Abstract The last uranium cycle started in late 2004 with an unprecedented rise in the uranium spot price from a long-established spot price of <US$20. This period corresponded to a massive increase in exploration spending. More than 50 major uranium deposits were explored and extensively drilled and about 20 of these could produce uranium in the next 5 years. Despite this enormous increase in spending, this last boom led to limited grassroots exploration success, with 14 new discoveries in the Athabasca basin, but only two significant new discoveries elsewhere. Much exploration, particularly junior company activity, concentrated on brownfield areas with delineation of 43-101 or JORC compliant resources at sites with previously known mineralization. In many cases, leases had been allowed to lapse after the uranium price collapsed in the early 1980s. Thus many projects with defined resources were available through acquisition and, in some cases, by staking open ground. On a more positive note, the availability of financing during this period allowed several companies, such as Paladin and Uranium One, to fast track deposit development in less than 5 years (e.g., Langer Heinrich, Kayelekera, South Inkai). In terms of development of new genetic models and research on uranium deposits, activity was limited because with the flurry of new activity, explorationists had little time to advance models and conduct relevant research as most were too busy relearning or reinventing older and existing concepts. Interest by academia, as well as industry support for academic research in uranium mineralized systems, also waned, with a few notable exceptions and the cycle was a little short to rekindle interest. The one exception is the progress in understanding of the Athabasca basin ore genesis, where research kept pace with the times, and a rigorous and well-constrained model was established and then very successfully applied, leading to a rash of new discoveries in the past 15 years. Unfortunately, applying this model to other areas with unconformity-related uranium potential has been much less successful. There were also some major advances in the understanding of sandstone- and volcanic-hosted uranium deposits, mainly as a result of the newly available literature and knowledge from Russia, Kazakhstan, Slovakia, China, and Mongolia. Volcanic-hosted deposits with new critical information for model development included the Streltsovskoye, Xiangshan, Johodna, and Saddle Hills and/or Dornod deposits, and a greater understanding of roll-front sandstone-hosted mineralization resulted from new information on uranium deposits in Kazakhstan. Rigorous application of these updated metallogenic models has just begun. Except for these deposit types, uranium research and the model development have not advanced substantially in the past 15 years. The exploration and mining business will remain cyclical, so now is the best time to prepare for next high-price cycle(s). New research efforts focussed on developing process-driven models that stress the mechanisms of uranium mobilization from source, and its transport and accumulation, are required for all uranium metallogenic models, but particularly volcanic, sandstone-, and calcrete-hosted deposit types, to give explorationists in the next cycle the edge to explore efficiently and effectively. It is important to develop these models to provide a solid geologic context for exploration programs, as development of new geophysical techniques (e.g., improved airborne electromagnetic EM, radiometrics and gravity techniques, three-dimension seismic surveys, downhole instruments for downhole Prompt Fission Neutron (PFN) detection, Induced Polarization (I.P.), and gravity) and geochemical techniques (e.g., partial extractions techniques, handheld field X-ray fluorescence (XRF) units) is continuing at a rapid pace.