Forum Moderators: skibum
Assume that a site had first class SEO, Design, Content/Editorial, and Marketing... simply as good as it gets.
Let's say the site was generating 1,000 unique users a day, and that it was an affiliate for any number of products (e.g. satellite dishes, pharmaceuticals, online dating, carpets, etc.).
With the understanding that different products can have widely different statitics, does the following seem 'in-line' with reality? (Oh, and one more thing, assume that the site was good, content driven, and captured focused users.)
1) Of the 1,000 daily users, approximately 30 (3%) would click on something that took them to the actual product/service/lead site for which we are an affiliate.
2) Of the 30 that went to that product/service/lead site, approximately 1.5 'took action' (filled out form, purchased, etc.)
3) Overall, this means that about 1.5 per 1,000 visitors (.15%) took action.
Does this seem reasonable as a general rule of thumb?
The reason I ask is because we are providing design and Seo for some affiliate sites, and they asked us to work with them to generate some ballpark figures.
Initially, I went about it by saying 'okay, what does it take at this traffic level to generate $100 per month, $300 per month, $500 per month & $1,000 per month?'
THEN, however, I noticed that when analyzing Commission Junction's 'EPC' -- which I naturally thought was 'Earnings Per Click', I learned that it was actually earnings per 100 clicks. As such, I took their statistics, including the 'action' amount, divided them and averaged them to come up with 5% average conversion (in other words, on average, 5% of the clicks from affiliate sites actually generate 'actions' for the CJ advertisers).
AS A RESULT, the numbers I had originally come up with, which seemed VERY conservative (e.g. 1 per every 1,000 visitors, 1 per every 700 visitors, etc. take action)... truly become much smaller -- I THINK -- maybe, in the grand scheme of things, it's all just a number, and using a number like 1 per 700 visitors to both CLICK to the AFFILIATE site and TAKE ACTION are reasonable... however, the way it looked to me was that I could divide those numbers above by .05, so those marvelous conservative figures become much more difficult. As such, 30 per 1000 click to the product/service/lead site, and then 5% of them actually 'take action,' again arriving at a final amount of 1.5 per 1,000 or .15%.
Now the 5% 'take action' rate is arguably conservative, at least according to this article [ecommerce.internet.com] -- though it may not be totally applicable...
So overall, does this seem appropriate? Please let me know your thoughts... I'm kind of torn here, because I feel confident that we can be successful with 1/2 of the above... (e.g. of every 1,300 people who come to the site, only 30 go to the product/service/lead site, and there, again, 5% or 1.5 'take action'...
Then, if anyone feels inclined to provide some general ballpark figures for numbers (e.g. experience with, say, 10% of site visitors actually clicking through to the product/service/lead page, or 5%, or 20%, or 50%) then I'd be greatly appreciative... as for the other number (5% who, after clicking through, actually 'take action'), I'm fairly confident in the validity of that particular average I came up with.
Thanks again!
[edited by: andrewrab at 11:09 pm (utc) on July 22, 2003]
Click thru from my sites to "affiliate merchant sites" runs between 5% and 15%.
The "action" conversion rate on the merchant sites, will vary between 1.5% and 10.8%, which is a pretty big range.
Factors to take into consideration when trying to arrive at "happy" statistics:
type of click - was it natural SERP, was it adwords / overture, was it pop under/ over?
That will impact conversion rates in a HUGE way.
What was the keyword? How specific? Degree of specificity, I've noticed, has a direct correlation with the conversion rate of the site - eg, if the keyword was "blue widgets in my area" and that's what the merchant sells, the affiliate is much more likely to get that desirable "action" to take place.
Just some guidelines - your numbers seem realistic, but keep in mind the above.
I actually saw that I had used some bad numbers... so the ones I just edited are 'actual' and are even more conservative than what I posted before... (every time I post some numbers-related question, I always screw up with the math!)
But I'm going to reread your response... because it was one of those things I just needed a little 'phewwwww' for... because I've put A LOT into this (we've got some fairly sophisticated and special things going on) and after recently losting my SHIRT on patented software we had endorsed by Microsoft (and for which I raised two rounds) I just became VERY WORRIED that my MONSTROUS hole after this recent debacle was about to get much worse!
I was thinking... oh no! I based assumptions that were 100x wrong!
Oh, by the way, these are all natural, in-context, appropriate and non-intrusive links, all of which are text (again, contextually placed)... no banners or popups or anything. And one more thing, they are all natural SERPs, and focused for the term, and not cloaked.
Thanks a lot... I'll sleep better tonight! :-)
When you say that you are "providing design and Seo for some affiliate sites" do you mean actual advertising affiliate sites (not the actual merchant themselves)? If so then this is a bit strange (unless its not only an affiliate site) as affiliates only earn a small commission - so would have little to pay for design/sep services?
0.3% click through seems a bit small - if these visitors are finding your site for, looking for a product, I think more users should be clicking through - after all your not asking them to buy it - only to see more info. How targeted are the users that come onto your site? Is it a kind of searh-engine/shopping type site?
Unless the traffic is free and low quality, I think that you'd need a much higher your-site to affiliate site, and a slightly higher conversion rate for users going to the merchant site (=outwith your control - just choose good merchants!)
[edited by: TravelSite at 11:07 pm (utc) on July 22, 2003]
Why you might ask? Because we aren't affiliates ourselves (don't really want to be, at least now) but we have made a lot of clients a LOT of money through major experience, expertise, sweat & tears, and we came up with a way to risk a little more but potentially gain a lot more...
I'm glad you chimed in letting me know these numbers are either realistic, or potentially low.
As for the users, they are extremely targetted... we provide (on a page-by-page basis) precisely what they were looking for... well, usually at least, we can't always get it right! One more thing... 30 per 1,000 is actually 3%, not .3%.... bad math again!
Sounds as though your doing something interesting!
...aah, just read the 3% thing..much better than 0.3%!
On an interesting note with regards to ppc... we've managed a couple campaigns and often find 1% to 4% clickthroughs for top-3 Google and Overture placement... man, that doesn't seem very good... either that, or people seem to greatly prefer SERPs 'au natural'
For example we get high listings in normal search results and do our best to make our title better than the adwords ones - as we DON'T want users clicking on them!
The folk in the adwords forum will tell you that its all about ROI, not click through - but from this post I see that your already fully aware of all this :)
CommissionJunctions earnings per click is a great feature that I wish all such networks had. It gives a good - but rough guide. As its an average you can often achieve much higher EPC than stated - if you send targeted traffic.
And you must assume, that you can make no assumptions. ;-)
Your numbers do seem conservative, which is what you want. Affiliate marketing is largely a gamble when trying new ideas, niches, and methods. Don't bank on success.
But with very conservative numbers, and looking at a worst case scenario, you may be able to judge what to expect minimum, in general. I still wouldn't put much on the line though.
