The UK's Office for National Statistics (ONS) is advancing its methods of collecting price data to offer more accurate and up-to-date inflation measures. Traditionally, the ONS deployed price collectors to supermarkets monthly, but it is pivoting towards utilizing algorithms to gather extensive pricing information. This new method involves scraping data from 100,000 prices daily, drawn from seven major UK supermarkets, which together account for 80% of grocery sales.
This innovation supports the development of a real-time grocery inflation index, facilitated by web scraping technologies and a large language model to categorize products accurately. Richard Davies and Finn McEvoy are spearheading this project, aiming to publish daily grocery inflation data and assess supermarket pricing strategies in near real-time.
This new system allows for a timely analysis of sudden market changes, such as those caused by Brexit or new grocery taxes, providing swift insights for the Bank of England and other stakeholders. Despite earlier challenges in categorizing products accurately, the system's current results closely align with official data, offering promising prospects for faster, more responsive inflation tracking.
The ONS plans to integrate scanner data from store checkouts into the official Consumer Price Index (CPI) by March of the following year, potentially providing even more detailed and accurate measures of consumer price inflation.