Acceso usuarios
María Paz Moral Zuazo, María Pilar González Casimiro, María Beatriz Plaza Inchausti 
Purpose - Online advertising such as Google AdWords gives SMEs access to new markets at reduced costs. Our aim is to analyse the visibility and performance of a website and to test the effectiveness of online marketing using the data provided by Google Analytics.
Design/methodology/approach - We use a class of econometric time series models with unobservable components, Structural Time Series Models. We allow for time-varying trends to take into account the non-stationary behaviour displayed by time series. We illustrate our model using daily data from a local tourist website. Three specific questions are addressed: do paid keywords campaigns increase the volume and quality of search traffic? Do paid keywords affect the volume and quality of the unpaid traffic? How do paid and unpaid keywords perform? Findings - The results for our case study show that: 1) online campaigns affect traffic volume positively but their effectiveness on traffic quality is uncertain, 2) paid keywords do not affect the volume and quality of unpaid traffic, 3) the increase in traffic volume is not always due to the paid keywords and the lowest quality visits come from paid traffic.
Practical implications - This analysis may help webmasters to design successful online advertising strategies.
Originality/value - This study contributes to the development of user-friendly methodologies to monitor website performance. The analysis shows that Structural Time Series Models is a suitable methodology to test the effectiveness of online campaigns and to assess the changes over time in the performance of a website.
