About Kardmatch
Kardmatch launched in 2008 as the first credit card comparison website in Mexico and has since helped more than 10 million Mexicans choose a card. Its blog, live since 2012, grew into the country’s largest credit-card-focused media property, with over a million monthly visitors and more than 150,000 subscribers. In 2018, the company launched Promos Kardmatch, an app that surfaces personalized promotions for existing cardholders. Founder and Managing Director Joel Cortes built the company around one mission: help Mexicans make better decisions when choosing and using their credit cards. Every part of that depends on accurate, current data pulled from dozens of bank websites.
Challenge
A credit card comparison platform runs entirely on other people’s data. Banks publish rates, fees, and offers on their own sites, in their own formats, and change them without warning. Building an accurate comparison layer on top of that created several problems:
Data trapped across bank sites:
Current rates, annual fees, rewards, and promotions had to be pulled from Mexico’s major issuers and refreshed often enough to keep every comparison accurate.
Scale beyond manual effort:
Promos Kardmatch alone needed weekly extraction from eight bank promotion pages covering roughly 1,000 individual offers, which was impossible to do by hand for a lean team.
Constant layout changes:
Bank sites redesign pages and restructure how offers are listed without notice. As Kardmatch put it, when you finally adapt to collect the information one way, they change the layout again. A script written against one version breaks the moment the page changes.
No spare headcount:
Every hour spent noticing, diagnosing, and rewriting broken scrapers was an hour not spent on the comparison experience, the blog, or the app.
A survey Kardmatch ran with 2,100 of its own users showed the stakes. Only 56 percent used their card promotions even twice a year, and of those who used them rarely, 40 percent were simply unaware the promotions existed. Closing that gap meant becoming the reliable layer between cardholders and dozens of bank sites, which made dependable data the foundation of the entire product.
Why PromptCloud
Kardmatch looked for a local Mexican partner first but could not find one offering a fully managed web scraping service at the scale it needed. After researching providers online and checking references from other customers, it chose PromptCloud, a partnership running since the last quarter of 2017. The decision came down to four things:
Fully managed service:
No in-house scripts to write or maintain, the one capability no local supplier could offer.
Scale across sources:
Weekly extraction from all eight bank promotion sites and their roughly 1,000 offers, handled continuously rather than in bursts of manual effort.
Maintenance absorbed:
Layout changes become PromptCloud’s problem to catch and fix in the background, not a task on Kardmatch’s roadmap.
Proven reliability:
Strong references from other customers, and a track record that has since held across nearly a decade of partnership.
PromptCloud’s fully managed web scraping pipeline gave Kardmatch exactly what the local market could not: current data from every bank source it needed, delivered on schedule, with no scraper code for the team to own. Data reliability moved off the product roadmap entirely.
Implementation
Where PromptCloud sits in the pipeline
PromptCloud runs the data acquisition layer beneath Kardmatch’s comparison platform and app:
Weekly bank data extraction:
Structured data from all eight bank promotion sources, delivered every week without Kardmatch touching any code.
Layout-change monitoring:
PromptCloud watches the source sites, catches redesigns early, and adjusts extraction logic in the background so the comparison data keeps flowing either way.
Feeding comparison and personalization:
That structured feed powers both the comparison site and Promos Kardmatch, which personalizes roughly 1,000 offers to how each cardholder actually uses their card.
Our objective was to extract weekly content from 8 promotion sites that contained around 1,000 offers, which was impossible to do manually. We looked for a supplier in Mexico, but we couldn’t identify someone who could offer a fully managed web data scraping service. So, with some research on the internet, we found PromptCloud.”
Kardmatch
Results
With weekly data collection handled, Kardmatch could operate and grow without firefighting broken scrapers:
Manual extraction eliminated:
No more hand-collecting offers across eight bank sites every week, removing a recurring task with no strategic upside.
Around 1,000 offers kept current:
Promotions tracked and refreshed on a predictable weekly cadence instead of an ad hoc one, keeping Promos Kardmatch accurate.
Engineering time redirected:
Team focus moved to the comparison platform, blog, and app rather than rewriting scrapers after every bank redesign.
A media property at scale:
A stable data foundation supported growth to over a million monthly blog visitors and more than 150,000 subscribers.
A new product launched:
Promos Kardmatch, which depended entirely on structured, current offer data, became possible to build and grow.
Get Started
Tired of scrapers that break every time a bank redesigns its site?
Get clean, structured web data delivered on your cadence from a managed pipeline built around your specific sources and schema.
• No contracts. • No credit card required. • No scrapers to babysit.
