Takeoff Copenhagen are a specialist sneaker store, selling Denmark’s widest range of hyped sneakers from both their online and physical stores.
They recently switched from Bright Data to Browserless.io to help them with their scraping and offered up details for this case study.
Keeping up with fluctuating prices
Sneakers have become collectors items, with highly fluctuating prices. These prices constantly vary based on supply and demand, limited editions, country and a range of other factors. That makes it a challenge to both stay competitive will protecting their profit margins.
Takeoff use scraping automations to stay competitive, checking how each sneaker is priced on a variety of sites to get a feel for the market.
Limitations with the low-code scrapers
They started off with Bright Data's basic tools, but quickly wanted to do more than the limited site interactions.
Upgrading to the their “Scraping Browsers” meant they could run Puppeteer scripts. That seemed great, although they had to upgrade their Vercel package to deal with the slow 25 second fetch times.
The real kicker was when the proxies stopped running correctly.
It was essential that they only see European pricing, not how much the sneakers cost in America, Asia or Africa. But, the proxies stopped being regionally limited and were loading from anywhere in the world.
Unfortunately, the Takeoff developers spent weeks of being bounced between support teams while missing out on the data they wanted, so they looked at switching vendors.
From discovery to scrape in an hour
One of Takeoff’s team found Browserless, which appealed to them due to the low pricing and our scraping API. Using our headless browsers meant they could perform the necessary site interactions, but without having to pay a premium.
To check they wouldn't get the same support headaches, they then messaged our team and were happy to get rapid answers. He copied across his puppeteer code, and had a scraper running within an hour.
Fully migrating over was then straightforward, including our team configuring the European residential proxies.
5x faster scrapes with minimal errors, for 1/3rd the price
Their move has been a great success. Their automations:
- Now take under 5 seconds, a five times speed improvement from the Scraping Browser's 20-25s
- Cost around 1/3rd as much, not even including the Vercel savings
- Run more reliably, with a jump up to 99.5% success rate
- Can support more complex scripts, such as their plan to screengrab sites with errors
Needless to say, they are very pleased with moving to Browserless.
Try Browserless yourself with our 7 day trial
If you want to test how your scraping performs on Browserless, go ahead and check out our free trial.
Our team are happy to help you transfer your scripts across for testing on our system, or to help with speed and detection optimization.