ComradeWolf's PPC Journal

Dude, 7Search blows bad. It's all retarded people who can't figure out when they've got caught in a loop of redirected parked domains, and bot traffic. It will convert, but very shittily. It's a waste of your efforts. It's pennies per click for a reason.
 


Yeah, to be honest, those cheap clicks seemed a bit suspicious.
Would you recommend PoF then?
Thanks a lot!
 
New test for the supplement - no sales (was testing new page layout). Now I am testing images for the previous LP. I hope that they will increase conversions.

New test for CB fitness product - I am now testing KWs positions, because I've read mixed opinions on this (some say that lower positions improve conversions, some say they don't) and now I want to test it.

I also plan on testing some more offers (not only in fitness/health niche - something that I've done before - but in other niches as well) since I have probably probably sticking to this one for too long.
 
Looks like the most the difficult thing in PPC apart from constant outflow of cash is the semi-depression/sadness (which sometimes overwhelms you) because you don't know when you will succeed.

I guess this is a bit similar to what travellers like Columbus or pioneers felt (even though it is obvious that what I am doing is definitely not as crucial and beneficial to society): you have no idea when you will see the land, you are not sure if you will actually see it at all, your position in society is getting shaky (sailors are about to fuck you up big time = you are losing money).

But I digress. No sales in the supplements test. Testing another image now. Hope it starts converting.

Still testing positions-conversions correlation on Bing for the CB product. No sales so far. Looks like higher position lead to more sales (which means that the hypothesis, that people who are more interested in the product, are ready to buy it and actually get to the4th-5th positions, is wrong).
 
@Minderwinter Thanks for that

However, the spreadsheet had some errors (formulas took the wrong cells)

Here is the updated version:
https://docs.google.com/spreadsheet/ccc?key=0AlsBupJOGhOAdGttSHUzWmlBMVBFRElnRU9NLUhzRkE&usp=sharing

I added a revenue field per conversion, so you can now also see the profit margins.

This is also nice to figure out where the limits are.

In the spreadsheet, you can see that 11 conversions will make you money even on the lower bound, 10 will not.

You can also play around with the costs

::emp::
 
Thanks for taking a look, emp, but I'm not sure what you're talking about. Can you be a bit more specific?

The one error I caught appears in your spreadsheet, too. The upper bound on the conversion rate should output 1 (not zero) if the upper bound is greater than 1.
 
Oh, actually I do see one way the spreadsheets are different. You have the "upper bound" on the cost per conversion as the smaller number, which doesn't make sense. The lower bound on the conversion rate gives you the upper bound on the cost per conversion, and the upper bound on the conversion rate gives you the lower bound on the cost per conversion.

But if you've caught any other errors in the sheet do let me know.
 
Ah... I might have misunderstood you, then...
For me, upper / lower bounds are tied to the confidence interval regarding conversions.
(just makes sense for me, I think)

So in that view,
Upper bound means more conversions, so cost per conversion is lower.

::emp::
 
Hey Comrade,

I scanned this journal... I really respect the hard work and props to keeping a journal!

I manage PPC profitably for many small businesses, including my e-commerce startup. One client spends over $100,000/m. I started PPC back in October 2012 and have been grinding my way up ever since. The first two months were rough, because my fundamentals needed improvement... so just hang in there!

Based on your posts, I think you have a pretty big flaw in your PPC fundamentals right now when it comes to keywords.

You're focusing on your landing page a lot, which is truly great, but it seems that you treat keywords as an afterthought.

It doesn't matter if you have the best landing page in the world, if you are targeting the wrong keywords, your campaign won't be profitable. Additionally, it's helpful to build landing pages based around keywords and what the searcher is looking for.

A lot of the advice you are reading is in regards to display advertising (POF/Facebook) and you are doing search advertising.

Think of keywords much like demo targeting on display advertising. Getting a campaign profitable on display advertising is all about tweaking and advertising the demo until you find the perfect demo. For Example, the perfect demo for Supplement A and Landing Page B may be Males 32-36 who are College Educated (indicates higher disposable income) and single (they may be interested in losing weight to attract a mate).

A similar concept applies to search advertising. Let's take two different search terms:

"Information about Diet Supplements"
"Best Diet Supplement to Lose Weight"

Keyword A is trash, Keyword B is much better.

The searcher in keyword a is a browser, he's just looking for information right now about diet supplements. He may convert, but it's going to be much tougher.

The searcher in keyword B has already been browsing, and now he's looking for the "best" supplement. This signals he is much farther in the sales process and is much more likely to convert.

Regardless, you don't have to be great at picking keywords right off the bat, it will take time and experience in the niche. What you do have to be great at is:

A. Tracking Keywords Properly.
B. Bidding on only a small set of keywords, so you can split test quicker.

So you've had a couple of sales already. What keywords did they come from? If you can't tell me this, then you're not tracking keywords properly.

Secondly, how many keywords are you bidding on? If you are bidding on 20 different keywords, then it will take you forever to figure out which keywords convert well and which keywords don't. Take a serious look over your keywords and pick 3-5. This will help you determine the winner and loser keywords much quicker.

Also, make sure you are properly using keyword matching. If you are using naked broad matching, then you're going to get hammered and not make money. Use Exact Match, Phrase Match, or Broad Match Modifiers, but never naked broad matching (especially if you're new to search advertising).

Lastly, You shouldn't have your 3-5 keywords all in the same ad group, even the same campaign. Each of these keywords should have a separate campaign. This is for many reasons, but most importantly so that you can create different ads for each keyword.

You don't want to run the same ad for keyword "How to Lose Weight" as you will for "Best Supplements to Lose Weight". This should be obvious to you, because Quality Score is heavily determined by AD CTR and AD CTR is heavily determined by ad relevance to the keyword.

If you're bidding on "How to Lose Weight", then your ad copy should mention the phrase "How to Lose Weight" AT least once, and it should probably be the Ad Title.

Long Rant, but hey, I think it will help you :).

TLDR: Proper Keyword Research, Campaign/Ad Setup, & Tracking is just as important, if not more, than Landing Page Optimization.
 
Hey K-Noop,

Huge thanks for this extremely valuable post!
You are right about KWs: I stopped thinking about them altogether and got obsessed with LPs. Will definitely change my approach and focus.

Thanks a lot!
 
As per the advice of K-Noop I have created separate campaigns for each KW and most of the QS are really high. Only 1 campaign has a QS of 4, but I am working on the LP now.

Now I am trying to get prosper202 to track my weight loss offer (for CB offer I use CB tid's).

I will wait for more clicks for each KW to figure out which ones convert and which ones don't and remove/keep them and test a new batch of KWs.
 
Still waiting for more clicks, though some of the KWs already have around 50-60 clicks. CTRs of ads have increased to around 3-4%.
Made 1 sale from 1 keyword.

I am still tuning prosper202: it didn't track Bing clicks so I tried changing settings and then came across a short line in FAQ where it was mentioned that an LP should be in .php format. I've changed some of the html LPs to php and I am now waiting for results. If it still isn't working I'll try doing the same with advanced LP tracking. If still doesn't work I'll buy a VPS since it looks like some people have problems with setting up prosper202 on HostGator shared hosting.
 
After days of futile attempts to get prosper to work on shared HG hosting I bought a VPS, transferred my site there and asked the tech support to install prosper.
The tracking seems to be working, at least when I check it myself. Now I am waiting for for my ads to start getting clicks on Bing so that I could be absolutely sure that it works.
By the way, now the pages load very fast, I would even say extremely fast. Feels good!

Today I've realized that this month I've saved around $200 dollars (more monies for campaigns): I eat porridge with milk or 2-3 boiled eggs with toast for breakfast and a cup of prunes/dried apricots/cashews for lunch. For dinner I fry fresh broccoli or make tomato-onion-cucumber salad or sometimes fry chicken thighs/drumsticks with brussels cabbage or make chicken/beef soup and save around $6-7 dollars daily. This is more or less healthy and fairly delicious.

I thought I'd write this to somehow help newbies who need more money for campaigns.
 
I have changed my LPs for fitness offer because of the exit pop-up on the offer page which could result in campaigns/account ban.
Now I am also working on changing the LP for the weight loss offer (again because of the exit popup). This won't take much time and the LP will be identical to the original one (except that there would be no exit pop-up) so I think that network AMs will accept it.

I have also noticed that Bing interface was changed a bit - now I don't have to go to campaign settings to choose between the accelerated/even ad displays. Much better that it used to be.

Looks like the conversion rate for my fitness KWs is pretty low. Besides KWs don't have enough volume to get 100 clicks within 4-5 days.

At least the weight loss offer gets clicks faster (when my ads aren't suspended lol). It is very likely that I will move on to the weight loss offer completely because I am tired of waiting for at least 6 days to get enough data.
 
Thank for subbing, IceToEskimos! I'll try to keep it updated.

I have made a new LP for the weight loss offer - I have never spent so much time on a landing page before. This time I approached the task extremely dilligently: I read all the Bing guidelines and made sure that nothing on my LP or offer page was violating them.

Then I did my best to get the highest QS possible - this is one of the longest LPs I've created. I optimized tags so that they would have the KWs I am targeting (this time I have a lot of images on the LP so it took quite a lot of time).
Right now I am targeting 3 KWs with quite high volume and unless my ads get suspended I will get the data within 3-4 days.

My AM recommended that I try running this offer on PoF saying that some affiliates get decent conversions there. So now I am working on several creatives which I will start testing tomorrow.

By the way, the CB fitness offer didn't get any conversions (I've already finished testing 3 KWs with the highest volume), so I guess I will stop running it and will try something else.

I have also noticed that if I stop my campaigns, their QS usually drops within a couple of days by 1-2 points, whereas if I just let them run with a smaller budget the QS doesn't decrease.
 
I am now testing 20 creatives for POF with copy #1, then I will test the same 20 creatives with copy#2. Next, I will choose the better performing copy and top 10 performing images that were combined with this copy and will test them against 2 new copies (using these top 10 images) and will choose the winning copy.

Campaigns are quite targeted: I am targeting sex, age, body type, browsers, country, login count. I don't specify session depth, instead I just change bids.
I am not sure if I will get enough volume but if I don't I will just make them more broad.

My cut-off rate for ads is 0.1% CTR after 5,000 impressions.
 
4 ads out of 20 have CTR that is at least 0.1%, 3 have CTR of around 0.07-0.08% after around 2K impressions.

I am now testing the offer across networks because whenever I ask guys at peerfly why 15-20% of clicks (not only in my PoF campaigns, but in Bing, too) are not registered in their tracking system even though the traffic is targeted and is from the US (and the offer is available only to the US visitors), they just scratch their heads or give some incoherent response.

But I guess it's natural, because I've read about this on some blogs before, when affiliates said that CRs might be completely different across networks and that sometimes clicks "disappear".

I also think that my ads are way too targeted so I will make it a bit more broad to get more impressions after I finish this test and tweaking my ads.
 
I am now promoting an offer in the cooking niche (on BingAds). I am targeting three KWs, and this time I indicate the price in my ads, trying to target customers that are in the buying mode. I guess my CTR will suffer, but probably my CR will increase.

I am also direct linking, assuming that based on the KWs I've chosen, customers don't need more information on the product. Right now I am working on a review page that would direct to the offer page to split test it against direct linking.

I had to temporarily stop my ads on PoF because the tracking was off and I am working on it now.
 
You must log in or register to reply here.
Top Bottom
Back