I have been running a website with advertising since 1999. I started out with Burst Media, I added Flycast, then FastClick, Tribal Fusion, Casale, and Adsense (plus a couple of other one-hit-wonders over the years). Things have worked out moderately well with that crop of ad networks (at least the ones that still survive - RIP Flycast!). However, I have noticed that the landscape seems to be shifting lately. Burst Media was purchased a couple of years ago, their passionate founder, Jarvis Coffin, hasn't been on the scene i. They shut down their own internal publisher message forum over a year ago, and they shifted to a net-60 on payment around the same time. I get horrid CPMs from them, so I barely send them any traffic. Another advertiser I was using, Olive Media, shut down their network a couple of months ago. They were primarily a Canadian network operated by the Toronto Star - I had been with them for almost six years. FastClick was purchased by ValueClick a couple of years ago, and they recently have shifted their identity to "Conversant" - though their user interface is still the old FastClick software. I see middling results from them, the CPM is never that high, and the advertisers can be sketchy (suggestive/sexual ads on occasion, and things like auto-play audio - though they are not as bad as they used to be). Tribal Fusion rebranded themselves as Expo9 a couple of years ago. They still perform pretty strongly for me, with a good selection of higher brand advertisers. They have their ups and downs though - I can never seem to achieve a good CPM with them for skyscraper ads, and they are very cyclical with CPMs, with January being really, really bad. They are much stronger on US impressions than on Canadian impressions (and my site is about 68% Canadian). Casale just rebranded themselves as "Index Exchange". I have been having good success with them too, but things have turned a bit downward in 2015. This could be related to the Canadian dollar (they are a Canadian company) which has lost about 30% in relation to the US dollar over the past six months. They have always been better for Canadian impressions. Google Adsense still chugs along, it never achieved a CPM comparable to the top-tier networks of Casale and Tribal Fusion, but lately the spread is smaller due to those other networks losing ground. I mostly use them as a fallback. Overall, the CPMs I have been seeing over the past 3 years are disconcerting. I keep meticulous stats on my revenue and impressions; I saw a 4% CPM increase in January of 2012, then a 2% decrease in January 2013, then a 9% decrease in January of 2014, and then a 20% decrease in January 2015. February is shaping up to be even worse - down 47% from last year. That trend does not seem promising. It has been less painful due to my traffic increases of 12% in Jan 2012, 22% in Jan 2013, 22% in Jan 2014, and 25% in Jan 2015 - but an 86% increase in monthly unique visitors from 2012 to 2015 has brought me a 0.6% decline in revenue! I suspect that part of the CPM drop is due to a shift to Mobile. In 2012, my Mobile traffic was about 25%. Today, it is about 60% (I believe that number, from Google Analytics, includes tablets). My site is not yet a responsive design, so I have to imagine that people are not interacting with the ads as much. [Although most ad networks report payments based on CPM, I suspect that ultimately these ads generate revenue based on blended formulas of CPM and click percentage - at the very least, the networks optimize their ad allocation based on performance.] I am planning on moving to a responsive design, but the main issue I am having is that it is very difficult to place CPM IAB ads on a mobile site. My site right now generally serves 2 ads per page - a Leaderboard and either a skyscraper or a box (depending on browser width and page design). There is an equivalent unit to replace the skyscraper on mobile devices, and I can get such a unit "above the fold", but I can't get 2 ad units above the fold - I would need to integrate a second unit within the content, below the fold, and I don't know if the companies I use will allow this. I'm worried that if I go responsive, I will see a huge drop in revenue due to fewer ads per page coupled with lower mobile CPMs - I think it could be as much as a 60% revenue drop, which would be debilitating. My site is over 10 million pageviews per month, on about 1.4 million monthly uniques. In January, I saw an overall effective CPM of under $0.75 - coming from 2 ad units per page. That strikes me as pretty low, but the only thing I have to compare against is my own history. Historically, I haven't seen CPMs that low since just before the stock market crash in 2008. So after all that, my question is, what I am I missing out on? In which direction should I go? Are there any other top-tier networks out there? Or are Casale and Expo9 the best? I get at least 3-4 emails weekly from companies I have never heard of who are "looking to buy inventory" - but they all seem questionable to me. I cut out popunder ads about 5 years ago, and don't like the idea of super-intrusive advertising (especially those mobile units that hang at the bottom of a page). Is that the direction that networks are moving? Can someone explain "exchanges" to me? Is this something I should be looking into? Are they better than the old-school CPM networks? When Olive Media shut down their network, they said that they would still be buying traffic "on the exchanges". This leads me to believe that I'm missing out on something. What options are out there for medium-size sites? Thanks!
Hey Ralphslate, You're completely right about mobile market, auditory is moving from desktop to mobile devices. So, there you'll be able to pick more visitors because mobile devices is simply more accessible nowadays. Have you tried Media Net? It's microsoft and yahoo company. Do you have a lot of unsold traffic?
For this one, I better receive some likes. lol Thanks for the tag btw sorry for being late to this, was on a business trip. So first lets address history. Before "networks" came along everything was what I like to call "Analog Direct" meaning each independent buyer would go to a site / publisher request ad space, the publisher would either set a random rotation, or would roadblock (occupy a placement for a set period of time). Over time publishers realized, wait, I don't have time to go out and sell all of my ad space for every country, every language, and still be productive in creating a good site with good content... incomes the birth of the ad network. The original ad network model which is called a Network 1.0 Business Model was entirely IO (Insertion Order) based. So every advertiser filled in an IO, every publisher got an IO from the network. Some networks skipped IO's for publishers making it easier. But Networks allowed publishers to source demand services and sales while focusing on what is important about their site. Now networks needed to innovate to be able to sell more inventory (a wider array of inventory) at higher rates so they can make margins to survive also. In comes the birth of Network 2.0 which now includes programmatic. With the Network 2.0 approach you have IO's still being filed but also programmatic (at this time super infancy not really called programmatic yet) / rtb buys taking place. What happened was the natural market realized that if networks sourced demand back and forth (exchange) that they would have fill / demand for geo's and users which might not have been previously covered under the Network 1.0 Model where it is straight IO buys. There are a lot of IO's to be had to fill 100% of 100% of your publishers inventory. So incomes the birth of the RMX era. Right Media Exchange. Allowed networks to sell inventory quicker to other networks and allowed publishers to get more for their money. Around this time, transparent buys became far more abundant until the industry made some heavy shifts away from transparent in the 2008 - 2009 period as a result of fraud. So here a system is a adopted which allows for more volume to push through at higher fills but is a safe haven for fraud since someone can put the tags anywhere and send anything to it, 2008 - 2009 is when this came to an industry headway and literally shifted buyers away from transparent and more toward site specific. Now what you are seeing in present day is where the model of networks is being broken down into 2 basic segments, Supply and Demand. Supply being the Publishers and their inventory. Demand being advertisers and their inventory. Demand Platforms are linked into exchanges, and demand is pumped across exchanges, the same demand might be trafficked direct for some networks, indirect for others. But what happens is, because it is such a tight knit industry, if a publisher is using 2 different networks both with the same advertiser but 1 is charging more than the other, the advertiser is going to stop buying the more expensive one, causing bid floors at that price point on the more expensive one to drop, and effectively generating less revenue for the publisher since the publisher is not able to let them directly compete as 2 auctions are taking place separately. As transparency and fraud become even bigger issues, fraud now makes up 30% of all digital media buys in an industry which makes up for more than $50,000,000,000 in yearly ad spend. Now, we are in the beginning stages of pressure on fraud standards and view-ability buys aswell as a slide toward programmatic direct. Especially as trading desks face increasing risk on buys. Industry leading buying desks and agencies like Xaxis, now make major buying decisions based off of view-ability where back in 2008 most networks didn't even track or know to track view-ability. What it has come down to, is buyers wanting to make sure they are getting what they are paying for, and they are getting their money's worth. When Olive Media shut down to buy programmatic from exchanges, they are doing it, because there is a wider assortment of inventory, they do not need to have the overhead of managing an account, book keeping, support, etc, they can go to a single platform, single partner, and buy from a wide array of inventory. This is the function of a demand side platform os DSP. It allows advertisers access to supply across the ad tech ecosystem. Someone in your position first off needs to know these things. You need to be able to tell the difference between a company which understands the landscape and has your best interests in mind vs a company which is blind to the understanding of the industry and is just looking to broker / make a margin where ever possible. I once heard a company tell me, we used to have a seat on AppNexus but their rates suck so we shut it down, and only work with AdX. I honestly smacked my forehead in disbelief. I said, that is the largest exchange in the world. Surely there is high end demand on there. Sure enough working at a place where I can regularly disprove many different theories about the industry, we put it to the test, and found when Bidding 2 exchanges against each other in an RTB Environment you are likely going to get higher rates than just 1 would provide, assuming the priorities and the bid auction is being processed correctly. So let me give you an example, if I have a publisher with a site that does 10,000 impressions per day, and 8,000 of them go to AdX at $1.00 on average and the other 2,000 go to AppNexus at $1.10 (because it out bid AdX) my combined revenue would be $10.20 with an eCPM of $1.02 which is actually higher than if I just used AdX alone. You do this on a mass scale and that $0.02 difference in rates could be the difference of thousands, or it could be the difference of profitability. So in your position, you want a network who understands that demand is important, and having enough demand to fill inventory even with ad quality rules and restrictions which might need to be set to meet the requirements of your site or users is extremely - EXTREMELY - important. Next there are aspects to media buying which most people engaging in the selling side (specifically publishers) are not privy to such as the need for a unique site design, clean history (no prior blacklists), a good alexa rank, a good comscore, they are starting to get bigger into nielson OCR Ratings, you need high visibility, and you need clear social engagement. Amongst those you also need accurate information in your Domain Whois. All of these are things which dictate risk, all of these things being perfect mean extremely low risk, extremely valuable inventory, which will cause an exchange / RTB environment to go through the roof. Couple these characteristics with a niche site in a high demand / value niche, we have seen sites with RPM's as high as $40+ Top tier networks - is a grouping of networks by sheer size of client base, there are many other networks out there which will likely out perform them if they are doing business correctly. Unfortunately there are other networks which lie about things such as number of impressions passed back, the actual amount of revshare you are taking, potential revenue based on demand, and a whole slew of other things including perpetuating fraudulent traffic to boost numbers and revenue. What has happened in the industry is it has shifted so much so fast, that the pace which networks can educate their employees is far out paced by the rate of change. Leaving technology falling behind. Mobile is a great example, mobile is huge right now, direct buys are through the roof and mobile specific networks are helping publishers make a killing, but the exchange environment in mobile has not adjusted, it has not adjusted to account for technology differentials like responsive design, and there are so many demand houses out there with segments demand, that have not figured out how to integrate into an exchange environment to open up that inventory. Video is another great example, video is growing quicker than the technology to support it, and due to fraud its mostly site specific, making it extremely difficult to run video in an exchange environment, companies haven't fully adjusted to mobile and are now being forced to adjust to video account for video also. In the same breath, publishers who do not keep up with technology who might be using youtube for hosting videos were never told the value proposition of running video content on direct stream to directly monetize prerolls or midrolls or post rolls. Long story short, this industry is a giant cluster of mistakes and bandaides and a lot of people who have no clue how to apply peroxide to clean it up. Big headache. How you can protect your self and adjust, is roll with the companies pushing the boundaries and looking out for their publishers and as the industry shifts, so will the company and so will the way your inventory is sold under their guidance. But having the knowledge, asking questions, and being an active part of the process is the way to determine if you are making the right choice. Its easy to tell when a company is embedding youtube channels in ads and subsidizing your ad space for additional revenue to justify higher rates. Those are not people you want, you want people who can tell you how much bot inventory you have coming through on your site, how much your viewability is, what some of the risk factors of your inventory is, because if they can provide it to you, and know what it means and know what to look for, you should know very well that they are providing that same information to advertisers thus increasing their value proposition to you and to advertisers. Most of the networks today, do not take responsibility for this. You made a great point though - this has been one of the roughest Q1's I have seen. Publishers also don't realize that pulling tags and signing up for 50 different networks isn't going to solve the problem when they are all more then likely puling most of their demand from the same place. Some may disagree with this but after being in the industry for over 15 years, these are my observations. I have taken publishers from 0.10 CPM's to $2.00 CPM's in under 30 days just with understanding the industry, the market, and the business. Any other questions, happy to help.
This is an awesome answer, and I am very grateful that you posted it. I wish I knew how to "like" it - perhaps that is a feature not granted to new users such as myself. Casale Media (which just renamed itself Index Exchange) just announced today that they will be moving from "an Ad Network model to a programmatically enabled Supply Side Platform". I suppose this means that they will not be trying to directly sell advertising to companies, and instead will make their inventory (supply) available to advertisers in some way. To me, there seems to be little distinction - I was not getting direct insertion orders from Casale, and I doubt that anyone buying on Casale was trying to place on my site. I suspect that Casale was instead saying "we have a good portfolio of sites, you can look at the list if you want, and we will spread your ads across them using our algorithms to optimize the response". What do you think will change? Access to better inventory? Is there a down side here for me? I switched my tags a couple of days ago and I saw an immediate improvement in rates on their new platform - including better scalability; due to external events, I sometimes get a spike of visitors to my site. In the past, those days did not scale well - if I was making $50 from 10,000 visitors, if I got 20,000 visitors, I might see $70 tops. I had such a spike yesterday, Casale scaled very well; Tribal Fusion did not. What other companies should I be looking into, besides Casale, Expo9 (aka Tribal Fusion), Conversant (aka ValueClick aka FastClick), Google Adsense, Burst Media? I expect that my inventory should be considered pretty decent - though my niche isn't super-hot (sports), my content is not-at-all forum-based, nor is it blog-based - it is a pure informational site with the ads placed very visibly, but not in an overwhelming way (maximum 2 per page). As I mentioned, I get multiple requests per week from people looking for supply, I am generally skeptical because they do seem to be slimy middle-men who ask me how much CPM I "need", which hints that they are just looking to feast on a margin.
You don't mention here if you run Google AdX. My guess would be that this would be your best monetisation option. It should outperform AdSense due to higher fill rates and better revenue share. Have you tried?
The distinction between Network 2.0 VS Programmatic is Direct Sales and IO's vs "Nearly" (carefully quoted) automated demand across an RTB Platform. See a lot of people today, instead of allowing a programmatic platform to fully work they decide to run waterfalls with 3 - 5 networks all passing down based off price points and fill, and what publishers don't realize is the demand overlaps in situations like that, actually waterfall the rates downward over time, because if there are demand intersections, and advertisers see it, they will drop fill on the most expensive placements, and increase bidding on the cheapest placements dropping the value of your inventory heavily. Another great way to scale rates, is to work with remarketing companies like Criteo, and Simpli.fi. A few other companies unmentioned would be OpenX and PulsePoint (a good 2-4th tier option). Ad Sense should be a low priority fill, don't EVER EVER EVER get attracted by the high rates google shows off, its a trap to send them the bulk of your volume at which point you will eventually get fried from them. We are a Tier 1 Google AdX Partner, and we are actually dealing with them right now trying to withhold several months of very large payments. Thats why google is not to be trusted, and is to be run on low priority, if you are going to run any google products make sure you are running a 3rd Party antifraud platform on all of your inventory sent to them so in the event of a dispute you have grounds to pursue legal options. If you are smart, you work with an RTB / Programmatic based network, and allow them to force the RTB demand to outbid platforms like google. We have clients on AdX and our external RTB demand outbids and fills inventory over AdX 50-80% on select sites.