Header bidding has revolutionized programmatic advertising by allowing multiple demand sources to bid on ad inventory simultaneously, rather than sequentially as in traditional waterfall auctions. This competition results in higher CPMs, better ad quality, and greater control for publishers. 

Today, header bidding is an essential strategy for maximizing revenue and improving the user experience. 

In addition, the auction enables publishers to present their ad inventory to several SSPs, ad networks, and exchanges simultaneously before involving the ad server. This creates a more competitive and transparent auction, as demand sources bid in real-time for the same inventory.

But let’s dive headfirst into this bidding auction. All puns intended. 

How Header Bidding Works

Header bidding enables multiple advertisers to bid on a single ad impression simultaneously. This process begins when a user loads a webpage, triggering the header bidding script to send ad requests to various demand sources like Supply-Side Platforms (SSPs), Ad Exchanges, and Demand-Side Platforms (DSPs). Each demand source responds with a bid and ad creative in real-time, after which the highest bid is selected and displayed to the user.

The real-time bidding (RTB) process enables advertisers to bid on ad impressions within milliseconds, helping publishers secure the best price for their inventory while allowing advertisers to reach their target audiences effectively.

Behind-the-Scenes Mechanics

Header bidding operates on a system that involves creating HTTP requests, handling bid responses, and leveraging user data for efficient targeting.

HTTP Request Generation and Processing

When a user visits a webpage, the header bidding script sends HTTP requests to demand sources. These requests, typically formatted in JSON or XML, contain parameters such as ad unit size, placement ID, and user data. Demand sources process these requests and return bid responses with a bid amount and ad creative.

Pre-bid and Post-bid Phases

• Pre-bid Phase: The header bidding script sends bid requests to demand sources, collects bids, and runs an auction to determine the highest bid.
• Post-bid Phase: The winning bid is sent to the ad server. Ad Server then updates the winning bid in the analytics and displays the ad on the ad slot using its ad-serving logic.

The Role of User Data

The user data is part of the bidding process, allowing selection of certain audiences based on their characteristics: age, interests, and behavior. Through cookies and pixels, the bid of demand sources can be adjusted according to the relevancy and the revenue. However, acts like GDPR and CCPA permit data and personal information to be handled responsibly and only with permission, data anonymization.

Key Components of Header Bidding

1. Header Bidding Wrappers

Header bidding wrappers are JavaScript libraries that sit on publisher websites and enable them to connect with demand sources like SSPs, DSPs, Ad Networks, etc. They help in orchestrating the entire auction using the below steps.

• Manage the setup
• Bids collection
• Ads creative placement

Popular wrappers like Prebid.js and Amazon TAM are used to standardize and simplify the integration of multiple demand partners.

These wrappers streamline bid requests and ensure that all the demand sources can participate in the auction. By connecting all the demand sources wrappers improve transparency and competition in the auction process.

2. Ad Exchanges and Ad Networks

Nowadays in adtech, it’s hard to differentiate between Ad Exchanges, Ad Networks, DSPs, and SSPs as companies are doing all types of activities. Some differences in the process are still there. Ad Exchanges & Ad Networks aggregate demand and connect to supply.

• They connect various advertisers and DSPs.
• They aggregate the demand and help in facilitating real-time bidding b/w publishers and advertisers.
• The real-time bidding ensures everybody gets the benefits 
  • Advertisers get the best-performing slot for their ads. 
  • Publishers get the best rates for their ad slots.

Ad exchanges and ad networks work hand-in-hand with wrappers to process bids quickly and efficiently.

3. JavaScript Implementation

Header Bidding relies heavily on JavaScript implementation, it executes bid requests and collects responses within the user’s browser. It’s crucial to implement JavaScript to ensure maximum profit.

• Proper optimization of JavaScript ensures more demand sources can participate, maximizing revenue for publishers
• To ensure a seamless auction, enable faster page load time & reduce latency.

Each component of header bidding like wrappers, ad exchanges, and networks interacts through this JavaScript infrastructure, to drive efficient and competitive auctions.

Header Bidding Architectures

Header bidding can be implemented through two primary architectures: client-side and server-side.

Client-Side Header Bidding: In client-side header bidding, the bidding process happens within the user’s browser. It is executed by a JavaScript wrapper. It sends bid requests to demand partners, then collects bids, and auction the ad slot based on bids and predefined configuration using RTB. While this method provides transparency and control, it may cause latency and scalability issues, as multiple requests must be processed by the browser.

Server-Side Header Bidding: Provided the Client Side Header Bidding issues, server-side header bidding was introduced. It uses a server to perform the bidding. The browser sends one request to the server, which then handles bid requests to demand partners and returns the highest bid to the browser. This reduces latency and improves scalability but can reduce transparency and control over the bidding data.

Tools to Check Behind the Scenes of Header Bidding

Developer Tools

• Chrome DevTools: Chrome DevTools is a fundamental yet crucial tool for monitoring and troubleshooting header bidding processes. It allows tracking of bid requests and responses in the Network tab, identification of JavaScript errors in the Console tab, and performance analysis in the Performance tab.
• Requestly: A browser extension that simplifies ad operations troubleshooting. Requestly intercepts and modifies HTTP/HTTPS requests, allowing you to fix malfunctioning ads, simulate geo-targeted campaigns, and test custom ad experiences. It also supports testing Prebid.js configurations in staging environments to catch issues early.
• Charles Proxy: A desktop-based application that works similarly to Requestly. It can monitor and debug HTTP/HTTPS traffic. It includes features like bandwidth throttling, session recording, and request/response inspection.

Analytics and Reporting Tools

• Prebid Analytics Adapter: Built into Prebid.js, this tool sends auction data to analytics platforms and tracks key metrics such as bid responses and win rates. It helps publishers optimize their header bidding strategies in real time.
• Professor Prebid: A specialized tool for Prebid users, providing detailed insights into header bidding performance metrics like win rates and revenue impact.
• Google Ad Manager: Google Ad Manager is widely used by publishers for tracking ad performance, impression data, and auction outcomes, offering a comprehensive view of header bidding activities.

Maximizing Revenue Through Optimized Header Bidding Strategies

Header bidding is crucial when it comes to generating maximum revenue with the help of better ad placements. Knowing how it works, how to make it faster and more efficient, and which tools to use, ad ops professionals can reach its full potential. Staying updated on best practices will be key to maintaining a competitive edge in the evolving ad tech landscape.

The post Behind-The-Scenes of Header Bidding and How It Creates Better Ad Quality for Publishers appeared first on AdMonsters.

Digital advertising has presented challenges and opportunities for publishers looking to future-proof their businesses. In a recent LinkedIn Live event, I sat down with Justin Wohl, the Chief Revenue Officer of Snopes, to discuss the key trends and strategies shaping the future of ad ops and the ad tech stack. 

Wohll offered a unique perspective on the shifting dynamics between buyers and sellers. For instance, the supply side has been catching up to the buy side, becoming more sophisticated and aware of their value. This newfound empowerment is crucial to the industry’s evolution as publishers strive to optimize their partnerships and drive greater efficiency in their ad tech stacks. 

Beyond the changing publisher-buyer relationship, Wohl also delved into the potential impact of emerging technologies, such as AI and generative AI, on the future of ad ops. 

But he heeds a warning about Google’s trial with the DOJ. Wohl explains that the decisions made in this case could profoundly influence the future of pre-bid and header bidding, potentially unlocking more efficient and pure programmatic opportunities for publishers. “Should that unlock for us, it’ll create plenty of movement and opportunity for publishers to, like better, optimize and become more efficient with how we run our auctions and sell our inventory.”

Want to get the full scoop? Watch the full live below: 

Publishers Grow More Sophisticated as Ad Tech Evolves

Over the years, the ad tech industry has significantly transformed, and publishers are still learning to find their footing. According to Wohl, the dynamic between the buy-side and sell-side has been shifting, with publishers playing a more active role.

“I think that the experience of always being on the supply side, always being on the publisher side of this business, has been one of catching up to the buy side,” he explains.

Wohll notes that publishers are becoming more sophisticated and aware of their value proposition, leading to a more balanced understanding of the tech stack between buyers and sellers.

“We’ve become smarter, more sophisticated, more aware of our value proposition. As tech understanding comes to parity on both the buy and sell sides, we should keep seeing this evolution of more sophistication and self-determination on the supply side, where we know the value of our inventory and can sell it successfully,” said Wohl. 

Snopes FactBot: Integrating AI in Ad Tech 

Interest in integrating AI and generative AI technologies is growing. Wohl points out the difference between traditional machine learning, which has been valuable for data analysis, and the new opportunities generative AI presents for publishers.

One example of Snopes’ use of AI is its fact-checking bot, designed to be transparent when it doesn’t have an answer. Rather than generating inaccurate responses, the bot informs users when it cannot address a query. Snopes then logs those unanswered questions and uses large language models to analyze recurring topics, providing valuable insights for the editorial team to expand their fact-checking coverage.

This strategy helps Snopes understand audience interests while expanding its content. However, Wohl also emphasizes the high costs of using generative AI. The expense per query, he explains, is significantly higher than the revenue generated from those queries — about five times more. As publishers consider adopting these technologies, Wohl advises carefully evaluating the potential benefits and costs, focusing on practical applications that can optimize efficiency and revenue.

Practical Strategies for Optimizing the Ad Tech Stack 

Drawing from his own experiences, Wohl urges publishers to understand their programmatic partners’ contributions to revenue versus the opportunities provided. 

“One of the recurring conversations I have with our SSP partners is about their performance. We run a direct-only auction with no reselling, involving around 20 SSPs, each with a direct seat. Every quarter, they all ask where they rank—whether they were our third-largest or smallest buyer.”

Wohl suggests that publishers should come prepared for industry events, such as the upcoming Pub Forum Scottsdale conference, with this level of partner analysis. 

“I want to optimize my partners and know who sits where so we can talk about the business and why that is.” By understanding the nuances of their ad tech stack, publishers can make informed decisions about optimizing their partners, determining which ones to include or exclude from their auctions, and ultimately driving greater efficiency and revenue. 

During his session at AdMonsters PubForum Scottsdale, Wohl plans to explore this concept of partner optimization.

The post LinkedIn Live Replay: The Future of Ad Ops According to Snopes’ Justin Wohl appeared first on AdMonsters.

Although Google delayed their third-party cookie cut-off, there have to be new and effective ways to aggregate consumer data in an ethical manner. 

Amongst buyers and sellers alike, there is already some discussion about the next phase of data collection practices and theories. So much so, that it could be a bit of information overload for the ad tech newbie or veteran. Regardless, the industry is moving full speed ahead to stay on top of these new regulations. 

One of the more widely spread solutions is Alternative IDs. This medium involves identifiers that are privacy-focused and allows publishers to access first-party data that is volunteered by the consumer. 

This can include data assets such as email addresses, clicks, and names. While the data is collected through cross-device usage, both publishers and advertisers have first-party IDs to find ad placement matches. As of now, there are 10 prominent alternative ID solutions and it is one of the most common practices for cross-site tracking. 

The majority of companies are encrypting their user data. Others are taking it a step further by using ID tokens that periodically alter, similar to the theorized Web3 tokens. 

So, what are the different ID solutions? From February 2022 – June 22, Emarketer and Sincera highlighted that thousands of publishers were using what they called the Top 5 Identifiers. including ID5, Unified ID 1.0, RampID, Prebid SharedID, and Criteo.

Here we look at 10 such solutions and explain how they’re supposed to work.

Top 10 Alternative ID Solutions

ID5 ID

The ID5 ID is specifically designed for publishers. This is an encrypted first-party data ID that gives publishers a unique device ID they can share with their partners and clients. With their ID5’s identity cloud publishers can:  

• Identify users in all digital advertising environments to increase monetization with the ID5 ID
• Ensure compliance with data privacy regulations with their privacy-by-design technology
• Prevent data leakage by permissioning user IDs to authorized partners only

This specific ID is unique because its technology, as mentioned above, is privacy focused by design. Browsers and applications that work with ID5 must disclose to the user that their data is being collected and what kind of information is being collected. Users are given consent, opt-out and do-not-sell signals from ID5 based upon their preference. 

ID5 has a relationship with about nearly 600 publishers and with their mass amount of data assets, ID5 can retrieve that data to create a unique ID. With the data in the ID publishers can use cross-site data collection, profiling, and data measurements, and share the ID with partners to target ads and sell them for higher prices. 

Unified ID 2.0

This specific ID solution was launched by The Trade Desk but has evolved more into a collaboration with Prebid, with Prebid.org taking on the role of operator of UID 2.0’s technical infrastructure. Much like the other solutions, the consumer’s email is used to build the ID and that allows them to receive personalized advertisements. In practice, UID 2.0 can facilitate cross-site app reach, create targetable audience segments, measure attribution and conversion of ad spend, and analyze ad campaign details.

For the benefit of publishers and advertisers, the solution targets ads to users giving both entities encrypted IDs that can be matched in the bid stream. The ID itself does not contain data, but it is transferred in the bid stream process. UID2 is a decentralized system that can be used by small and large publishers and advertisers who want to use the encrypted IDs for matching and who have a large database of users’ emails. 

This solution has become one of the more popular Alternative IDs. In fact, the popular streaming service, Tubi, became the first CTV publisher to implement the ID in July of 2021. 

In terms of its development, The Trade Desk’s Chief Revenue Officer, Tim Sims, said that “Unified ID 2.0 is a collaboration among everyone within the advertising ecosystem to pioneer an interoperable, open-source ID solution that helps advertisers connect with audiences while keeping consumer control and privacy in mind.”

Ramp ID

LiveRamp’s ID solution is self-described as “a people-based map connecting de-identified offline touchpoints and online devices.” It has one of the largest U.S.-based device data assets and works to design its technology to collect that data in an ethical way. 

There are the aspects to the process of creating their ID: 

• Offline PII merging: Resolving separate emails, postal addresses, and phone numbers to a single individual
• Online device linking: Matching disparate devices to people-based anonymous identifiers
• Offline to online: Merging these offline and online identity spaces into a unified, privacy-conscious, people-based ID space

While many ID solutions have interoperable properties, this specific solution is known for its interoperability. It can connect with IDs stored with premium publishers such as Google and Facebook, direct internet deals, and marketplace activations. After the ID is created, it offers cookie syncing to match in DSPs. For optimal user experience, publishers and advertisers who want to outsource ID creation and greater interoperability should be on the lookout for RampID. 

SharedID 

This community-focused ID was created by Prebid and  is inherently unique from the other solutions because it is owned and operated by Prebid.org’s publisher members rather than a vendor. The community-led technology is privacy focused and targeted to work across different publishers and DSPs. 

To start the ID process, publishers write the SharedID as a first-party value. After creating the unique user ID with the data, the publisher can make it accessible to a bidder or “use the user ID as a building block for publisher controlled first party self-declared data without the use of an identifier.” 

The main functionality of SharedID is acting as a storage center for first-party data. On the other hand, unlike many of the other ID solutions, SharedIDonly offers cross-site tracking within the publisher’s domain. Thus, it is recommended that only tech-savvy publishers with fewer resources that want to share their first-party data with DSPs should use this ID solution. 

ConnectID

Yahoo’s alternative ID solution, ConnectID, “uses consent-based, first-party data to bring audiences to life in a world without cookies.” One of the biggest pulls for this specific solution is Yahoo’s large consumer base that consists of 500 million unique user profiles and about 200 billion daily cross-screen user intent signals. In addition, the company has direct consumer relationship data across sites such as Yahoo News, Sports, and Finance. This helps publishers create in-depth identity-based user profiles. 

While the data is expansive, ConnectID only works with Yahoo’s DSP which has audience activation and media measurement with assets such as CTV, programmatic audio and much more. Experts agree that it should be used by publishers or advertisers that have a lot of resources, direct user relationships, and who want to outsource ID matching. 

To start the process, a Yahoo user signs in using their email address and gives consent for the site to use their information. Their address is encrypted after being sent to Yahoo and is then sent back to the publisher as a ConnectID to share for bidding. After the ID is created, the publisher and advertiser’s ConnectID are matched on Yahoo’s DSP which initiates the process across multiple media channels. 

Panorama ID

This specific ID is a privacy-complaint solution for the open web that caters to multiple inputs such as CTV, web, and mobile. Created by Lotame, the solution is powered by the brands’ patented graphing technology and it connects “different types of device identifiers, associated individual behaviors, and privacy choices into a single view, without dependence on cookies.” 

For privacy compliance, Lotame has included a universal opt-out option for consumers. If they opt out once, the decision will be reflected in real-time and maintained across every device and touchpoint associated with the ID. In addition, Panorama ID has access to 90 platform partners and data from 180 providers across 58 countries. For each anonymized ID, there is an average of 199 web and 89 mobile attributes that both publishers and advertisers have access to. 

Since it uses multiple data points, it can be utilized across SSPs and DSPs to match with ads and other IDs. Publishers who want to obtain user content and who have a large user base to provide active consent to be tracked and shown personalized advertisements should look out for this ID. 

CORE ID

CORE ID reaches about 200 million consumers in a privacy-safe way, according to the creator of this ID solution, Epsilon. It offers publishers and advertisers deterministic data rooted in offline name or address data. Publishers and advertisers with resources to outsource ID resolution and matching should use the solution

To start the process, after a user is tagged on the publisher or advertiser website, they call the CORE ID service to see if the user has a CORE ID. If the user does, their email address is pushed through the hashing process and shared with the service provider. Afterward, that data is paired with other data connected to the user that can create a unique profile identifier. The matching process occurs through Epsilon’s DSP and their data processing system called Agility. 

Unlike the previous IDs mentioned, CORE ID does use cookies as a part of the unique ID. Despite this, CORE ID asserts that third-party cookies going away will not affect their process. Since it is “people-based, CORE ID is anchored on deterministic data elements, making it highly reliable and stable. The data is pseudonymized before it enters the digital ecosystem, keeping consumers’ information properly safeguarded and aligned with evolving regulations.” 

Fabrick ID

Neustar launched their unified identity ecosystem in 2020. The Fabrick ID is an “innovation in identity-powered media and measurement that improves omnichannel marketing effectiveness, while future-proofing brands, publishers, and platforms in a privacy-first, post-cookie world.” 

Like many of the ID solutions mentioned, publishers and advertisers can use the ID solution to conduct cookieless audience targeting and data personalization. According to Neustar, Fabrick ID has already fostered success for their clients: 

• With Fabrick’s Multi-Touch Attribution and Marketing Mix Modeling, a retail client increased the accuracy of their cross-media forecasts, reduced data sparsity, and evolved into a 100% data-driven marketing organization. 
• A telecommunications firm increased their ad spend by 4x, increased their on and offline sale conversion rate, and bolstered their ad creative. 

Designed as a cookie replacement, Fabrick ID is a programmatic token made up of a culmination of publisher-provided data. With the PII, clients who use the solution build marketing programs. Once the campaign is executed, Neustar performs measurements using Fabrick and other IDs. It is perfect for publishers that use multiple IDs and want to connect those data points across one platform for transactions and measurements. 

nonID 

LiveIntent’s nonID is unique because it does not require publishers or advertisers to conform to their specific ID solution. According to the brand, their “nonID data processing service is able to safely bridge whatever ID you’re using to the wider digital ecosystem via an active, encrypted email address.” Not only does the solution offer cross-site identification, but it can also help determine which email address is best used for the multi-device process. 

All ID solutions are powered by first-party data, but LiveIntent has data collection assets with 200 million unique email addresses and 25 billion authenticated data signals. 

In addition, due to LiveIntent’s partnerships which allow them to place ads within publishers’ newsletters, they have large access to zero-party data. Therefore, the ID solution is recommended to publishers with newsletters that want to outsource ID creation and matching and who want to generate revenue by putting ads into newsletters. 

SWID

This user-first ID solution was created by a network of operators called the Secure Web Addressability Network (SWAN). 

The Secure Web ID standards are strict as SWAN requires all operators to sign a binding contractual agreement and follow their standardizing terms. Based on the consumer’s data sharing preferences, the SWAN operators created the SWID on their behalf. Transparency is the core of their system. It is best used by publishers and advertisers of no particular size or resources that want a decentralized ad tech system that centers around consumer consent. 

As mentioned above, consumers can choose to opt in or out of sharing their data. On the other hand, consumers can change their preferences any time they please. To start the ID creation process, a pop appears on the publisher page that uses SWAN where the consumer can choose their preferences. If they choose to share their email address, the SWID is created. The SWID is stored into a cookie in a single browser where both the sender and receiver can increase transparency through the ad transaction supply chain. 

Unlike some of the other solutions, publishers cannot buy or sell profile or media data. Publishers can only create personalized profiles for the data which they can sign in to view. 

The post What are the Top 10 Alternative ID Solutions and How Should You Use Them? appeared first on AdMonsters.

Google’s incorporation of prebid is supposed to make ad ops easier by allowing publishers to manage their header bidding relationships through “yield groups.” This will enable publishers to identify some portion of their inventory for Open Bidders to target, and prebid bidders can now do the same.

Ad servers will no longer need to be filled with line items to heighten header bidding demand. Publishers and programmatic executives over at Bustle, AccuWeather, and Cafe Media are all currently testing out this new feature and seem to be in favor. 

This Google Ad Manager prebid link-up does help to eliminate some discrepancies.

“Currently, prebid rounds all the prices before submitting them to GAM, but in the new setup, GAM is reading the price from prebid before prebid has done any rounding,” said Patrick McCann, SVP of Research at CafeMedia and Chair Prebid.js. “You won’t see the rounding errors. You end up with a better auction outcome when people are bidding closely to each other thanks to the rounding effect in prebid.”

Will This Interrupt Pre Existing Header Bidding Set Ups?

It’s looking like it just might. Yesterday on LinkedIn, we saw ad tech vendors questioning CafeMedia’s CSO, Paul Bannister, after he posted about the “bridge to prebid.” 

“There is an existing setup with the orders and line items which publishers are efficiently running and is a proven model,” said Dikshant Joshi, Director of Publisher Development at AdPushup. “While this seems to be a promising tool introduced by Google, there is no proof the performance would be at par or better or less. Google mentions the setup of yield groups (an additional and different setup). Still, the experts in the industry would want to benchmark the performance of the header bidding yield groups before they make a complete move or stick with what they currently have.” 

Google also mentions an offering of a “NetworkMinimumBidToWin” metric that publishers can access via Data Transfer Files. Nonetheless, this is a paid feature with recurring costs, so you know what that means; many won’t be able to access it if they do not wish to pay. 

If “NetworkMinimumBidToWin” were a tool to help publishers and the ecosystem, we would have seen this as a part of standard metrics accessible through APIs and the query/reporting tool.

What Is the Future of Google Ad Manager and Header Bidding?

I have asked the Universe tons of times to provide me with a crystal ball, but until then, we will have to ride the wave to see how this pans out. When we spoke to Dikshant, he mentioned that ad tech is an ever-evolving space, and with this promising change, it will be interesting to see “the actual value it brings against the current set up.”

Although many publishers will have to change their ways of using GAM’s services, others can see the benefits.

“The timeline is easy to criticize since header bidding has had such widespread publisher adoption for so long, but this still feels like a step forward,” said Emry Downinghall, SVP Programmatic Revenue, and Strategy at Unwind Media. “Google publicly acknowledges that +90% of publishers use header bidding, and they are working to support that. If this provides mediation transparency and simplifies setup in GAM, publishers win.”

The post Will Google Ad Manager’s Link-up With Prebid Reinvent the Wheel? appeared first on AdMonsters.

Brought to you by

[gated-content]

[/gated-content]

The post AdMonsters Revenue Strategy Sessions On Demand appeared first on AdMonsters.

Tom Levesque, VP, Product at OpenX, and Emry Downinghall, VP, Advertising at Chegg, sat with us to talk about Prebid as a unifying, open-source solution for Publishers and SSPs and why some companies are still reluctant to make the shift. 

In the earlier days of ad monetization, many publishers chose to overlook the page latency issues created by dropping ad code on the page because it drove significant yield for the business.

But as the demand for header bidding grew, many publishers were faced with a tough choice on how to manage it: Continue down their existing, proven (but sometimes flawed) path or find a new, more efficient solution that would continue to drive revenue.

From that need, PreBid was born solely to manage desktop display ads offering technology that was good, albeit somewhat limited.  

For Chegg, Prebid was simply a better option than their home-grown solution. 

But unlike Chegg, many organizations that are part of the programmatic ecosystem are still reluctant to get on board with Prebid. According to Downinghall, the disconnect may be due in part to the belief that it’s too difficult to implement.  

“As every publisher knows here, there’s a big cost of switching technology,” Downinghall said. “You have to test into it. You have to relearn from an engineering perspective – like how it fits, how it doesn’t fit.”

With Prebid offering access to people who Downinghall describes as “the savviest publishers” across the open exchange for driving monetization solutions centered around what’s best for the publisher, however, “difficult to implement” can be removed as an excuse not to use Prebid solutions.

Organizations that are already on board may also not be taking advantage of the full breadth of services that Prebid offers them. That’s because Prebid has now grown from display ad management to something far greater, including: 

• Prebid Mobile for iOS and Android
• Prebid Server which plugs into pretty much anything if you want server-side options
• Video and native accelerated mobile pages.
• Access to insights from thought leaders and the overall roadmap of the Prebid community.

The development plans don’t end there. The evolution of Prebid continues with the goal of expanding the breadth and depth of open-source monetization tools to include:

• UID2
• Standardizing taxonomy
• Helping publishers with single-sign-on technology
• Testing standards established by IV Tech Lab

“We’ve been increasingly fielding requests for Prebid,” said Levesque, “If you have a tool that you’re building for publishers that helps them in one way or another with their monetization and you think it’s important for that tool to be open source and community-driven, we have committees and working groups for [those initiatives].”

The post Prebid.org: The Braintrust of Open-Source Advertising Solutions appeared first on AdMonsters.

“It was like a magic reset button or a pause, one pub said at a recent Think Tank with AdPushup. “Everything had stopped. So let’s redo our website. Let’s make it faster, let’s go through and do that. And one of the resulting things has been a little bit more strategic conversation.”

As the Think Tank progressed, it became clear that pubs are truly taking a ‘nowhere to go but up’ mentality this year, making most of the dregs befalling ad tech to conduct thorough audits of their partners, re-evaluate their place in the client-side vs server-side header bidding ecosystem, improve their Google Core Web Vital scores and, ultimately, offset revenue losses from 2020.

Pandemic Times Allow for Thoughtful Latency Audits 

Asked how they see latency affecting revenue in days leading up to the cookie’s demise, one pub joked, “All I know is that, instead of 10 SSPs hitting me up to get integrated, now I have 10 ID solutions hitting me up to get integrated.”

He added that he believes ID solutions are the right way to move, but that there needs to be more consolidation.

Another pub agreed, explaining how the pandemic allowed his team to reset with the dev team and show them what was truly valuable to them now.

“Ad dollars had slowed down for a time period, and [it allowed us]  to really strategize a roadmap that really kind of helped benefit us in the second half of the year,” they said.

“All the advertising dollars kind of came back to us. We pushed our sales team to really dive into programmatic and guaranteed a lot of lean on our efficiencies there. And it really worked out for us. And we’re seeing this year start off really strong already. So that pause, seeing who is benefiting us and optimizing our stack was a big benefit of this.”

For mobile-first publishers like Ed Arrandale, Head of Programmatic at WeatherBug, figuring out an ID solution, and how that will impact the user experience, as we move closer to entering a post-IDFA world has been top of mind. “We’ve looked at pretty much every identity solution there is and talked to every partner that we could talk to. And really for iOS, there is no perfect solution. We think realistically it’s going to be in the 20% range [of users] that will actually say yes, and that is with a thoughtful prompt and being a brand with dedicated loyal users.”

Overall though, consolidation was a big theme of the Think Tank, with a focus on pandemic audits and a re-evaluation of the partners who will help pubs better navigate latency, particularly as it impacts revenue.

Reassessing Latency Around Google Core Vitals 

One pub said that an audit of his prebid stack with 10-15 demand partners prompted the question, “why do we need so many resellers?”

“So we’re kind of going down that path of speed, supply path optimization, as well as making sure the ads don’t do that fun CLS thing for Core Web Vitals, and a couple of other things just to make sure the ads aren’t slowing down the page.”

For this publisher, over the past nine months, it’s been important to take the approach of quality over quantity when it comes to speed and efficiency.

So we're kind of going down that path of speed, supply path optimization, as well as making sure the ads don't do that fun CLS thing for Core Web Vitals, and a couple of other things just to make sure the ads aren't slowing down the page.

Another pub explained how the Google Core Web Vitals conversation also ties back to their team conducting a lot of A/B testing. Some of that testing has included looking at how ads load on page and making sure they’re loading separately from content.

They explained how the site has a right rail with additional sponsored content suggestions that are being populated by an internal algorithm.

“Our site is not always sponsored, but it’s sort of recommended reading for everyone and that was starting to push the ads down, which also then pushed the rest of the content down,” they explained. “And so we actually had to go back and reconfigure that part of our sites, because not only did it affect our Google Web Vitals score, but it actually completely killed my CPRs for my half-page ad. So we told them that that wasn’t gonna work.”

Another pub said that they’re finding success with Google by making sure all ad placements have a fixed height, no matter what they contain. 

“We actually took out our 970 x 250 in favor of a 720 x 90 in the masthead, which didn’t really impact revenue that much,” they said. “Most of our traffic is mobile. So it was hurting us on desktop. It wasn’t a huge portion of what was being sold. And, you know, as we have direct sales coming, that will probably need to come back. But we’ll just figure out how we need to do it because it was all about CLS and speed.”

Why There’s No Set Healthy Number of Prebid Partners

One publisher, acknowledging his privilege of having a Google team he could go to audit his site and provide the dev team with a list of action items so that those things could be fixed on the site. He acknowledged that the dev team is great, but has different ideas about how things should work compared to how Google thinks things should work. 

When it comes to A/B testing, the pub discovered that creating custom high-impact units they could sell became a problem, especially with page latency and Google-heavy ads. It was important to create different versions of these units that work within the guidelines from Google. This involves multiple testing templates, such as biweekly updates to individual ad units based on what the reporting might say. 

It's really just being able to have enough data and identify where you can make updates rather than having some overall look at what Google says on latency.

Also by working with Pubmatic’s OpenWrap for Prebid, they have built-in A/B testing on any wrapper.

“I can go into any site across our network and say that I want to test a new time out for 10%, or 25% of the traffic on just this site for this time frame,” the pub said. “And they automatically are able to filter that traffic. So to create an A/B testing scenario for me, I don’t have to work on my dev team at all. I’m also able to do it with other SSP partners.”

“It’s really just being able to have enough data and identify where you can make updates rather than having some overall look at what Google says on latency,” he added.

The same pub went on to say that working with OpenWrap has allowed them to see, for instance, if one partner is timing out much higher than others. This ultimately helps them better consolidate partners within Prebid, making for a simple and effective way to reduce latency on page.

He added that there shouldn’t necessarily be some arbitrary rule for how many are on page, because every site is different.

For AdPushup, the future lies in building optimization back for pubs. They’ll build rule engines, marry their research and analytics to the machine learning they have for their testing algorithms.

“Right now there is a lot of manual testing going on in the Prebid stack,” said Dikshant Joshi, Director of Publishing Development, AdPushup. “But marrying the data that we bring into the analytic solution and the capabilities to test it out with machine learning algorithms is something that we are looking at as something beyond the standard Prebid conversations.”

Moving Toward a Hybrid Solution For Client-Side and Server-Side Bidding

Another pub said that they see pros and cons to working using client-side and the server-side, and as such, they use both.

“We have a complex setup where we kind of tap into different partners in different places,” they said.

Other pubs agreed with this approach. “I think that the solution for publishers is just [having] as much data as you can, look at it, and see what works best for you,” another added. “And I’m definitely confused sometimes because I do hear this conversation and I have vendors reaching out to me, [saying] that the benefit of using them is their server-side.”

And if you’re thinking one of the benefits of Prebid is that Amazon isn’t involved, you might want to rethink that.

According to one pub, Amazon’s demand is enough of a benefit to stay with them. “They just kill it on the open exchange,” they said. “So even if I just stop using TAM, I would just keep Amazon’s demand up there if I could.” 

“I think the hybrid solution is something that you’re indicating towards, with Amazon running client-side bidding and open bidding that’s all configured,” said AdPushup’s Joshi.

“Yeah, it’s kind of unfortunate that we do have to have all these integrations, though” replied the pub. “It would be nice to have more consolidated solutions, but we’ve created this ecosystem where we’ve had to have all this diversification and if you don’t have it, then that’s when you get hurt the most.”

The post How the Pandemic Helped Pubs Think More Strategically About Latency appeared first on AdMonsters.

I’ve thought of describing it as an embargo. Maybe it’s more of a wall or a security checkpoint. Let me describe the situation and perhaps you can help me come up with a name:

The revenue side of the business will be in perfect synchronicity with the development team. Adjustment to existing code is being made on time, new money-making products are being rolled out. Things are flying along, and then… they aren’t.

It can be a number of things. It can be conflicting priorities with other departments. It can be that something ad ops thinks is simple, is in fact not simple at all. It can be that someone makes a mistake and a new layer of approvals need to be implemented.

Whatever caused it, the result is an ad ops-to-development team traffic jam.

Here’s the thing — if there is anything we know about operating in the digital world is that the lack of speed kills. That logic applies to the speed of your page and that same logic applies to how your departments work together.

If I can update prebid today, I can make thousands of dollars more than if I have to wait two weeks.

If I can update compliance code for GDPR quickly, there’s less chance of some EU regulators knocking on my door. I’ve only met one EU regulator in my life and I don’t need to meet another one (story for another time).

The part of this that can be so frustrating is these are two departments that most often speak the same language. Operations people typically aren’t coders, but they understand code to some extent. Troubleshooting a campaign is not unlike debugging code.

So what do you do when your process for updating ad tech is more similar to the “cash-only” slow lane instead of the E-Z pass lane?

Well, you build more E-ZPass lanes for one. To mix my analogies further, the real key is building a bridge with all E-Z Pass lanes between the two departments so that information and implementations can move back and forth more efficiently. Both teams should work to create this together: Developers will need a strong bridge, so they know they can support this new infrastructure and ad operations need it to be quick moving – traveling at the speed of business.

How do you do this?

Ad ops should talk revenue to business but build trust with developers. Make sure management knows that what you want more than anything else is to get revenue going up. If your process with developers is slowing you down, make sure they know. The CEO and the CTO set priorities. You won’t always win, and you won’t always get your way, but at least you should be heard.

That revenue argument, however, is not going to work as well with the development team. What they need is security. Code changes mean risk. Your requests can endanger the delivery of the site – which is their job #1. If you build trust – that you know what you’re doing and that you’ve done everything you can to mitigate risk, there is a chance they’ll open up an EZ-Pass lane for you. As an ops person, you can’t build this on your own. Loop developers into the process from the start.

John Shankman, CEO of Hashtag Labs, introduced me to the concept of an Advertising Management System. Much like a Content Management Systems (CMS) is for editors to update the website, an Advertising Management System (AMS) like HTL BID allows ad operations to manage header bidding settings and even keep up on the latest Prebid configurations all within one centralized UI and without needing the development team to do code changes. You can optimize viewability at the ad slot level.

What I particularly like about this concept is that something like updating an ad slot isn’t even worth an additional email. Adjusting existing ad code in order to optimize the ad serving and tech setup is something that operations people should be able to do on their own and an ad management system enables that. Developers have plenty of other things to work on without these tasks. They just want to make sure the bridge is safe and secure.

Process is key and where a lot of publishers get tripped up is the connection between departments. Operations to Development is no exception. However, a good Ops/Dev relationship is a differentiator and a revenue generator. If the walls go up, you need to bring them down and establish a good way to work together. Maybe it’s the New Jersey in me, but I’ll take E-ZPass every time.

The post Advertising Management Should Be More Like E-Z Pass Lanes Between Dev and Ops appeared first on AdMonsters.

Recently named to the Board of Directors at Prebid as a publisher representative, CafeMedia’s Patrick McCann is tasked with ensuring that publishers’ interests are heard throughout the organization and to communicate those interests to developers.

McCann took to the main stage at AdMonsters PubForum Virtual a few weeks back to provide an overview of the latest rev of Prebid.js, Prebid 4.0.

“Prebid 4.0 is a pretty big release in the sense that it adds a focus on trust. Trust in our partners and trust that the software is compliant with various regulations and that communication is standardized and you don’t have to do customizations for every partner that you have,” said McCann.

Watch his full talk in the video below to learn more about the updates to Prebid.

You might also want to read McCann’s recent article on AdMonsters, “Sellers Need Buy-Side Transparency,” where he wrote:

At Prebid.org, much of our recent work has been around standardizing how buyers and sellers communicate with each other. In the past, prebid adapters have consumed information in very different ways, but as that type of information becomes common to transmit, publishers tire of configuring it for each demand source in a different way…With Prebid 4.0, much of the standard ways of communicating will become recommended behavior, with an eye towards making them required in future major releases.

The post PFV Video: Under the Hood With Prebid 4.0 appeared first on AdMonsters.

In late 2017 and early 2018, the buy-side made huge inroads against domain spoofing with widespread adoption of ads.txt. No longer is not knowing if a traditional display bid request is misrepresenting its domain a widespread concern, and sellers have begun to focus on transparency of the intermediaries in a transaction with the sellers.json standard and extending this transparency into the mobile application and connected television markets.

Unfortunately, sell-side participants—publishers—often have difficulty obtaining basic levels of transparency in reverse when using header bidding tools to extend demand.

Transparency Is Pubs’ Partner

Transparency is critical to sellers for several reasons. Top of mind for many sellers is not scaring away their audience or decreasing their engagement. This can happen if brands and content do not align—for example, a meat advertiser on a vegetarian recipe site or an online gaming advertiser appearing on a competitor’s site. It can also happen if malvertising is redirecting users to another site, or with invasive advertising for untrustworthy e-commerce platforms.

Transparency is also key in relationship management: perhaps the seller has a deal with an endemic advertiser and wants to make sure all buying from that advertiser goes through that channel. Sometimes sellers are looking to discover which advertisers prefer their inventory so they can pitch custom ad products.

Ad operations teams are particularly interested in transparency. Certain creatives may drive discrepancies between an SSP and a publisher’s transaction counts. Another reporting dimension would be very helpful for teams tracking down these discrepancy sources.

Often a wayward mis-categorized ad that sneaks through existing protections is reported by a reader and the ad operations team has difficulty discovering its origin. A high VAST error rate might be plaguing a video demand source without an easy way to identify the source creative asset.

Within Google AdManager, there exists rich protections, reporting, and creative review of AdX, and to a more limited extent Open Bidding demand. In the header, vendors have tried to fill this gap: Confiant, Clean Creative, GeoEdge, and others offer malvertising and creative content or advertiser blocks. While malvertising protection is well developed, the latter is certainly a work in progress.

Communicating Across the Aisle

Adomik, Ad-Juster, and others offer tools to see who is buying your inventory across your many header demand sources. The IAB provides some specifications for communicating buyer seat and domain, but they are only recommended. Adoption of the Ad Management API (which will provide the buy-side equivalent of sellers.json) is a work in progress.

At Prebid.org, much of our recent work has been around standardizing how buyers and sellers communicate with each other. In the past, prebid adapters have consumed information in very different ways, but as that type of information becomes common to transmit, publishers tire of configuring it for each demand source in a different way.

Examples include consuming video parameters from the ad unit and not the bidder parameters, consuming consent, identity, and floors from their respective modules, and reading from a common first-party data interface. Conversely, publishers are demanding a more standard and richer bid response.

With Prebid 4.0, much of the standard ways of communicating will become recommended behavior, with an eye towards making them required in future major releases. One of these recommendations is called the bid.meta object. It is the part of the bid response with information such as the advertiser domain(s), buyer seat, and disclosures on rich media formats. Prebid analytics adapter providers can report aggregations of these fields to publishers who have their adapter installed.

A header bidding key for advertiser domain(s) will soon allow publishers to set universal price rules and protections in their ad server. Publishers who define the ‘hb_adomain’ key and advertiser domains of interest as values will be able to easily obtain reporting on header advertiser impression count and price paid cut by demand source. Publishers with the ad server logs or analytics adapters will have full transparency into adomain and buyer seat when header demand partners are setting it.

Demand More From Your Demand!

Adapters setting this information remains the key; the bid.meta object in the bid response object currently only has limited adapter support. Your communications with your SSP partners to keep their adapters modern (eg supporting the floors module, the identity module, the bid.meta object, and reading video ad unit parameters) is critical to the success of the Prebid project.

Publishers with other header bidding integrations should be demanding their provider keeps up, with the end goal being full transparency of buyers.

The post Sellers Need Buy-Side Transparency appeared first on AdMonsters.