Short term prediction of the magnitude, time, and location of earthquakes is currently not possible. In some cases, however, behaviour has been documented that has been retrospectively considered as precursory. Some models hold that on a timescale of several years, increasing levels of background seismic activity may signify enhanced damage generation affecting a broader area in the lead up to the future earthquake. Localization of seismicity and/or aseismic deformation towards a single fault or multiple fault branches in these models leads to spatial coalescence of segments, stress transfer and non-local interaction of foreshock events, and a nonlinear increase in seismic activity. Perhaps due to the structural complexity of fault zones, however, proposed seismic and aseismic preparatory deformation processes, when present, are strongly variable and still not well understood. Here we present evidence for an extended earthquake preparation process starting in June 2022 and lasting approx. 8 months prior to the occurrence of the February 6th, 2023, MW 7.8 Kahramanmaraş earthquake on the East Anatolian Fault Zone. The apparently precursory activity ahead of the earthquake is composed of a handful of isolated spatio-temporal clusters within 65 km of the future earthquake epicentre. Some of these clusters display accelerating seismic activity starting ca. 8 months before the mainshock, non-Poissonian inter-event time statistics and distribution of magnitudes in time, as well as low Gutenberg-Richter b-values. Close to the mainshock epicentre and during the weeks prior to its rupture, seismic quiescence is observed. Our observations suggest a different initiation mechanism compared to the cascade of close (<200 m) foreshocks observed before the MW 7.6 Izmit 1999 earthquake. The trends of seismic preparatory attributes for this earthquake follow those previously documented in both laboratory stick-slip tests and numerical models of heterogeneous earthquake rupture affecting multiple fault segments. With more comprehensive and effective earthquake monitoring, it may be possible to recognize a preparation phase before at least some significant earthquakes.