The injection of water into geothermal systems is an important procedure required to recover subsurface water resources and enhance permeability for increasing the reservoir volume. The injected water often leads to microseismic events during migration, which can be used to directly track the location of the injected water. However, in rare cases, unexpectedly large induced seismicity occurs after the injection termination. For risk control, understanding the differences between cases that cause post‐termination seismicity and those that do not is necessary. For this purpose, we used microseismic monitoring to examine the behavior of water during two injection tests, including their post‐termination periods, in Okuaizu geothermal field, Japan. In this field, a new remote microseismic cluster, apart from the injection well, was created in the post‐termination period of the first injection test. However, this cluster was not well activated in the second injection test. As a result, we revealed that this microseismic cluster was created on a structure that was different from the target fracture of the injection, possibly owing to pore‐pressure migration in the post‐termination period of the first injection. Its inactivation in the second post‐termination period may be attributable to the lower magnitude of pore‐pressure migration derived from the smaller amount of injected volume compared with that of the first injection test. The lower pore‐pressure migration was insufficient to reactivate the seismicity. We concluded that the occurrence of seismicity after injection termination may depend on the magnitude of pressure in the injection well at the shut‐in time. The Kaiser effect (i.e., a fault is not reactivated by stress that is less than the maximum stress loaded previously) could explain the observed phenomena.