Advertising network

An ad network, or online advertising network, is a business that links advertisers with websites or apps that want to show ads. Its main job is to find ad space that is available from a lot of publishers and match it with advertisers who want to reach certain groups of people.

Ad networks used to mean networks for TV or print ads, but now they mostly mean networks for online ads. The main difference between traditional and online ad networks is the technology they use. Online networks use a central ad server to send, follow, and measure ads in real time. Compared to older, non-digital media, this lets you target better, track performance better, and get more detailed reports.

This section needs to be updated. Please help update this article to reflect recent events or newly available information. (March 2014)

The advertising-network market remains large and growing. Worldwide total media ad spending is forecast to reach around US$ 979 billion in 2025, up roughly 4.9 % from 2024.Of that, digital-platform ad sales are projected at about US$ 715 billion in 2025.

The inventory of online advertising space comes in many different forms, including space on the desktop and mobile websites, in RSS feeds, blogs, instant messaging applications, mobile apps, adware, e-mails, and other media. The dominant forms of inventory include third-party content websites, which work with advertising networks for either a share of the ad revenues or a fee, as well as search engines, mobile, and online video resources.^[1]

An advertiser can buy a run of network package, or a run of category package within the network. The advertising network serves advertisements from its central ad server, which responds to a site once a page is called. A snippet of code is called from the ad server, that represents the advertising banner.

Large publishers often sell only their remnant inventory through ad networks. Typical numbers range from 10% to 60% of total inventory being remnant and sold through advertising networks.

Smaller publishers often sell all of their inventory through ad networks. One type of ad network, known as a blind network, is such that advertisers place ads, but do not know the exact places where their ads are being placed.

There are several criteria for categorizing advertising networks. In particular, the company's business strategy, as well as the quality of the networks' traffic and volume of inventory can serve as bases for categorization.

Online advertising networks can be divided into three groups based on how they work with advertisers and publishers:

1. Vertical networks: They represent the publications in their portfolio, with full transparency for the advertiser about where their ads will run.^[2] They typically promote high-quality traffic at market prices and are heavily used by brand marketers. The economic model is generally revenue share. Vertical Networks offer ROS (Run-Of-Site) advertising across specific Channels (example: Auto or Travel) or they offer site-wide advertising options, in which case they operate in a similar fashion to Publisher Representation firms.
2. Blind networks: These companies offer good pricing to direct marketers in exchange for those marketers relinquishing control over where their ads will run, though some networks offer a "site opt out" method. The network usually runs campaigns as RON or Run-Of-Network. Blind networks achieve their low pricing through large bulk buys of typically remnant inventory combined with conversion optimization and ad targeting technology.
3. Targeted networks: Sometimes called "next generation" or "2.0" ad networks, these focus on specific targeting technologies such as behavioral or contextual, that have been built into an ad server. Targeted networks specialize in using consumer clickstream data to enhance the value of the inventory they purchase.^[3] Further specialized targeted networks include social graph technologies which attempt to enhance the value of inventory using connections in social networks.^[4]

Ad networks can also be divided into first-tier and second-tier networks. First-tier advertising networks have a large number of their own advertisers and publishers, they have high quality traffic, and they serve ads and traffic to second-tier networks. Examples of first-tier networks include the major search engines. Second-tier advertising networks may have some of their own advertisers and publishers, but their main source of revenue comes from syndicating ads from other advertising networks.

While it is common for websites to be categorized into tiers, these can be misleading because tier 1 and tier 2 networks might perform differently based on different metrics, such as reach versus impressions.

One aspect of ad-serving technology is automated and semi-automated means of optimizing bid prices, placement, targeting, or other characteristics. Significant methods include:

• Behavioral targeting — using a profile of prior behavior on the part of the viewer to determine which ad to show during a given visit. For example, targeting car ads on a portal to a viewer who was known to have visited the automotive section of a general media site.^[5]
• Contextual targeting — (also known as Semantic Marketing) refers to the optimum ad placement as a result of analyzing information from the entire Web page where the ad is being served. This concept was introduced as a way of improving the ‘keyword approach' to ad placement were issues surrounding ambiguity in relation to a word's meaning in the prescribed context. The concept of analyzing the ‘entire' Webpage in order to promote relevant advertising material is to benefit both the viewer of advertising content and the source of the ad. Keywords (or Adwords) are not always relevant in the context in which the word is intended. Therefore, by analyzing the entire page rather than just the keyword, the ambiguity is removed and a more relevant and accurate ad is promoted into the advertising slot on the Web page.^[6]
• Creative optimization — using experimental or predictive methods to explore the optimum creative for a given ad placement and exploiting that determination in further impressions.

Ad networks often support a wide spectrum of ad formats (e.g. banners, native ads) and platforms (e.g. display, mobile, video). This is true for most ad networks. However, there also are ad networks that focus on particular kinds of inventory and ads:

1. Mobile ad networks, focus on the traffic generated via mobile web and mobile apps, and work with the corresponding ad formats.
2. Video ad networks serve ads via inventory, associated with online video content.

Video and mobile ad networks can be acquired by larger advertising companies, or operate as standalone entities.

1. Placement / lack of clarity in placement A lot of ad networks don't fully say which websites or apps their ads are on. This means that advertisers may not know exactly where their ads are shown, and if their ad is shown next to content they don't want, it could hurt their brand.
2. The risk of malware and malvertising Some networks or their partners may let bad or low-quality advertisers on, which can lead to ad placements that deliver malware or unwanted behaviour (also known as "malvertising").
3. Price transparency: Although an ad network may sell inventory to an agency at a fixed cost per thousand impressions (CPM), many of the impressions within the network come from lower-quality publishers at significantly lower CPMs. As a result, the network's margin may be high and the effective value (eCPM) may be significantly lower than the amount the agency was given.
4. Ad relevance: More often than not, the ads were out of relevance with the website content as a fall out of point 1, and also because there weren't intelligent contextual engines built into the ad servers (the server system that churns out the ads) of these ad networks.
5. Privacy & tracking limitations: With third-party cookies being phased out, increased regulation of user-data tracking, and more control being placed in “walled gardens” (big platforms), ad networks are under pressure to show transparency in how they collect data, target users and measure outcomes.
6. Algorithmic opacity / measurement issues:Advertisers may find it difficult to comprehend how algorithms determine placements, CPMs, or audiences, which raises questions about fairness, bias, and effectiveness. Research indicates that many ad-tech systems (including ad networks) use sophisticated machine-learning / AI for targeting and pricing.

Most online ad-network platforms offer website owners and marketers to signup as advertising publishers. Publishers can then display ads shared by the advertising network and the revenue is shared between both the advertising network and publisher. When the beginners could not pass through the minimum criteria for publishing advertisements, ad placement services could ban the publisher for not fulfilling the requirements. Some networks demand strict terms and conditions while there are other ad publishing alternatives times commissions vary on what sells otherwise user still to earn a good commission when one matches the criteria, and the publisher is allowed to display and share ads provided by the platform earns a good revenue. Getting approved as a publisher of the best advertising platform is a thorough process. Websites with a clean interface, more traffic and engagements are preferred to be selected as ad network publishers by the advertising platforms.

On October 27, 1994, the first online ad was posted. It was a banner advertisement added to a web page. The website that posted the first ad was a precursor to today's tech site, Wired.The first central ad server was released by FocaLink Media Services and introduced on July 17, 1995,^[7][8] for controlling the delivery of online advertising or banner ads. Although most contemporary accounts are no longer available online, the Weizmann Institute of Science published an academic research paper documenting the launch of the first ad server.^[9] The original motherboard for the first ad server, assembled in June 1995, is also preserved. FocaLink re-launched the ad server under the name SmartBanner in February 1996. The company was founded by Dave Zinman, Andrew Conru, and Jason Strober, and is based in Palo Alto, California. In 1998, the company changed its name to AdKnowledge and was purchased by CMGI in 1999.^[10] The AdKnowledge name was subsequently purchased by a company in Kansas City in 2004, which now operates under the brand name AdKnowledge.

The first local ad server was released by NetGravity in January 1996^[11] for delivering online advertising at major publishing sites such as Yahoo! and Pathfinder. The company was founded by Tom Shields and John Danner, and was based in San Mateo, California. In 1998, the company went public on NASDAQ (NETG), and was purchased by DoubleClick in 1999. NetGravity AdServer was then renamed to DART Enterprise. In March 2008 Google acquired DoubleClick. Google has continued to improve and invest in DART Enterprise. The latest version of the product was renamed and shipped as DoubleClick Enterprise 8.0 on September 28, 2011.^[12]

• Affiliate marketing
• Ad exchange
• Contextual advertising
• Digital marketing
• Landing page
• Online advertising
• Pay-per-click
• Pay per play
• Surveillance capitalism
• Website monetization
• View-through rate