Early detection and effective treatment are crucial for tuberculosis control, but global case detection rates remain low. The diagnosis of paediatric and extrapulmonary disease is problematic and there are, as yet, no rapid screening tests to assist active case finding in the community. Progress has been made in clinic-based detection tools with the introduction of Xpert MTB/RIF, a nucleic acid amplification test that combines sample processing and analysis in a single instrument to provide a diagnostic result and detection of resistance to rifampicin in under 2h. Enthusiasm for Xpert MTB/RIF has been high and global rollout has been facilitated by donor agencies. However, concerns remain about access and sustainability due to the high cost and infrastructure requirements. Although more sensitive than smear microscopy, early studies suggest the impact of the new test on case detection rates and patient survival has been limited. Alternative technologies are being developed, including non-sputum-based tests to assist the detection of extrapulmonary disease. Evaluation studies are needed to provide evidence of the impact of the new technologies on patient outcomes. This will enable appropriate placement of new diagnostic products in the healthcare system to support the control and eventual eradication of tuberculosis disease.