Industry Expertise

CPG & Consumer Behavior

Consumers expect more personalized and curated products, content, and offers. Brands require more insight from the market and buyers to deliver the desired consumer experience. The data science needed to provide a personalized and curated experience continues to have an increasing impact on retail and CPG. For companies who do this well, this can be a source of competitive advantage. Our solutions have included automated taxonomy generation, detection of product sentiment, and AI guided campaigns and offers.

Data science to recommend CPG campaigns to targeted audiences to maximize ROI and sales lift

Train machine learning models to predict consumer purchase behavior to optimize CPG campaigns

Read more

Data science to recommend CPG campaigns to targeted audiences to maximize ROI and sales lift

Train machine learning models to predict consumer purchase behavior to optimize CPG campaigns

The Challenge

Our client, a digital marketing platform, enables CPG brands to reach customers via retail locations. The brands need predictive insights on which campaigns are most likely to generate sales lift and/or ROI, based on past behavior.

Outcome

We aggregated point of sale data from 6,000+ locations over a three-year period to identify campaigns based on price discounts. Seasonal sales trends were modeled to identify true sales lift for the campaign. Campaign costs were estimated to generate a campaign ROI / margin. Ensemble machine learning methods were used to learn past campaign behavior to predict sales lift and ROI. Our client saw a 4x increase in sales leads from a major industry conference based on the newly introduced campaign insights.

Read more
Data science to recommend CPG campaigns to targeted audiences to maximize ROI and sales lift
The Challenge

Our client, a digital marketing platform, enables CPG brands to reach customers via retail locations. The brands need predictive insights on which campaigns are most likely to generate sales lift and/or ROI, based on past behavior.

Our Solution

We aggregated point of sale data from 6,000+ locations over a three-year period to identify campaigns based on price discounts. Seasonal sales trends were modeled to identify true sales lift for the campaign. Campaign costs were estimated to generate a campaign ROI / margin. Ensemble machine learning methods were used to learn past campaign behavior to predict sales lift and ROI. Our client saw a 4x increase in sales leads from a major industry conference based on the newly introduced campaign insights.

Develop predictive models to power the sales process

Use data science to identify the highest value sales targets

Read more

Develop predictive models to power the sales process

Use data science to identify the highest value sales targets

The Challenge

Our client is a provider of hotel-inspired services for multifamily communities. When engaging with a new community, their sales team needed to focus and personalize their marketing efforts on the residents most likely to become customers. They needed a data-driven way to identify the residents most likely to become high revenue customers. 

Outcome

We aggreated multifamily community data and historical revenue. We employed data-science-driven clustering analysis to identify common traits of high revenue customers and trained a model to predict sales outcomes for new communities. Our analysis confirmed that users of our client’s services were much more likely to re-sign the lease, conferring significant savings to multifamily community owners on churn, leading to an acceleration of sales.

Read more
Develop predictive models to power the sales process
The Challenge

Our client is a provider of hotel-inspired services for multifamily communities. When engaging with a new community, their sales team needed to focus and personalize their marketing efforts on the residents most likely to become customers. They needed a data-driven way to identify the residents most likely to become high revenue customers. 

Our Solution

We aggreated multifamily community data and historical revenue. We employed data-science-driven clustering analysis to identify common traits of high revenue customers and trained a model to predict sales outcomes for new communities. Our analysis confirmed that users of our client’s services were much more likely to re-sign the lease, conferring significant savings to multifamily community owners on churn, leading to an acceleration of sales.

Understand how discounts affect consumer behavior

Analyze purchase history trends to determine the impact of discount campaigns on customer visits and purchases

Read more

Understand how discounts affect consumer behavior

Analyze purchase history trends to determine the impact of discount campaigns on customer visits and purchases

The Challenge

Our client is a Fortune 500 consumer product goods (CPG) company. The client’s retail partners had suggested that BOGO offers may alter customer behavior, resulting in fewer trips with less overall revenue to the retail stores. Our client needed to understand the impact of discount campaigns on traffic and basket size to be able to continue engaging their retail partners to support their product campaigns. 

Outcome

We aggregated basket size and visits per retail location. We identified historical price discount campaigns from row level POS data across many retail locations. We evaluated store performance as a result of the campaign, with consideration for seasonal and general trends. We applied data science techniques to determine if certain discount types resulted in reduced basket size or store visits. Our analysis confirmed that campaigns of all types are beneficial to traffic and basket, confirming the CPG overall-marketing campaign strategy.

Read more
Understand how discounts affect consumer behavior
The Challenge

Our client is a Fortune 500 consumer product goods (CPG) company. The client’s retail partners had suggested that BOGO offers may alter customer behavior, resulting in fewer trips with less overall revenue to the retail stores. Our client needed to understand the impact of discount campaigns on traffic and basket size to be able to continue engaging their retail partners to support their product campaigns. 

Our Solution

We aggregated basket size and visits per retail location. We identified historical price discount campaigns from row level POS data across many retail locations. We evaluated store performance as a result of the campaign, with consideration for seasonal and general trends. We applied data science techniques to determine if certain discount types resulted in reduced basket size or store visits. Our analysis confirmed that campaigns of all types are beneficial to traffic and basket, confirming the CPG overall-marketing campaign strategy.

Automate taxonomy construction and product classification

Develop an algorithm to transform 2.3 Tb of text for ~8.5 million unique products into a hierarchical taxonomy and unified naming convention

Read more

Automate taxonomy construction and product classification

Develop an algorithm to transform 2.3 Tb of text for ~8.5 million unique products into a hierarchical taxonomy and unified naming convention

The Challenge

Our client, a grocery and retail pricing platform, collects billions of pricing records from 100s of online and brick & mortar retailers to provide sales and pricing insights and forecasts to consumer product goods (CPG) brands. Our client needed a way to automatically classify and organize products, in particular, to be able to automatically identify similar products, even if named or described differently. 

Outcome

We trained a deep learning, natural language processing (NLP) model on 2.5 TB of text including product name, description and store categories for 8.5 million products, leveraging word vectors to auto-generate a taxonomy. We translated to a hierarchical taxonomy and unified naming convention. The client software engineering team has incorporated the taxonomy and unified naming convention into their platform, and has deployed data engineering pipelines to pre-process text for input into the deep learning model to classify new products.

Read more
Automate taxonomy construction and product classification
The Challenge

Our client, a grocery and retail pricing platform, collects billions of pricing records from 100s of online and brick & mortar retailers to provide sales and pricing insights and forecasts to consumer product goods (CPG) brands. Our client needed a way to automatically classify and organize products, in particular, to be able to automatically identify similar products, even if named or described differently. 

Our Solution

We trained a deep learning, natural language processing (NLP) model on 2.5 TB of text including product name, description and store categories for 8.5 million products, leveraging word vectors to auto-generate a taxonomy. We translated to a hierarchical taxonomy and unified naming convention. The client software engineering team has incorporated the taxonomy and unified naming convention into their platform, and has deployed data engineering pipelines to pre-process text for input into the deep learning model to classify new products.

Automatically extract & organize information on 250,000+ food products

Use machine learning algorithms to extract and organize food product ingredients from 2 million+ product label images and classify products into 2,000+ categories

Read more

Automatically extract & organize information on 250,000+ food products

Use machine learning algorithms to extract and organize food product ingredients from 2 million+ product label images and classify products into 2,000+ categories

The Challenge

Our client, a food product data transparency platform, needed to automate the pipeline of data ingestion and quality control of ingredients, brand name, and nutritional facts for food products to ensure food product claims match product ingredients. Data was stored in more than 2 million images of product labels, and they were receiving data on more than 15,000 products per week. The client had OCR algorithms to parse the product label images into text.

Outcome

Our team trained a specialized deep learning, natural language processing model to classify and cluster 250,000+ unique products into 2,000 categories of aisle, shelf, and food type. We partnered with the client engineering team to incorporate the data engineering pipelines and classification and clustering algorithms into their internal platform. 

Read more
Automatically extract & organize information on 250,000+ food products
The Challenge

Our client, a food product data transparency platform, needed to automate the pipeline of data ingestion and quality control of ingredients, brand name, and nutritional facts for food products to ensure food product claims match product ingredients. Data was stored in more than 2 million images of product labels, and they were receiving data on more than 15,000 products per week. The client had OCR algorithms to parse the product label images into text.

Our Solution

Our team trained a specialized deep learning, natural language processing model to classify and cluster 250,000+ unique products into 2,000 categories of aisle, shelf, and food type. We partnered with the client engineering team to incorporate the data engineering pipelines and classification and clustering algorithms into their internal platform.