The characteristic timescale of variations in the X-ray luminosity of accreting black holes is seen to be proportional to black hole mass and inversely proportional to accretion rate (in Eddington units). Therefore, monitoring of nearby AGN with low black hole masses, and with reasonably high accretion rates, is expected to trace variability processes that if scaled up to the most massive AGN might take >millenia. The nearby galaxy NGC4395 harbours a very low black hole mass AGN (only ~1e5 Msol) and is thus a prime target for study. Past observations of NGC4395 with ASCA, Chandra, XMM and Suzaku have revealed dramatic variations in X-ray flux and spectral shape. We have undertaken a year long monitoring campaign using the Swift satellite with the intention of tracing these variations over the course of a year. Our Swift dataset caught NGC4395 undergoing a month-long period of exceptionally high activity, peaking at roughly ten times the long term average 2-10 keV luminosity, before declining to a sustained very low state. The huge dynamic range in luminosity of NGC439 is accompanied by dramatic changes in X-ray spectral shape, indicating either that the intrinsic X-ray continuum shape is varying, or that we are seeing the nucleus through a screen of obscuring material having variable column and/or ionisation level. We test these possible explanations with a detailed time-resolved X-ray spectral analysis of the Swift-XRT data, together with archival XMM observations. I will show that the rapid X-ray spectral variations seen in NGC4395 cannot be explained easily by changes to the properties of a simple combination of neutral or warm absorbers, but that substantial changes in the spectrum of the intrinsic emission are also required.