The UK online puppy trade has rapidly outgrown the current legislation, aided by the anonymity provided by classified advertisement platforms. In an effort to meet increased demand, some unregulated and regulated breeders may have employed practices that negatively impact canine welfare. A paucity of up-to-date empirical data, necessary to characterise the scale and nature of this industry, makes intervention challenging. This study quantifies the online puppy trade via web-scraped online classified advertisements, providing empirical data that reveal market trends, along with spatial and temporal patterns. A total of 17,389 unique dog advertisements were collated and analysed over a 2-year period (1 June 2018 to 31 May 2020). The second year included the COVID-19 Lockdown (23 March 2020 to 31 May 2020). Statistical comparisons between dependent and independent variables were performed by linear regression. In the case of a single continuous variable, a one sample t-test was used. Of these advertisements, 57.2% were sourced from a pet-specific classified advertisement website (Pets4Homes, n = 9948), and the remaining 42.8% from two general classified advertisement websites (Gumtree, n = 7149, 41.1%; Preloved, n = 292, 1.7%, respectively). England exhibited the greatest number of advertisements (n = 10,493), followed by Wales (n = 1566), Scotland (n = 975), and Northern Ireland (NI; n = 344). Scaled for estimated human population density, Wales possessed as many advertisements per million inhabitants (489.4) as the other three combined (England = 186.4, Scotland = 177.3, and NI = 181.1). Across both years, 559 unique breeds were advertised, yet 66% of all advertisements focused on 20 breeds, and 48% advertisements focused on only 10 breeds. Regional breed popularity was suggested, with French Bulldog as the most advertised breed in England (7.3%), Scotland (6.8%), and Wales (6.8%), but Schnauzers were most popular within Northern Ireland (6.83%). Within the 559 unique breeds advertised, only 3.4% had links to conformational disorders CD); however, these breeds were among the most commonly advertised, totalling 46.9% of all ads. Across all regions, price density peaked between GBP 300 and GBP 1000, with Bulldogs presenting the greatest cost (mean = GBP 1461.38, SD = GBP 940.56), followed closely by French Bulldog (mean = GBP 1279.44, SD = GBP 664.76) and Cavapoo (mean = GBP 1064.56, SD = GBP 509.17). CD breeds were found to be GBP 208.07 more expensive, on average, than non-CD breeds. Our results represent a buoyant online market with regional and seasonal fluctuations in price, advertised breed frequency and total counts. This suggests a market influenced by consumer trends, with a particular focus on breed preference, despite links to illness/disease associated with conformation. Our findings highlight the value of utilising online classified advertisement data for long-term monitoring, in order to assist with evidence-based regulatory reform, impact measurement of targeted campaigns, and legislative enforcement.
Keywords: animal welfare; canine welfare; conformational disorder; data science; puppy trade.