bernson, Daniel Bridges, Joe Mako, Ken Black

 

Awards Category: Best 'Alteryx for Good' Story

 

PATH is the leader in global health innovation. We are driven by an unshakeable commitment to health equity and a bold belief in the power of innovation to improve health, save lives, and transform the future, especially for women and children. Working with partners across the public and private sectors, our trusted experts bring together the right tools and people at the right time and place to accelerate innovative solutions to the health challenges facing vulnerable women, children, and communities worldwide. By tenaciously supporting solutions all the way from early concept to large-scale use, PATH’s work translates smart ideas into tangible change for millions of people.

 

Describe the problem you needed to solve 

In Africa, malaria is the leading cause of death in young children. For the past decade, the Zambian National Ministry of Health and the National Malaria Elimination Centre (NMEC) with support from PATH have used data to fight malaria—saving thousands of lives. In 2015, PATH teamed up with Alteryx and six other technology partners to launch Visualize No Malaria—a campaign focused on integrating new tools and systems for data use to support Zambia’s Ministry of Health in their effort to eliminate malaria by 2021.

 

The PATH-supported malaria laboratory in Lusaka, Zambia, located at the NMEC is implementing a range of molecular tools in the quest to achieve elimination. Every month, thousands of blood samples from around the country are processed to answer key questions around issues such as diagnostic efficacy, drug resistance, and transmission intensity.

Prior to our partnership with Alteryx, the lab was ill equipped to deal with these data. At the time, our laboratory information system was costly, cumbersome and necessitated building a custom database for each study. We needed a flexible, agile tool that could grow with our needs, namely, a system that would provide a simple mechanism for reporting data, aggregating it, and producing summary outputs.

 

Describe the working solution

Alteryx software has transformed how we access data. As we experimented with Alteryx, it quickly became apparent that we could almost instantly curate, process, and archive large amounts of data to inform our field activities and to better understand what’s really happening with malaria transmission patterns and dynamics in near real-time.

 

Our partnership will empower frontline health workers with the critical tools to prevent, track, and treat malaria cases to help eliminate this deadly disease.  We are improving data accuracy and making real-time critical data-informed decisions about how and where to tackle outbreaks. We are also building the skills of district and facility health teams to combat the disease at community level. If successful, this model could serve to inform global efforts to end malaria for good. 

 

PATH and its data science partners are using Alteryx in a number of ways to help lead the charge toward elimination including:

 

1. Automating workflows that help us process lab results and notify health workers of positive malaria cases faster in targeted study sites. Prior to the introduction of Alteryx, we essentially were unable to provide feedback on results in any meaningful way. Now instead of waiting months, we can process results on a daily basis with automated outputs sent to the end-user for dissemination and to inform decisions.

 

PATH Workflow 1.png

 

2. Responding to survey data during field data collection. Survey data is usually evaluated at the end of the process, which is fine if everything has gone well, but potentially catastrophic if errors were introduced. Through Alteryx, we have been able to access survey data as it is collected through a cloud API and then validate and summarize data to ensure that we are on track. The speed of this process meant that we were able to not only identify and resolve issues during the survey, but could rapidly and iteratively interrogate the data to ensure high data quality.

 

PATH Workflow 2.png


3.
 Developing a workflow to enable automated prediction and forecasting of malaria cases in the Southern Province of Zambia that can be batched on a weekly schedule.

 

PATH Workflow 3.png

 

 

4. Leveraging Alteryx for geospatial data processing and analysis to create modeled boundaries such as facility catchment areas, entity relationship, and resource optimization diagrams, such as health worker to Facility network diagrams, and input layers for other geographic processing and rendering tools, such as Mapbox.

 

PATH Mapbox.png

 

5. Pre-processing and merging of geospatial and epidemiological data for ease of use in front end rendering and analytic tools, such as Tableau.

 

PATH Tableau Viz.png

 

 

Describe the benefits you have achieved

The net result from this work has been a more efficient workflow process, more insightful analyses, and the development of a more holistic approach to understanding the complete malaria picture in Southern Province. Alteryx put the power in the hands of the user so we can build exactly what we need when we need it.

 

Since deployment of these tools, along with a number of strategies to further accelerate the gains against malaria, since 2014 there has been a 90 percent decrease in malaria cases in the Southern Province. As transmission rates decrease and malaria cases become increasingly localized, combining the existing interventions with capabilities of data analytics tools like Alteryx will play an essential role in combating the disease at the community level and bringing the number of cases to zero.

PATH Benefits 1.png

Author: Andrew Simnick, Head of Strategy

Company: Art Institute of Chicago

 

Awards Category: Best 'Alteryx for Good' Story

 

The Art Institute of Chicago is one of the world's greatest art museums, staying true to our founding mission to collect, preserve, and interpret works of art of the highest quality from across the globe for the inspiration and education of our audiences. Today we face new competition for visitor attention, a continued responsibility to expand our audiences, and an increasingly-challenging economic environment.  Alteryx has allowed us to quickly overcome our data and resource constraints, develop a deeper understanding of our local audiences, and strike a balance between mission- and revenue-driven activities to continue to deliver on our mission for Chicago.

 

Describe the problem you needed to solve 

We, as do other museums, face the challenges of growing our audience while  maintaining a strong financial foundation. Our strategy to navigate this has been to increase visit frequency from our core visitor segments in the near term and use this increase to further expand outreach to new local audiences. However, our challenges to achieving this have been three-fold. First, visitor segmentation in the arts and culture space is a relatively recent concept, and general segmentation schema are not always applicable to Chicago at a granular level. Second, we have very useful data but in inconsistent formats, ranging from handwritten notes and Excel documents to normalized but disconnected databases. Third, we are resource-constrained as an institution and cannot dedicate large amounts of time or money towards dedicated analytics or external consulting.

 

Describe the working solution

First, we built a database describing the Chicago CBSA at the census block group level, providing the nuance necessary for a city where demographics change block-to-block and limit the utility of ZIP code analysis. Alteryx allowed us to get to this new additional level of detail and make our analysis relevant to Chicago. Using the Allocate Input and Calgary Join, we applied information from the US Census as well as Experian data sets. We utilized basic data such as population, income, and education, as well as proprietary Experian segments such as Mosaic groups and ISPSA (Index of Social Position in Small Areas) to describe these census blocks.

 

Second, we brought together our disparate visitor data into a blendable format. Some of our datasets are well defined, such as our membership CRM which resides in a relational database on MSSQL Server, whereas others are more ad hoc, such as our Family Pass users, which are transcribed from pen and paper into an Excel document. The Join tools in Alteryx provided a simple way to bring these data together without commanding significant time from our small analytics team.

 

Third, each of these datasets was CASS Encoded and geocoded using the US Geocoder tool, providing us a spatial object. We then utilized the Spatial Match tool to find the intersection of these objects with our universe of Chicagoland Block Groups. Each of these distinct streams were then normalized and combined to the block group aggregation level resulting in our final dataset. We also utilized a shared public custom macro which allowed us to convert these block groups into polygons for visualization in Tableau.

 

Finally, we utilized heatmaps and scatterplots to identify which proprietary Experian segments correlate with our different offerings. This informed our choice of variables for our final Decision Tree tool analysis, which identified prime target block groups associated with our different offerings. These bespoke segments created via machine learning were more applicable to our own audiences and required a fraction of the time and cost of other segmentation methods.

 

Describe the benefits you have achieved

This approach has given us a framework and the supporting intelligence from which to make institutional decisions surrounding visitor outreach and programming, allowing us to focus our resources on actions which we believe will have the greatest impact towards increased participation, attendance, and/or revenue. For example, we can now tailor membership messaging more effectively and quantify the effects on repeat visitation. We also can identify gaps in our geographic coverage of Chicagoland and test different outreach efforts to engage new audiences. Most importantly, we can unify our approach to audience development across departments using a common baseline and methodology. These combined efforts enabled by Alteryx will help us to build our audiences and fulfill our civic responsibilities well into the future.

Author: Andrew Simnick, Head of Strategy

Company: Art Institute of Chicago

 

Awards Category: Best 'Alteryx for Good' Story

 

The Art Institute of Chicago is one of the world's greatest art museums, staying true to our founding mission to collect, preserve, and interpret works of art of the highest quality from across the globe for the inspiration and education of our audiences. Today we face new competition for visitor attention, a continued responsibility to expand our audiences, and an increasingly-challenging economic environment.  Alteryx has allowed us to quickly overcome our data and resource constraints, develop a deeper understanding of our local audiences, and strike a balance between mission- and revenue-driven activities to continue to deliver on our mission for Chicago.

 

Describe the problem you needed to solve 

We, as do other museums, face the challenges of growing our audience while  maintaining a strong financial foundation. Our strategy to navigate this has been to increase visit frequency from our core visitor segments in the near term and use this increase to further expand outreach to new local audiences. However, our challenges to achieving this have been three-fold. First, visitor segmentation in the arts and culture space is a relatively recent concept, and general segmentation schema are not always applicable to Chicago at a granular level. Second, we have very useful data but in inconsistent formats, ranging from handwritten notes and Excel documents to normalized but disconnected databases. Third, we are resource-constrained as an institution and cannot dedicate large amounts of time or money towards dedicated analytics or external consulting.

 

Describe the working solution

First, we built a database describing the Chicago CBSA at the census block group level, providing the nuance necessary for a city where demographics change block-to-block and limit the utility of ZIP code analysis. Alteryx allowed us to get to this new additional level of detail and make our analysis relevant to Chicago. Using the Allocate Input and Calgary Join, we applied information from the US Census as well as Experian data sets. We utilized basic data such as population, income, and education, as well as proprietary Experian segments such as Mosaic groups and ISPSA (Index of Social Position in Small Areas) to describe these census blocks.

 

Second, we brought together our disparate visitor data into a blendable format. Some of our datasets are well defined, such as our membership CRM which resides in a relational database on MSSQL Server, whereas others are more ad hoc, such as our Family Pass users, which are transcribed from pen and paper into an Excel document. The Join tools in Alteryx provided a simple way to bring these data together without commanding significant time from our small analytics team.

 

Third, each of these datasets was CASS Encoded and geocoded using the US Geocoder tool, providing us a spatial object. We then utilized the Spatial Match tool to find the intersection of these objects with our universe of Chicagoland Block Groups. Each of these distinct streams were then normalized and combined to the block group aggregation level resulting in our final dataset. We also utilized a shared public custom macro which allowed us to convert these block groups into polygons for visualization in Tableau.

 

Finally, we utilized heatmaps and scatterplots to identify which proprietary Experian segments correlate with our different offerings. This informed our choice of variables for our final Decision Tree tool analysis, which identified prime target block groups associated with our different offerings. These bespoke segments created via machine learning were more applicable to our own audiences and required a fraction of the time and cost of other segmentation methods.

 

Describe the benefits you have achieved

This approach has given us a framework and the supporting intelligence from which to make institutional decisions surrounding visitor outreach and programming, allowing us to focus our resources on actions which we believe will have the greatest impact towards increased participation, attendance, and/or revenue. For example, we can now tailor membership messaging more effectively and quantify the effects on repeat visitation. We also can identify gaps in our geographic coverage of Chicagoland and test different outreach efforts to engage new audiences. Most importantly, we can unify our approach to audience development across departments using a common baseline and methodology. These combined efforts enabled by Alteryx will help us to build our audiences and fulfill our civic responsibilities well into the future.