by Luca Passani, CTO @ScientiaMobile
In the first draft of this blog post, I explained how real-time auctions typically take place in your browser each time you, as a web user, visit one of the many high-traffic websites that rely on ads to monetize their content.
I ended up scratching the whole paragraph explaining Prebid.js. After all, if you’re reading this and you don’t know what Prebid.js is, then turning you into an adopter of WurflRTD would be pretty unlikely.
On the other hand, if header bidding is your bread and butter, please keep going. You’re gonna love this, I promise.
Introducing WurflRTD for Prebid.js: Clean device signals in the OpenRTB bid request, at the moment of auction.
The idea is simple: inject accurate device intelligence into the OpenRTB bid request at the moment of auction, client-side, before a single SSP sees it. The industry is rebuilding itself around traffic shaping and curation — SSPs filtering and packaging inventory on signal quality, rather than blasting every impression into the auction and hoping something sticks. That approach only works if the signals being filtered on are actually reliable. The device object in a raw bid request frequently isn’t. WurflRTD fixes that at the publisher’s own Prebid.js setup, which means cleaner traffic shaping decisions downstream, more defensible curated deals, and less of the QPS waste that nobody wants to pay for anymore. A few lines of JavaScript, and suddenly the device signal in your bid stream is something everyone downstream can actually trust.
What is WurflRTD for Prebid.js?
WurflRTD is a Real-Time Data module for Prebid.js. When a publisher installs it, every bid request leaving that page carries accurate, structured device intelligence — not the guesswork a truncated user-agent string allows, but the real thing. Form factor is the easy part: smartphone vs. tablet vs. desktop vs. CTV. The harder part is everything downstream of that. WurflRTD can distinguish specific iPhone and iPad models (a notoriously hard nut to crack, as anyone who has tried will confirm), identify which smartphones were released in the last 24 months and retail for $600 or more, and expose the full set of WURFL capabilities that buyers have historically relied on for fine-grained targeting and bid-time decisioning.
WurflRTD also handles the HTTP dance that delivers user-agent client hints (UA-CH). UA-CH is how modern browsers now volunteer device information, and it is not something you want your publishers to have to figure out on their own.
The point of all this is not more data for its own sake. The OpenRTB device object that SSPs and DSPs rely on is increasingly unreliable — user-agent strings are frozen, reduced, or spoofed, and downstream systems end up making bidding and filtering decisions on signals that aren’t trustworthy. WurflRTD fixes the signal at the publisher, before the auction starts. What SSPs and DSPs receive is something they can actually act on.
Why WurflRTD Matters for Publishers
Publishers sit at a peculiar spot in the programmatic supply chain. They own the inventory, but they don’t control how that inventory is described to buyers. The user-agent string that leaves the browser is whatever the browser vendor decides it should be — which, these days, is less and less. By the time the bid request reaches a DSP, the device signal has been truncated, reshuffled, or misclassified by any number of intermediaries. Premium mobile inventory routinely arrives at the bidder looking indistinguishable from commodity mobile inventory. iPads disguise themselves as MacBooks. CTV impressions get called desktop. Nobody is happy.
WurflRTD gives publishers a way to fix this at the source. Specifically:
- An accurate device signal in every bid request. Form factor, model, hardware generation, screen characteristics, and the full set of WURFL capabilities — in the OpenRTB device object, where bidders actually look for it. Device age and price point are in there too, for the buyers who care about those dimensions.
- A defensible story when you sell into curated deals. SSPs are increasingly packaging inventory into curated PMP deals that make explicit quality claims — “premium mobile, verified device type, viewability above X”. Those claims only hold up if the underlying device data is trustworthy. WurflRTD gives you something to stand behind.
- Fewer wasted bid requests on the SSP side. Traffic shaping models work off the signals in the bid request. Feed them clean ones and they route your inventory to the right demand. Feed them noisy ones and your impressions get filtered out of auctions they should have won.
- A privacy-safe foundation. WurflRTD identifies device models, not people. No persistent identifiers, no cross-site tracking, no user profiling. The kind of signal that keeps working as the rest of the identity stack gets harder.
Who Pays for What
Publishers add WurflRTD to their Prebid.js configuration in a few lines of JavaScript. Small publishers do this at no cost, as their traffic is likely to fall within the boundaries of a free tier of enriched bidding opportunities. Large publishers may want to pay to get their full bid stream enriched once their monthly free tier is exceeded. SSPs may also elect to pay for the enriched signal because that is where the value actually lands in their stack: cleaner inputs to their traffic shaping models, more defensible curated deal inventory, higher-confidence decisions downstream.
Where WurflRTD Sits in the Auction Flow
The enrichment happens client-side, in the browser, before bid requests leave for the SSPs. No extra network round-trips, no server-side lookups, no added latency. By the time the first SSP sees the bid request, the device signal is already in the OpenRTB device object — accurate, structured, and visible to every bidder in the auction. That’s the point of doing it in Prebid.js rather than further downstream: one enrichment, one signal, everybody works from the same data.
Integrating WurflRTD into your Prebid.js setup is simple and takes just a few steps. Here’s what you need to know:
- Install the WurflRTD Provider: Download and integrate the WurflRTD module into your existing Prebid.js configuration.
- Configure Your Settings: Customize the provider’s settings to align with your specific goals, whether it’s optimizing for device type, OS, or other device characteristics.
The payoff is a bid stream that downstream systems can actually trust.
For detailed instructions on how to get started, check out the WurflRTD Integration Guide.
Whether you’re a publisher trying to get the most out of your bid stream or an SSP building a more defensible supply story, WurflRTD is a small piece of infrastructure that does real work. Accurate device intelligence in every bid request, client-side, at the moment of auction. If that’s useful to you, get in touch — we’re happy to walk through what a deployment looks like in your setup.
Curious whether your bid stream could use it? Let’s find out.