Sarah Raven is an online gardening retailer. We used machine learning to quickly identify where it should spend its limited budget and achieved a 194% increase in transactions.
The Sarah Raven website was not achieving its full revenue potential, largely due to a lack of traffic. We identified opportunities for boosting both category and product pages, but were faced with the additional challenge of a very limited budget so we needed to ensure that we made every penny count.
We decided to compare a range of pages to identify those that were statistically most likely to convert. With a huge amount of data available, limited time and restrictive budget, we used machine learning analysis to help us yield some quick gains.
We used four different algorithms to develop a machine learning model that could predict, with a 70% or greater degree of accuracy, the likelihood that a page would convert. The calculation was drawn from a correlation of large numbers of conversion indicative metrics. The analysis identified that the dahlias category page (rather than individual product pages) was the one to promote.
We improved the page meta data to better match buyer intent and improve the call to action. This was designed to encourage a higher click-through rate.
We quickly achieved significant increases in traffic, organic transactions, and revenue – a trend that continued for months afterwards. At the time of writing, click-through rates are 76% higher than previously, and transactions are up by 194%.
Without machine learning, it would not have been possible to achieve such impressive results in such a short space of time on a limited budget. We are happy to say that Sarah Raven has since expanded its relationship with Vertical Leap to also include PPC.