Abstract: Significant breakthroughs have been made in precursor experiments for the exploration and development of deep coalbed methane, showing good prospect for development. However, deep coalbed methane has relatively high buried depth with strong heterogeneity. Lots of geological and engineering factors may affect the development results, and reasonable development countermeasures are still not determined. This paper is a case study of No. 8 coal seam in Ordos Basin, and analyzes its accumulation laws and development characteristics. The results show that No.8 coal seam has high maturity, stable distribution in the whole basin, and huge hydrocarbon generation potential; the average volume proportions of micropores, mesopores, macropores, and microfractures in deep coal reservoirs are 78.0%, 6.8%, 2.1% and 13.1%, respectively, as being a typical multiple pore-fracture system with superior hydrocarbon accumulation conditions; deep coalbed methane is located below the critical depth, characterized with a small-scale structural uplift, relatively tight reservoirs, undeveloped faults, weak hydrodynamic forces, and better preservation conditions. There is a high content of deep coalbed methane in the study area, which coexists with adsorbed gas and free gas. The coal structure is generally well developed, which is more conducive to reservoir stimulation by hydraulic fracturing. The production of gas wells quickly increases at early stag, characterized with high early production and rapid decline; according to desorption laws, the whole process can be divided into three development stages:free gas production, stable production, and decline. To address the challenges faced in the exploitation of deep coalbed methane, based on the experiences obtained during the development of tight gas and shale gas, four targeted suggestions are proposed:(1) initially applying 3D seismic technique; (2) establishing a reservoir geomechanical model; (3) building an industrialized large-scale well cluster construction mode; (4) keep doing precursor experiments. Finally, taking the Daning-Jixian block as an example, the understandings obtained from the precursor experiments have been summarized, will provide reference for further development of deep coalbed methane.