Purpose Gully erosion is the most serious form of soil degradation that occurs in agricultural areas. Gully head retreat is an essential process in gully erosion. In some respects, gully head activity reflects gully head retreat. This paper aims to present a method for quantifying gully head activity, understanding the effects of its influencing factors and analyzing the relationship between gully head activity and gully growth rates. Materials and methods Based on a field survey of 80 gully heads in five small watersheds, hilly-gully region of the Loess Plateau, this study introduced a gully head activity index (GAI) and calculated it with a semiquantitative method, which employs seven indicators. Theoretically, the GAI ranges from 1 to 3, and the larger the GAI is, the more active the gully head is. Pearson correlation coefficients (r) were used to evaluate the individual relationships between the GAI and upslope drainage area, upslope length, vegetation coverage, and slope aspects. One-way ANOVA was used to compare the GAI in different vegetation types, soil and water conservation measures (SWCM). Finally, Pearson's correlation was used to evaluate the relationships between gully growth and the GAI. Results The GAIs of 80 gully heads ranged from 1.122 to 2.943, with an average value of 1.765. Positive correlations with the GAI were found for both upslope drainage area and upslope length. There was a significant negative correlation between vegetation coverage and the GAI. No significant relationship was found between slope aspects and the GAI. In the drainage area, the average GAI values were 1.801, 1.895, and 1.457 for grassland, shrub-grass land, and tree-grass land, respectively. The average GAI values were 1.507 and 1.498 for terraces and fish-scale pits, respectively, compared with a GAI value of 1.826 when no soil and water conservation measures were taken. The average annual gully growth rate (A(g)) between 2007 and 2016 was significantly correlated with the GAI. Conclusion The GAIs of 80 gully heads reveal the relatively lower activity of the gully heads. Vegetation types and SWCM have positive effects in terms of reducing gully head activity. SWCM combined with proper tree species can effectively reduce gully head activity. Gully erosion processes are complex, and gully head activity can only explain part of the variation in gully growth.