Results 1 to 13 of 13
  1. #1
    Newbie
    Join Date
    January 18th, 2005
    Posts
    1
    We're all losing money. -Here's why...

    I believe that due to EPC (Earnings per click) being the most important stat to Advertisers to make their program look attractive versus EPM (Earnings per banner impression), that the banners that COULD be making Publishers the most money are not being promoted.

    Let me explain:

    Advertisers want their EPC stats to be as high as possible, because when potential “heavy hitter” new publishers surf CJ looking for advertisers to promote, that’s the main stat they use to judge the attractiveness of a program.

    BIG PROBLEM, here’s why:

    Let’s say an advertiser has two banners they can make available to their publishers.

    Banner1 is viewed by 10,000 people and gets a 1 in 100 impressions clickthrough rate and a 10% product purchase rate for those who click through. Let’s say the product pays $5 commission, that’s a $50 EPC. (Because for every 100 people who click through, 10 buy @ $5 commission apiece)

    In the above example, 100 people clicked through the banner and 10 of them bought, earning you $5 apiece = $50.

    Banner2 is viewed by 10,000 people and gets a 10 in 100 clickthrough (ten times the clickthrough of banner1) and a 5% product purchase rate for those who click through.

    In this second example, 1,000 people clicked through the banner and 50 of them bought, earning you $5 apiece - $250. Much better, right? Except the EPC on this second banner is $25, only half the EPC of the first one.

    Here’s the problem with Commission Junction’s stats of only reporting EPC and not EPM when people are looking to sign to promote a particular program:

    Even though the advertiser AND the publisher would make more money using Banner2, it will HURT the publisher to use Banner2, because it’ll make their EPC look twice as bad, which means that when new publishers are browsing 1,000 programs to make quick choices of who to promote, their reduced EPC from using Banner2 (which is more profitable but looks worse from an EPC-only point of view) will hurt their rating and possibly make new potential advertisers pass them over in favor of their competitor (who may be using a less profitable banner that simultaneously gives them a better EPC).

    End result? Advertisers ultimately would prefer to have a higher EPC which will be attractive to new potential “Super Affiliates” than to use a banner that actually earns more money overall for their publishers and themselves – Because the advertisers will benefit in the aggregate from having more affiliates sign up to promote them, even if the earnings of EACH affiliate isn’t maximized.

    By the way, I realize that EPM is available to individual publishers in their reporting once they start promoting a specific program. I also realize that if aggregate EPM was reported right along EPC when selecting companies, that non-targeted publishers heavily promoting a specific company could really screw up the EPM for that company. I’m not saying that there’s not reporting drawbacks to EPM as well, but by not being able to factor EPM in at ALL when choosing advertisers to promote, there will continue to be strong motivation on the side of the advertisers NOT to create banners that maximize EPM stats while hurting EPC stats, which ultimately cause all individual publishers (as well as advertisers) to lose money.

    What are your opinions on this?

    (Ultimately, EPM really IS our most important stat as publishers. For every banner view, how much do I make?)

    _____________________________
    (C) 2002 All Rights Reserved

  2. #2
    2005 Linkshare Golden Link Award Winner  ecomcity's Avatar
    Join Date
    January 18th, 2005
    Location
    St Clair Shores MI.
    Posts
    17,328
    Your logic is correct and shows why merchant AM's look for any way to boost EPC while dropping the commission payout. Seems strange they'd do that since the affiliate not earning money ceases to showcase them. Most AM's work on all forms of their company advertising and at CJ many are the company owners. The almighty click ..not the sale is what drives them. They spend all their time and effort on post visit sales activity.

    Simple cure for all networks is to publish a merchant conversion ratio based upon networkwide stats. Number of clicks divided by number of sales = conversion ratio. All affiliates then could try to match or beat this conversion ratio with pre-sell efforts. All merchants would then have to clean up their diversion tactics, reporting tags and make better landing pages to recruit new affiliates and hold on to the old ones.

    1/55 means one sale per 55 clicks. 1/1648 means you have to generate 1648 clicks to get the first sale. Of course my feeling is that no merchant should be allowed to continue in a "pay-per-performance" program if they generate the above stat. They are a waste of web ink and web traffic. Dealing with "incent" traffic affiliates then comes into play, since without parasitic BHO traffic diversion plug-in, their own traffic is usually worthless.

    WebMaster Mike

  3. #3
    Full Member
    Join Date
    January 18th, 2005
    Posts
    290
    Hi TAG,

    That was what I also thought about a year ago when CJ first launched this 'Open Marketplace'. Those 'trick' banners (i.e. window alerts) always get a higher than normal CTR but low CR thus EPC is lower but you can earn more than normal banners because they are clicked more often.

    Those were the days when I was still running a content site. I used mainly banners [img]/infopop/emoticons/icon_eek.gif[/img] Thank God I found out about CJU during that time otherwise I'd be toast by now. [img]/infopop/emoticons/icon_smile.gif[/img]

    P.S. Don't take the EPC too seriously, it's just a guide that helps you sort the diamonds from the sand more easily [img]/infopop/emoticons/icon_razz.gif[/img]

  4. #4
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    1,087
    Positively brilliant Tag. I only wish I could've presented it the way you have when I have tried to explain to a few advertisers in the past. Maybe then they would not, or could not, have taken offense to my observations.

    Can I use your example in the future when I attempt to show merchants a few errors in their ways of thinking?

  5. #5
    Ad Network Rep ToddCrawford's Avatar
    Join Date
    January 18th, 2005
    Location
    Santa Barbara, CA
    Posts
    2,162
    CJ presents both EPC and EPM to publishers and advertisers in the reporting. Each publisher should look at all the data to make their decisions on what is most effective for them.

    The reason CJ does not show EPM instead of EPC is because some publishers (often the largest ones) do not generate impressions (they are serving their own ads) and therefore EPM would not be accurate for an overall program metric. Since there is no way to get around the click (every publisher generates clicks), CJ chose EPC as a more accurate metric for program performance.

    Todd Crawford
    Commission Junction

  6. #6
    2005 Linkshare Golden Link Award Winner  ecomcity's Avatar
    Join Date
    January 18th, 2005
    Location
    St Clair Shores MI.
    Posts
    17,328
    Todd were well aware that the Dupers and parasites do not generate impressions and in many cases not even a click to poach a sale. They just devised a way to swap affiliate ID#s or plant cookies while the shopper is going from point A to B.

    Please comment on my above simple "conversion ratio" stat that bypasses all EPC or EPM fallacies.

    WebMaster Mike

  7. #7
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    1,086
    One of the deals missed in the equation is the type of company and content of the site. The reason that inkject ink has such a high EPC is that people really don't go to an inkjet ink unless they are interested in inkjet ink. It is kind of like a product link verses branding banner sort of thing. 1-800-Contacts has a really high EPC, because it is a very targetted audience.

    Okay, banners for sites like hotjobs, eBay, etc., will get a lot of idle clickers who are just interested in the subject...not buying (I think this is part of Fredericks of Hollywood's problem, there are some people, who enjoy looking at underwear models. They don't know any women to buy underwear for...but they sure like the models.)

    The main thing the EPC does for me is it shows that the company is paying. Other than that, it is a game of matching your traffic to a strategy.

    Missoula - Short Stories

  8. #8
    2005 Linkshare Golden Link Award Winner  ecomcity's Avatar
    Join Date
    January 18th, 2005
    Location
    St Clair Shores MI.
    Posts
    17,328
    Yintercept makes a good point. Niche' merchants should have a higher EPC and conversion ratio than general giftshops or highly competitive categories like computers and home electronics.

    A site like www.suspenders.com doesn't get shoppers who are just looking over 50,000 items for gifting ideas. Scott could pull the CJ merchant records for the 2 years Holdup Suspenders was at CJ. They recorded and paid out 1 sale for every 15 clicks. No other product merchant in the history of affiliate marketing has come close to that figure. At that time CJ wasn't interested really in sales merchants as the kiddie click scripters were raping them royally. To make up for the extra expense of monitoring cheaters they raised all merchant fees. I pulled Suspenders.com out of CJ prior to EPC and meaningful reports, because they were buried underneith a pile of 1600 "not ready for primetime" merchants and couldn't get CJ to showcase their success.

    CJ finally listened to me and kicked all the CPC crowd out, but let the "lead" merchants remain. They did improve their recruiting functions and established EPC to leapfrog the other 2 sleeping giants. Now the script bandits come back wearing suits and push parasitic traffic hijacking schemes on them under the guise of reward/incent/Freebee lead merchants who also are fronts of the Opt-out and UCE spammers and Telmarketing groups.

    Scott could make the next bold move in affiliate marketing and make CJ a product only network. Think of the savings in compliance staff and customer support issues. Look at your merchant deadpool for the last 3 years and see which types of merchant went belly up the most. This sure won't sit well with many ABWers who a "lead" affiliates but those pseudo merchants will migrate to the 2nd level networks.

    I'll send you a bill if you think CJ needs any more management advice. I'm tired of working for nothing pulling your butts out of the fire. My CJ sales are off 50% across the board due to bad calls on your end of things and your merchants parasitic partnerships.

    WebMaster Mike

  9. #9
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    817
    EPC is the sum result of many factors. Aggregate EPC is a good guide, but not perfect. Your OWN EPC is more useful.

    Here is an example of how poor management can impact EPC. I recently assisted in auditing a program at CJ. Good merchant, good products, no tracking problems. EPC was horrible and not a true reflection of what they do.

    During the audit I found two affiliates sending "junk" traffic to the tune of thousands of clicks. None of this would ever convert and they were removed.

    In this case the manager was inexperienced and thought traffic = good. When in reality it would have been better to build a much stronger EPC and attract good affiliates with these numbers.

    Affiliates with strong EPC should also use that to their advantage at CJ. If you turn in high EPCs you will push up their EPC and in turn bolster their recruiting efforts. Use this to negotiate an additional point on the deal or a special promotion.

    They can see the numbers, you must let them see the potential advantages.

    best,
    Wayne

    Wayne Porter
    V.P. Product Development
    AffTrack LLC.
    http://www.afftrack.com
    http://www.revtrends.com
    Advanced & Automated Data Analysis for Performance Marketers.

  10. #10
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    2,419
    I use the overall ECP numbers on a merchant basis. Once I get in looking at links I pay very little attention to it.

    You can see two different text links one that converts big time and another that doesn't. I end up modifying my text usually anyway so the actual numbers dont mean squat to me at that level.

    I use the links I see as attractive and fit the overall space I'm filling in my page.

    The only concern I have is sometimes I wonder if there is a problem with the link itself and the tracking associated with it. But there is an easy way to correct that. If it doesn't work change it. If that doesn't work - yank um.

  11. #11
    ABW Ambassador BareNecessities's Avatar
    Join Date
    January 18th, 2005
    Posts
    519
    quote:
    Originally posted by yintercept:
    They don't know any women to buy underwear for...but they sure like the models.)



    Hehehe...

    Michael
    973-621-6211 x 5040
    michael(at)barenecessities.com

  12. #12
    ABW Ambassador cusimano's Avatar
    Join Date
    January 18th, 2005
    Location
    Toronto, Canada
    Posts
    1,369
    quote:
    Originally posted by Wayne Porter- AffTrack:
    EPC is the sum result of many factors. Aggregate EPC is a good guide, but not perfect. Your OWN EPC is more useful.



    I agree that one's own EPC is useful, however, it is also not perfect. For example, I have some merchants were I was lucky to make a few sales on a few clicks (and thus register a high EPC) however, my overall revenue was relatively low (because I only made a few sales). On the other hand, I have had some merchants where I have done well in terms of overall revenue, however, a lot of clicks were needed to generate those sales (and hence the EPC was very low).

    Theoretically, I should be concentrating more on the merchant where I have the higher EPC because in theory for the same number of clicks I should generate more revenue compared to my low EPC merchant. However, in reality this is not necessarily true because straight comparisons between merchants are difficult to make as there are many factors involved as to why people click and/or buy from one merchant compared to another.

    quote:
    Originally posted by Wayne Porter- AffTrack:
    In this case the manager was inexperienced and thought traffic = good. When in reality it would have been better to build a much stronger EPC and attract good affiliates with these numbers.


    If a merchant deletes all of their ineffective affiliates and as a result the merchant's EPC increases, how does that impact on the sales generated by any particular affiliate? I don't see how any change in the merchant's EPC (through the deletion of ineffective affiliates) would increase a particular affiliate's bottom line sales. It may be useful in attracting new affiliates and saving on traffic costs, but unless those new affiliates can produce sales, the EPC will just go down again.

    At the end of the day, I am much more interested in the merchants that can generate the most revenue for me, and not necessarily the ones that generate the highest EPC for me or the ones that show they have the highest aggregate EPC.

    Yours truly,
    Cusimano.Com Corporation

    per: David Cusimano

    affiliate scripts for amazon.com, allposters.com, gocollect.com, and more.

  13. #13
    ABW Ambassador
    Join Date
    January 18th, 2005
    Posts
    817
    David,

    Purging ineffective (not inactive) affiliates in an aggressive fashion usually ensures EPC will go up. A high EPC usually attracts other big hitters. This method works.


    I think dropping ineffective affiliates is a good idea. In all fairness they aren't getting a good return on their investment and should probably try another relationship anyway.

    best,
    Wayne

    Wayne Porter
    V.P. Product Development
    AffTrack LLC.
    http://www.afftrack.com
    http://www.revtrends.com
    Advanced & Automated Data Analysis for Performance Marketers.

  14. Newsletter Signup

+ Reply to Thread

Similar Threads

  1. Featured: Losing Tons of Money Because of Server Problems?
    By isellstuff in forum Analytics, Multi-variable Testing & Optimization
    Replies: 17
    Last Post: May 31st, 2014, 01:11 PM
  2. Conspiracy Theory, or are affiliates really losing money?
    By RickPlmr in forum Blocking Tips/Advice/scripts
    Replies: 25
    Last Post: September 19th, 2007, 11:32 AM
  3. Fix Reporting, I'm losing lots of money
    By kelly626 in forum Commission Junction - CJ
    Replies: 197
    Last Post: March 3rd, 2005, 03:48 PM
  4. Advice: You could be losing money using invalid links...
    By Ron Bechdolt in forum Commission Junction - CJ
    Replies: 16
    Last Post: February 20th, 2004, 12:37 AM
  5. CJ 6.0: Where are impressions, clicks, EPM, EPC, etc.?
    By Andy in forum Commission Junction - CJ
    Replies: 9
    Last Post: April 7th, 2003, 10:27 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •