Case study

Convincing AI to help us make more money.

Creating a ML model to improve the conversion rate for property bookings.


To date approaches that the company has tried didn’t provide expected results. The extant A/B tests showed no significant gain between random selection and their tried approaches. They needed expertise in ML to drive the process in the right direction. To increase the conversion rate.


Two ML solutions were initially delivered for A/B testing. The one with the highest impact on conversion was elaborated. Codes and models. All shared with the team to be used in making predictions and enhancing effectiveness of the search engine's results. Significant improvement in ROI was shown for dedicated pages.

Technology & Tools

Models created in Python using libraries: tensorflow, keras, scikit learn, pandas; ML algorithms included xgboost, neural network, NLP and computer vision.


A London-based trusted online marketplace for travellers with over 60 000 holiday cottages, lodges and apartments. Home to the largest collection of holiday rentals in the UK.


An increase of the conversion rate feeds directly into the bottom line. The client’s goal was to receive a solution that’s delivered relatively fast and provides results almost immediately, so that they can grow and implement more projects with time.

Client intended on solving the problem with 2 potential solutions.


The first option involved a better selection of images (of higher quality). The second is search ranking optimization. Both were tested. 

The former solution focused on optimising the miniature images appearing to users, to potentially improve CTR with their attractiveness and subsequently the conversion rate. For this purpose several approaches were tested, including those based on regression and classification. 

The second option’s goal was optimising the ranking of the search results. The best features for the problem solution were selected by our data scientist and in delivery NLP was utilised together with a deep network to obtain optimal results. 

The second solution proved significantly more effective, therefore to most efficiently support conversion rates, we continued with this project. The goal was to optimise it for the company’s search options.

The implementation process went as follows: At first the focus was on getting a clear understanding of the problem being faced as well as the solutions that were already tried and tested. Part of the available data was provided for investigating this matter. Once the knowledge was acquired, the planning phase started. Result oriented options only. Our data scientist was providing the solutions and monitoring results delivered by the models. Once the solutions were investigated, the selected model was actively investigated. The data scientist monitored what went wrong and how to improve the model.

Download graphic


We boosted the conversion rate for several sites. Future upkeep ensured. We provided scripts and guidance to the client so that they can retrain models in the future by themselves. 

Better predictions. Now all the end clients can enjoy a more pleasant booking experience this Christmas season. Relevant results mean less friction. A win win.


Principal Engineer, Accommodation Marketplace

“I appreciated that they investigated several solutions and delivered them to us with an explanation.”

More case studies

Case study

Urgent project rescue.

Re-inventing the faulty solution for managing machine learning models.

learn more.
Case study

Fresh approach to an old problem

Monitoring the EU roaming – the application.

learn more.

Get in touch to see
how we can generate
value together

let's talk.
Join our clients