In this article, the electromagnetic (EM) field in gap-mode tip-enhanced Raman spectroscopy (TERS) is investigated theoretically and experimentally for a range of commonly used and unusual metal and nonmetal substrates. By approaching a metal tip to a substrate, both form a coupled system that confines the EM field created at the tip apex. The influence of the substrate onto the EM field enhancement is observed in a top-illumination gap-mode TERS setup for different metal substrates. These include Au, the most commonly used substrate, and also a wide range of rarely or previously unused TERS substrates (Cu, Ag, Al, Pd, Pt, Ni, Ti, Mo, W, stainless steel, Al2O3, SiO2). Self-assembled monolayers of thiols and brilliant cresyl blue thin film samples are investigated experimentally on nine metal substrates, all showing considerable TERS enhancement. With finite difference time domain and finite element simulations used, the article provides a good estimate of the EM field enhancement for a wide range of substrates for users to estimate how well a substrate of choice will perform in a gap-mode TERS experiment. The reduction in EM field strength |E2| compared with Au is less than an order of magnitude for many metals (Calculations: Cu 92%, Ag 81%, Ni 53%). This article experimentally shows that a wide variety of conductive substrates can be used, when one is willing to trade a fraction of the EM field enhancement. TERS was seen on all metal substrates including stainless steel, yet quantification was not always possible. These qualitative results were complemented with intensities from calculations. The wider variety of substrates will increase the applicability of TERS and evolve it one step further towards use in standard analytics. Copyright © 2012 John Wiley & Sons, Ltd.