Due to a range of constraints, the availability of biodiversity-related information varies considerably over space, time, taxa and types of data, thus causing gaps in knowledge. Despite growing awareness of this issue among scientists, it is still poorly known how, and whether, scientific efforts have contributed towards overcoming these information gaps. Focusing on spatial gaps in global biodiversity data, we show that accumulation rates of non-bird species occurrence records stored in the Global Biodiversity Information Facility have not improved, and have even potentially declined, over the past three decades in data-poor, often biodiversity-rich, regions. Meanwhile, one citizen science project, eBird, has been making a considerable contribution to the collection and sharing of bird records, even in data-poorest countries and is accelerating the accumulation of bird records globally. We discuss the potentials and limitations of citizen science projects for tackling gaps in biodiversity information, particularly from the perspective of biodiversity conservation.