Title: How to measure galaxy-galaxy-galaxy-lensing with higher precision and accuracy

Abstract: Galaxy-galaxy-galaxy lensing (G3L) is a powerful tool for constraining the three-point correlation between the galaxy and the matter field and thereby models of galaxy evolution. We propose three improvements to current measurements of G3L, designed to improve the precision and the accuracy by using the galaxies' redshifts and removing biases of the estimator. We further show how to account for lens galaxy magnification by the cosmic large-scale structure and how to convert the G3L signal from angular to physical scales. The improvements are tested on simple mock data and simulated data based on the Millennium Run with an implemented semi-analytic model of galaxies. Our improvements increase the signal-to-noise ratio by on average $35~\%$ at angular scales between $0.1'$ and $10'$ and physical scales between $0.02$ and $2 \, h^{-1}\,\textrm{Mpc}$. They also remove the bias of the G3L estimator at angular scales below $1'$, which was originally up to $40\,\%$. The signal due to lens magnification is approximately $10\,\%$ of the total signal.