How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

Client Overview

Lumen Global is a comprehensive energy efficiency partner that works with corporations, private equity firms, and real estate investment and management companies to lower energy expenses and create lasting value. Their services go well beyond LED upgrades, covering solar, circuit monitoring, ESG analysis, and compressed air systems. They invest their own capital into the process and deliver turnkey solutions, with the goal of creating a direct, measurable impact on client EBITDA and business valuation.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

The Results

The Results

The Results

Lumen Global went from weeks of incomplete manual research to a fully ranked, facility-level intelligence workbook delivered in hours. The system is also reusable, adding a new PE firm to the pipeline takes seconds, so the value compounds every time they run it.

96+

96+

96+

Facilities Profiled in the First Run Alone

18M+ sq ft

18M+ sq ft

18M+ sq ft

Building Footprint Mapped and Growing

30,000+

30,000+

30,000+

PE Firms the Pipeline Can Scale To

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

Client Overview

Lumen Global is a comprehensive energy efficiency partner that works with corporations, private equity firms, and real estate investment and management companies to lower energy expenses and create lasting value. Their services go well beyond LED upgrades, covering solar, circuit monitoring, ESG analysis, and compressed air systems. They invest their own capital into the process and deliver turnkey solutions, with the goal of creating a direct, measurable impact on client EBITDA and business valuation.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

The Results

The Results

The Results

Lumen Global went from weeks of incomplete manual research to a fully ranked, facility-level intelligence workbook delivered in hours. The system is also reusable, adding a new PE firm to the pipeline takes seconds, so the value compounds every time they run it.

96+

96+

96+

Facilities Profiled in the First Run Alone

18M+ sq ft

18M+ sq ft

18M+ sq ft

Building Footprint Mapped and Growing

30,000+

30,000+

30,000+

PE Firms the Pipeline Can Scale To

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

How Genta AI Built an Automated Market Intelligence Pipeline for an Energy Efficiency Company

Client Overview

Lumen Global is a comprehensive energy efficiency partner that works with corporations, private equity firms, and real estate investment and management companies to lower energy expenses and create lasting value. Their services go well beyond LED upgrades, covering solar, circuit monitoring, ESG analysis, and compressed air systems. They invest their own capital into the process and deliver turnkey solutions, with the goal of creating a direct, measurable impact on client EBITDA and business valuation.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Challenge

01

Lumen's team knew exactly what a good target looked like: an industrial facility with 100,000+ sq ft of space, running 40+ hours a week, paying above average electricity rates. The problem was finding and qualifying those targets at scale. Their process involved manually visiting PE firm websites to find portfolio companies, Googling each one for facility addresses, digging through county assessor records for building sizes, checking utility maps by zip code, and reviewing Google Maps one facility at a time to spot legacy lighting. For a pipeline covering 30,000+ PE firms, each owning multiple companies across multiple sites, that process took weeks and still produced an incomplete, already-stale picture. And there was no scoring system to tell the team where to focus first.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

The Solution

02

Genta AI built a fully automated market intelligence pipeline. The only thing Lumen's team needed to provide was a PE firm name and their website URL. Everything else happened automatically. The system scraped and structured portfolio company data directly from each PE firm's website. It then found physical facility addresses using Google Maps Places API and web search, returning coordinates for every site. Building square footage was calculated using a multi-source approach combining the Google Geocoding API, Shapely, Overture Maps, and Google Solar API, with a connected-component algorithm making sure multi-section industrial complexes were counted as one footprint, not several. Electricity rates were pulled by location through the EIA API. Google Street View images were then captured for each facility from multiple angles, and GPT-4 Vision analysed every image set for wall-pack fixtures, parking lot poles, and daytime lighting, the three main signals of legacy infrastructure and retrofit opportunity. Finally, every facility received an automated fit score from 1 to 10 based on all the data combined, so the sales team could immediately see where to focus. The output was a clean, five-tab workbook covering every PE firm, portfolio company, and facility, fully enriched and ranked, delivered in hours instead of weeks.

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

Technologies Used

03

• Google Maps Places API • Google Geocoding API • Google Street View API • Google Solar API • Overture Maps (Microsoft Building Footprint Data) • EIA API v2 • GPT-4o Vision • Python / Shapely • Custom AI Agents

The Results

The Results

The Results

Lumen Global went from weeks of incomplete manual research to a fully ranked, facility-level intelligence workbook delivered in hours. The system is also reusable, adding a new PE firm to the pipeline takes seconds, so the value compounds every time they run it.

96+

96+

96+

Facilities Profiled in the First Run Alone

18M+ sq ft

18M+ sq ft

18M+ sq ft

Building Footprint Mapped and Growing

30,000+

30,000+

30,000+

PE Firms the Pipeline Can Scale To

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.

We’re Here to Help

Ready to transform your operations? We're here to help. Contact us today to learn more about our innovative solutions and expert services.