Prospect identifier to Swimming Pool maintenance services

Swimming pool water.

Use Google Earth images to identify your leads. By using AI to analyze the images, TechRivo developed a prospect identifier for a swimming pool service provider. As a result, the client could know which residences are in need of pool maintenance and offer their services directly.

Use Case


Statistics shared by the Association of Pool and Spa Professionals (ASPS) suggest that only in the United States exists 10.4 million residential swimming pools. The competition for market share in this billion-dollar industry is high with few opportunities to innovate and differentiate the service.

Apart from construction, a big slice of the market value is linked with maintenance services. Furthermore, suggests that the annual maintenance costs for a single swimming pool can vary from $960 USD to $1800 USD.

A set of health problems caused by poor water quality can be associated with inadequate maintenance of swimming pools. Moreover, the lack of infrastructure maintenance increases the probability of accidents and the costs of resuming the use of the swimming pool afterwards. Thereby, there is an economic incentive to not neglect the maintenance of a swimming pool.

The challenge faced by the maintenance services providers relies on the identification of new prospects. Most of the time, their marketing initiatives are inbound and restricted. The lack of information does not allow them to target prospects with higher conversion chances. Resulting in a significant decrease in the return on investment of those marketing initiatives.


There are visible patterns through neglected swimming pools that can be seen in Google Earth images. For instance, the watercolour, water limpidity, aspect of the surfaces, and state of the surrounding areas (grass, floor state, etc) are indicators of a lack of maintenance.

The Google Earth app launched in 2001 created a vast number of opportunities. The platform provides good-quality satellite images of the entire planet, allowing their extraction and analysis for an inexpensive price.

The most recent advances in computer vision with machine learning algorithms enable the tracking of large quantities of data. Consequently, the analysis of large data sets of satellite images from swimming pools creates disruptive opportunities for the first technology adopters.


TechRivo created a platform to help an online business in the swimming pool market find new business opportunities.

The application collects data from Google Earth for selected geographies, analyses the relevant data points, and provides users with the address of swimming pools in need of maintenance.

Maintenance detector powered by Computer Vision.

Google Earth updates data in an average periodicity of 1 to 3 years. Thus, the algorithms rely not only on current images but also on the historical data of the properties to perform the classifications.

Tech aspects

The designed platform used Auto ML Vision to perform the image analysis and classification. The tool delivered extremely accurate results in the identification of neglected swimming pools.

The use of Google Earth as a source provides a business opportunity and an accurate prospect’s address to follow. In addition, this information is made available to the end-users through an integration with the client’s CRM (

Results of the Prospect Identifier

The business development using the built customized software showed disruptive results on prospect identification. In summary, by knowing exactly who is willing to buy the services, providers invested their time and resources in high conversion opportunities. Outbound marketing strategies rely entirely on the leads generated by the system and allowed the company to expand the business to new geographies.


The use of open-source data disrupted the customer acquisition process in the swimming pool niche. Indeed, these opportunities are overflowing throughout the internet to foster inexpensive proofs of concept and minimum value products. Examples of this open-source data are RODA (Registry of Open Data on AWS), governmental data, UNICEF dataset, UCI Machine Learning Repository, and DBpedia.

Savvy entrepreneurs excel in adapting the same ideas to a different context. Get to know some of those business ideas to entrepreneurs.