The Central Bureau of Statistics has announced that the next census, to be completed by 2021, will be the most comprehensive one yet, surveying not just households but also livestock, household amenities and community infrastructure. Moreover, instead of using a sampling method to infer larger trends in the population, the bureau plans to reach out to every household and collect data directly, thereby creating a clearer picture.

This is great news. A comprehensive survey will provide a snapshot of the country—how it looks and what its needs are—which is especially necessary as Nepal has just transitioned into a federal model. A census containing accurate microdata, data collected at the individual level, will help development agencies, the government at all levels, and the private sector cater to the needs of every region in a targeted manner. However, how the bureau carries out the census—from the designing of questions to the training of data collectors—will determine the data’s accuracy and usability.

Nepal is constitutionally obligated to conduct a census every 10 years. But due to cost constraints, and perhaps not understanding the need for microdata, the country in the past few censuses has been using the sample survey method. In the 2011 census, the bureau only surveyed 12 percent of the households in Nepal, and extrapolated trends and proportions regarding the population from the data. A national outlook at policy development, made by a centralised governance structure, meant that this method was adequate in the past. However, with provincial and local governments shouldering more responsibility and power in the federal system, the need for detailed microdata cannot be overstated.

The bureau’s plans to include questions about access to public and private goods, besides recording the physical and social infrastructure available, is a significant and positive decision. Such detailed statistics can help governments implement targeted policies and strategies. Not all localities and regions benefit from a similar policy approach. Such data can also help keep electoral constituency delimitation relevant. Nepal’s electoral constituencies are determined by taking into consideration population and geography, to be reviewed every 20 years. Moreover, this will help the government, as well as development agencies and non-governmental organisations, take a better approach to tackle issues such as education and health. Even budgetary allocations could become more efficient. Businesses, too, can design and develop products and services according to the needs of the local population.

The Central Bureau of Statistics has estimated the cost of the next census to be over Rs4.5 billion. This is a whopping figure, considering the last one only cost Rs1.5 billion. While the benefits of such a comprehensive collection of data, for the first time in Nepal’s history and coinciding with the establishment of federal Nepal, would ideally far outweigh the costs, questions regarding survey design and implementation remain. The people hired to conduct the census need to be properly trained.

The questions themselves need to be designed thoughtfully and scientifically. The bureau says it wants to accurately take into account minute details, such as the differentiation of residential and non-residential households. But in a country full of multi-use buildings, many with undeclared income sources, capturing factual data will be challenging. Moreover, while the bureau wants to be thorough, it has not even been able to decide whether to include more than two gender options—something that the constitution itself has guaranteed. The bureau must not waste this opportunity to capture Nepal accurately.