Bark created a B2B Predictive model that steers their tROAS bidding to drive High Value Service Provider sign-ups.



Increase in LTV revenue


Increase in LTV ROAS


Bark is the UK’s largest & fastest-growing services marketplace, connecting consumers and businesses with the right professional for their projects in over 1,000 unique categories.

Founded in 2015 and headquartered in London, Bark operates globally in the UK, France, the US, Ireland, Canada, South Africa, Australia, New Zealand & Singapore.

Business Model

Bark’s business model consists of putting private individuals in contact with professionals from different sectors; cleaning, personal trainers, web design with more. Individual customers can post an advert on the Bark website, requesting a professional helper, free of charge. Bark then sends this advert to associated professionals in the surrounding area, who, if interested, acquire these leads with previously purchased credits, obtaining the email and telephone number of the individual who has published the advert.

In addition, professionals can optionally pay a monthly fee to Bark to appear as Elite Pro, which are professionals that Bark recommends.


Bark’s objective was to target its campaigns to attract professionals who use credits on a recurring basis.

Their primary pain was that the volume of conversions, professionals who clicked on the Search campaign and ended up buying credits within seven days, was so low that Google’s algorithms could not learn and, therefore, could not optimise for these users.


Making Science proposed to add an additional type of signal to the ones Bark already had. These new signals are generated when users/professionals subscribe, ensuring priority is given to new users, users who have never subscribed before. Our machine learning model assigns the user a value that coincides with the expected revenue in terms of credit consumption plus a monthly fee, in 90 days.

The project was activated in the US campaign and was achieved by multiplying the volume of signals (new professionals with potential recurring credit consumption) by 20 and tagging them based on their value. This allowed Google Search Smart Bidding campaigns, with tROAS bidding strategies (ROAS with expected value), to have enough data to learn and optimise towards the users/professionals that Bark had as KPIs (higher value prospects).


Implementing high-value customer signals in the Value-Based bidding for bid strategies permitted Bark to:

  • Increase 23% the 90 day LTV revenue vs 7 day revenue multiplier compare to the control
  • Increase 33% the 90 day LTV ROAS maintaining the 90 day LTV revenue