Many of you already know that FMTC collects data from networks, which is then processed by our data engineer team. The result is thousands of deals being available in our deal bank. But wouldn’t you like to know more details about how your deals get from point A to point B? I’d like to walk you through the process and provide tips to help your offers reach our feeds quickly with the most accurate information.
To aggregate the deals, we use the technology of multiple custom software programs. This technology allows us to:
- Gather data from multiple sources
- Detect dates, prices, deal types and merchant details
- Identify issues with formats, links, and spelling
- Many of these are automatically fixed, and the remaining issues are fixed by the Data Engineering team.
- Categorize content
- Remove expired deals
- Format links to include Affiliate ID
- Match offers with Skimlinks URLs
Manual Data Processing
After the automated processing is complete, the deals then make their way to our queues where they are processed by our dedicated data engineering team. There are different queues that sort the data for different types of processing. A few of the main ones are:
- Coupon Queue – All deals with a coupon code go to this priority queue. Here, each code is tested to be sure it is valid, and the proper discount is given.
- Deal Queue – All deals without coupon codes go to this queue.
- Change Queue – If a deal has already been added to our system, but a change is applied at the network, our system will detect that change and the deal will then go to this queue to be updated.
- Review Queue – This queue has deals that were processed before they went live. Data engineers double check all deals in the review queue to make sure everything is working properly.
In each queue, you can be sure our DE team is testing links to make sure they go to the right landing page, fixing spelling errors, tagging deals with correct deal types, categorizing each deal, and standardizing labels to provide clean and consistent data.
If there is ever an issue when testing deals, we have a special hold queue to place deals while DE team members reach out to program managers to inform them of the issue. Once the deal is fixed, or the right information is received, we then process the deals as usual. To avoid deals being delayed from having to contact program managers, please ensure all the data and links are accurate and include important information like dates, codes, and terms.
All of this processing is extensive work that takes an average of 1,300 hours each month by our Data Engineering team. You can be sure your deals will be tested for accuracy and processed efficiently by our data team, especially when you provide the best information at the networks for each deal. Now you know how we process deals to achieve the finished product of standardized content that can be used by numerous publishers.
Thanks for reading and have a great week!