Pricesearcher

» see pricesearcher.com as it lived, on the wayback machine

~ the obit ~

Samuel Dean started it in 2011 as a self-funded project with one goal, which was to index every product available to buy online. He hired freelancers off PeoplePerHour in the early years. Raja Akhtar, a PHP developer, joined in 2014 and worked on it in his spare time; a freelance DevOps engineer named Vlassios Rizopoulos came on in 2015 to speed up the indexing and became chief technology officer two years later. The public site launched in 2016 with the first outside seed money behind it.

The model was the part that made people look twice. Every other price comparison site in Britain charged retailers for the listing or took a commission on the sale. Pricesearcher charged nothing. A retailer uploaded a product feed in whatever format it already had, or simply let the crawler find the shop, and the products appeared for free. A search engine that indexes retailers the way Google indexes pages, with the billing problem deferred, is a coherent idea right up until the deferral runs out.

It ran out on 12 March 2020. Pricesearcher Ltd went into administration. Eight days later, on 20 March, Dean and Jack Sundt incorporated Pricesearcher Technology Group Ltd.

~ the rap sheet ~

~ one point one billion ~

By early 2018 the index held more than a billion products and the company was processing around 1.5 billion price changes a day on AWS. Search Engine Watch ran a piece that February headlined as the biggest search engine you have never heard of, which was accurate in both halves. Amazon and Argos were both in the index inside the first year, and so was most of the British high street behind them.

The research it published off the back of the index was genuinely useful and slightly melancholy: average product title, 48 characters. Average description, 90 words. Fewer than half of listings included shipping cost. Two in five gave no size or color. Between December 2016 and September 2017 it logged four billion price changes worldwide, and Britain changed its prices more often than anywhere else, once every six days.

~ the crawler ~

PriceBot did what its name says. It found online retailers and crawled their sites for anything for sale, and retailers could submit themselves to be crawled. By February 2018 it was pulling 2,500 UK shops that way, with another 4,000 pushing feeds in directly.

~ coming soon ~

The last two years read like a company doing everything right. September 2018, selected for the London Stock Exchange's capital-raising program. October 2018, Amazon UK's former head of pricing joins as chief analytics officer. January 2019, Google CSS partner. March 2019, a joint product with Pi Datametrics. July 2019, a deal powering price comparison for a discount platform with three million users. None of it was revenue, and none of it reached Turkey, where the shopping search I used had a different set of names on it.

The last full capture of the site in the Internet Archive is 11 October 2018. The front page carries the line 100% Unbiased Search Engine for Prices over twelve country subdomains running from Australia to Nigeria. In the corner there is a beta badge and a small banner promising a new site soon. Through 2019 the archive gets only redirects. After March 2020 there is nothing to capture.

The domain resolves today. It serves one line of text under the wordmark: Pricesearcher AI, coming soon, with an email address spelled out in words to dodge the scrapers. Six years on, the second company is still promising the new site the first one promised in 2018.

@}-->--

              )
             (_)
            .-'-.
            |   |
            |   |
            |   |
            |   |
          __|   |__   .-.
       .-'  |   |  `-:   :
      :     `---'     :-'
       `-._       _.-'
           '""""""

« previous  ·  random  ·  next »