Ad trading via auction was the best decision ever made. No matter what price you put, there is always someone willing to pay more. But what is the best way to not mix up the types of programmatic bidding and get prime impressions? The ultimate header bidding vs waterfall comparison is waiting for you below.
5 December 1766. London's Pall Mall is full of British noblemen in anticipation of the first Christie's auction in history. Soon, a former print warehouse will open its doors, welcoming all who arrive to celebrate the best art pieces from England and beyond. They are all put on display and set for sale, but with one condition: whoever has the highest bid gets the item.
The appeal of art auctions has not decreased over time; it's quite the contrary. Christie's grew into the largest auction house with 10 salerooms and a presence in 46 countries. The auction proved to be the best way to sell something without a definitive price.
Online advertising took almost 250 years to adopt the same rule. In the early days of ad tech, ads were served through a daisy-chaining process, aka waterfalling, where the bid of each advertiser should meet the minimum price set by the publisher. As technology advanced, the programmatic advertising ecosystem shifted to header bidding auctions, in which the highest bidder wins, just like what happened at Christie's.
The numbers show how settled the debate has become. In Red Volcano's March 2026 crawl, a single header bidding partner was set up in the Prebid configurations of more than 550,000 publisher websites.
Running your own campaigns next to Google? Epom Ad Server hosts your Prebid tag and puts your campaigns into the same auction as the SSPs. Header bidding is included in your plan. Start a free trial.
• Header bidding vs waterfall is a choice between asking buyers one at a time and asking them all at once.
• Waterfall bidding, or daisy chaining, offers each impression to demand partners in priority order. The first to meet the price floor wins, even when someone further down would pay more.
• Header bidding asks every partner at the same time, so the highest bid wins. Digiday research found 20–30% more revenue than waterfall-only setups.
• Waterfalls still have three jobs in 2026: ordering direct deals by priority, chaining passbacks on niche and adult inventory, and in-app mediation for networks that can't bid in real time.
• A waterfall's weak spot is the blank ad. End every chain with a house ad or default code, so unsold impressions don't stay empty.
What is Daisy-Chaining or Waterfall Bidding?
This one we know as the old school way of ad serving. It was slow, yet it raised publishers' fill rates in the years before automated auctions.
Waterfalling, or daisy-chaining, is a method to sell the publisher's inventory in which the publisher sets the priority for each advertiser or ad network to which they're connected. They also set a price floor, the minimum accepted price for the placement.
The slot is offered to demand partners one by one, in order of priority. Once someone meets the price floor, the impression is sold.
Inventory passes the route from top to bottom, like water in a river flowing down a cliff into a basin. That's why it's called a waterfall auction, although it's not quite an auction: the impression often goes to whoever is first in line, not to the highest bidder.
In an ad server today, a waterfall looks like a ladder of priorities. Sold campaigns are checked first, then lower tiers, with house ads at the bottom.
When a partner in the chain has nothing to show, it hands the impression back with a passback tag, and the next one in line gets the request.
Overall:
- The waterfall model is simpler to implement and requires less technical expertise than header bidding, which can be complex and resource-intensive.
- Waterfall bidding is ideal for smaller publishers with limited technical resources due to its straightforward setup and management.
- The waterfall model offers simplicity and better control over partner prioritization, making it easier to implement for selling remnant inventory.
A Waterfall Ads Example
A sports site sells a 300×250 slot through this chain:
- A direct sponsor at a fixed $8 CPM
- Network A, with a $3 floor
- Network B, with a $1.50 floor
- A house ad
A visitor has already seen the sponsor's banner three times, so frequency capping skips it. Network A has nothing for a mobile user in Brazil and passes the impression back. Network B fills it at $1.60, while a buyer outside the chain would have paid $2.40.
The numbers are illustrative, but the pattern is how waterfall ads lose money.
What is a Header Bidding Auction?
Comparing header bidding vs. waterfall, the key difference is that header bidding is an automated auction in the truest sense of the term: publishers offer each impression to several SSPs and ad exchanges at once, and the highest bid wins. Most header bidding runs on Prebid, the free open-source wrapper, so you will hear "Prebid" used as shorthand for it.
A practical header bidding example: a news publisher with 2 million monthly visitors connects to five SSPs via a Prebid wrapper. When a user loads an article, all five bid for the ad slot at the same time. Bids that arrive before the timeout, usually set around one second, enter the auction, and the highest one goes to the ad server, where it competes with direct deals and Google demand.
For setup, client-side vs server-side and timeouts, see our header bidding guide.
RTB vs. Header Bidding in Programmatic
Real-time bidding and header bidding are often used as synonyms, and a modern RTB often uses header bidding technology. Still, RTB is wider: without header bidding, it runs through an older ad tag model, or through an SDK in apps. Before header bidding, the programmatic waterfall was the dominant model, and publishers still use it for direct deals and remnant inventory.
First-price vs. Second-Price RTB in Programmatic
Header bidding auctions used to be second-price, where the winner paid slightly more than the runner-up's bid. Since Google rolled out first-price auctions to all Ad Manager partners in September 2019, first-price is the norm: the winner pays its own bid, which is simpler and more transparent for both sides.
A first-price header bidding auction is more beneficial for publishers, as the highest bidder pays exactly the price equivalent to their bid. It's the simpler framework that allows for maximizing yield from each inventory sale. Some experts, like Verizon's Media Deepti Phatak, believe that first-price auctions also give advertisers an edge, as this model is more transparent and eliminates the hidden fees & manipulations inherent in second-price auctions.
Is Waterfall Bidding Still Viable in 2026?
The waterfall vs header bidding question used to be a genuine debate. Today, for any publisher running meaningful traffic volumes, it is largely settled.
If you're going to exclusively use the waterfall model, or waterfalling, as your method to automate your media buying, then it's a resounding NO. There is no reason in 2026 to lose revenue share and prioritize low-paying demand partners.
When the first advertiser doesn't meet the publisher's price floor, the Ad Server passes the request to the next in line. And if, say, the first three advertisers bid below the floor, the waterfall model may increase ad-serving latency.
The only viable case in which a publisher can use waterfalling is when they need to organize their equally paid direct partners in a specific order. Also, publishers may prioritize some advertisers over others and start with a waterfall before entering a real-time bidding auction.
Ad networks often use an ad server and its waterfalling feature for premium placements, while programmatic tools are used for regular and remnant inventory. It doesn't mean that you can't find premium partners while trading ads via RTB; you can't be 100% sure. Everything is automated and based on the advertiser's targeting preferences, not on their overall credibility.
Header Bidding vs Waterfall: What Is the Most Effective Solution for Publishers?
Before we start, here are some stats to consider:
- Studies reveal that over 70% of digital publishers have shifted from the traditional waterfall model to header bidding to enhance revenue potential.
- Even five years ago, header bidding typically resulted in revenue increases of 20% to 30% compared to the waterfall model.
- Header bidding can increase revenue by 20–50% for publishers compared to traditional methods.
Given how header bidding works, this RTB advertising method increases healthy competition and reduces time spent on campaign management. Thus, publishers gain a clear advantage by making more money with a smaller time investment.
Managing header bidding alongside direct deals and programmatic backfill works best when everything runs through a single platform. Epom Ad Server hosts your Prebid tag and gives you control over direct campaigns, partner bids and delivery logic from one dashboard.
Summarized header bidding benefits for publishers:
- Header bidding ensures that inventory is sold to the highest bidder, while waterfall sells it to the first-in-line advertiser.
- Header bidding happens in real time, while waterfall uses a sequential auction model that increases latency.
- Header bidding needs a one-time ad server setup, whereas waterfall tags must be updated constantly.
- Header bidding implies a wide range of demand sources, whereas waterfall implies only sources you already know.
- Header bidding ensures a better fill rate, while waterfall sometimes fails to sell inventory on time
It may seem that header bidding only counts for publishers, as it's a publisher's method of data transmission, but that's not quite true. The revenue impact is measurable. Publishers that implement header bidding typically report higher CPMs and stronger relationships with their demand partners.
💡 Based on Epom experience with our clients, publishers who switch from waterfall-only setups to header bidding with three or more connected demand partners typically see fill rate improvements within the first 30 days, often before any significant CPM change is measurable.
Let's review some benefits for advertisers brought to the table by header bidding:
- Header bidding gives advertisers access to more traffic sources, while waterfall involves them in auctions organized by partner publishers.
- Header bidding allows bidding on premium inventory, while waterfall is usually used to sell remnant inventory.
Both header bidding and waterfall models have their use cases, but header bidding is increasingly seen as the dominant method for maximizing ad revenue.
How Header Bidding Increases Revenue Compared to Waterfall
Header bidding earns more for one reason: every buyer prices the impression before the ad server decides. In the waterfall example above, Network B won at $1.60. With header bidding, the $2.40 buyer bids in the same auction and wins.
More buyers per impression means more bid pressure, and fewer impressions fall through to house ads. Digiday research put the uplift at 20–30% compared to waterfall-only setups.
Header Bidding Disadvantages
It’s not hard to understand who is more handsome in a header bidding vs. waterfall couple. Yet it doesn’t mean that header bidding is flawless.
- Latency. The page waits for bids until they arrive or the timeout ends. Every extra bidder adds a request.
- Setup. A wrapper is slightly more complicated to implement than a simple ad tag, and it needs ongoing tuning.
- Duplicate bidding. Advertisers buy through several DSPs, and publishers connect many SSPs, so one buyer can end up bidding against itself. Fewer, direct paths to each buyer reduce it.
Most latency comes from client-side header bidding, where the user’s browser calls each SSP. Server-side moves the auction to a server, so the page makes one request, at the cost of weaker cookie matching. Many publishers run a hybrid.
To cut latency without losing revenue:
- Set the timeout from your own data, not a default.
- Drop bidders that rarely win or answer late.
- Move slow or low-value bidders server-side.
- Load the wrapper asynchronously, so the page renders while bids arrive.
Header bidding’s disadvantages are manageable. For most publishers with meaningful traffic, the revenue upside outweighs the technical overhead. For server-side setup, see our header bidding guide.
Waterfall vs Bidding-First Mediation in Apps
Apps went through the same shift as the web. Mediation platforms started as waterfalls, calling ad networks one after another by expected eCPM. Most now run bidding first: networks bid at the same time, like header bidding on the web.
Waterfalls didn't disappear. AppLovin MAX describes its bidding networks as competing in a real-time auction, while other networks respond one by one in a waterfall ordered by CPM. AdMob still lets publishers set up waterfall ad sources.
Platforms keep the waterfall because not every network can bid. If you run waterfall mediation and wonder whether to switch, keep the waterfall below the bidders and compare eCPM and fill per network for a few weeks. Our ad mediation guide covers the platforms.
Header Bidding and Waterfall in One Stack
In 2026, most publishers don't pick one. They run both, each where it pays:
- Sold campaigns go first, in the order of their priority. This is a waterfall, and it stays.
- Header bidding prices the rest. The best Prebid bid competes with Google demand in the ad server, and the higher one wins.
- Whatever is left falls down a short chain: house ads, or a passback to a partner that buys what others don't.
This is the order Epom Ad Server uses after Release 151: your sold campaign, then the best price between partner bids and Google, then house ads, then fallback markup.
Passback Tags: How Waterfalls Avoid Blank Ads
A passback tag is the piece of code that keeps a waterfall moving. When a partner gets a request it can't fill, because of targeting limits, exceeded capping or no ads at all, it calls the passback tag and returns the impression to the publisher. The publisher's ad server then requests the next partner in line, or shows a house ad.
Three terms get mixed up here:
- Passback: the impression goes back, so another partner can try.
- Fallback (default code): what the ad server shows when nothing in the chain fills, such as a house ad or the publisher's own markup.
- Blank ad: nothing fills and no fallback is set, so the slot stays empty.
Why Niche and Adult Networks Still Rely on Passbacks
On mainstream inventory, header bidding finds a buyer for almost every impression. On niche and adult inventory, the buyer pool is smaller.
Google's publisher policies keep AdSense and Ad Exchange off adult content, and many programmatic buyers exclude these categories too. Networks in these verticals work outside Google and chain direct advertisers and specialised networks through passbacks, so each impression gets more than one chance to sell.
How to Set Up a Passback Chain Without Blank Ads
- Order partners by what they pay, not by habit. Check eCPM and fill per partner every month.
- Keep the chain short. Every hop is another request and more waiting time for the user.
- Always end the chain with a fallback: a house ad or default code. A chain without one turns unsold impressions into blank slots.
- Track passbacks per partner. A partner that passes back most requests is slowing the chain without adding revenue.
How Passback Works in Epom Ad Server
In Epom Ad Server, passback is switched on per third-party banner. After saving, the banner gets a passback snippet for the partner to call when it can't fill. Epom then drops that banner for the request and picks the next one.
Default code is the fallback markup for site placements, used when nothing in the chain fills. For video, a backfill campaign at House ads or Non-guaranteed priority does the same job.
The Advanced report counts valid, rejected and fraud passbacks per partner. The No ads column counts requests that ended with nothing to show.
Conclusion: Header Bidding vs Waterfall Comparison Table
| Feature | Waterfalling | Header Bidding |
|---|---|---|
| Automation | The lack of automation constantly requires manual adjustment. | Needs only one manual action, then becomes automatic. |
| Core Element | Standard Ad Tag | Header Tag |
| Winning Bid | First bid that meets the price floor | The highest bid |
| Priority | Advertisers are prioritized in a specific order. | No priority, all advertisers are equal. |
| Process | Sequential | Real-time |
| Latency | High | Still persists, yet is eliminated in server-side header bidding |
| Fill Rates | Low to Medium | High |
| Revenue | Lower than could be | Optimized |
| Optimization | Manual | Automatic |
| Competition | Low | High |
| CPM | Doesn't matter much; what matters is priority. | Matters a lot, the higher the better. |
Header bidding is a new standard in programmatic ad serving, while waterfall is becoming increasingly obscure. Still, it doesn't mean that waterfalling will irrevocably fade into oblivion. Some publishers that prefer direct deals will likely use it to prioritize those deals, and niche networks still chain partners through passbacks.
Yet, many publishers are adopting a hybrid approach that combines both header bidding and waterfall strategies to optimize ad revenue. Overall, though, you should consider header bidding if you haven't implemented it yet.
