Home > Technical & Creative Skills > Tracking Campaigns

---

What am I doing wrong?? (20)

---

Ok, so I joined this forum a couple of weeks ago.

I figured out my priority should be setting up Prosper
for tracking.

I got so many internal server errors that I can't even tell where they are coming from anymore.
Tried to troubleshoot, wasted a lot of time, still did not work.

Then I figured out I should go with the Mobaff tracker, since it is going to be a while
until I will have 1 Million of clicks in there.

Wrong, now I am having problems setting up the affiliate networks in there.

I was trying to set up neverblue and mobpartner. I just can't figure it out.
And yes, I searched up the forum for all of the above.

I have one campaign running on buzzcity which with a little bit of targeting
I could make it a winner, but I don't have proper tracking.

I am terribly frustrated.

Can anyone help me? Maybe on skype or through PM?
I feel like I will never get up and running at this pace...

---

Not sure which hosting are you using but you can go with BeyondHosting and they will setup prosper for you. The BeyondHosting guys are in here, search for them.

As for the camp I would stop it (unless profitable) until you get your tracking sorted out.

---

Yes, definitely, if you're having trouble setting up tracking get Beyond or another company to set it up for you. Then you can follow the STM guides to getting stuff set up in Prosper - see my signature.

Here's the Beyond Hosting special offer: http://stmforum.com/forum/showthread...hlight=hosting

That'll get you up and running!

---

If you have tech questions, there are people here in the forum that could try helping you out. Everyone has different levels of expertise from coding to network administration.

The tech stuff is probably one of the more frustrating activities of affiliate marketing so post your questions!

---

Ok, I will post my questions.

I am not sending tons of traffic yet, so I think my shared hosting should be able to handle the traffic, right?

Second, everything is fine - logged in, create campaigns, everything is smooth until I get to getting the tracking link.
Sometimes the tracking link works. Other time it does not. tried with both - using direct linking and landing page...

It is intermittent, so then I do not want to send traffic to a link that works & stops. It would be a waste of money.
That's when my Prosper window gets filled up with server errors for every space on the page.

Finally, I believe that the PHP version on the server (I had so many headaches since I asked them to upgrade my PHP version and they moved me to a different server) and the memcache extension not being enabled.
I spoke to them a couple of days ago, and the guy kept asking me why I need memcache installed.
They still did not enable it...

Could memcache be the problem with the intermittent redirects?
My prosper shows also that memcache is not installed and also that the geoip dat file is not there.
About the geo database file, I am NOT getting the location error for the file.
It just does not show up when I look at the prosper settings.

Now which one is the real problem?

PHP Version: 5.3.27
MySQL Version: 5.5.32-cll-lve
PHP Safe Mode [?] Off
Memcache Installed [?] No
Memcache Running [?] No
Geo-Location DB Installed [?] No
> this is the geo db that prosper comes with for the prosper paid version, right? Or it is the one that I paid $25
to download and install? LOL Totally confused now...

---

I used Digital Ocean as my VPS ($5/month) and then set up the server using tutorials and youtube videos (about 1-2 hours). I had some trouble setting up the Prosper, but I found someone on Fiverr.com to set-up my tracking. He can also set-up your Digital Ocean VPS for you for $5 if you like, but it's not that hard to do when you have the instructions/tutorials to follow. I have my VPS and tracking set-up, now I just need to figure out how to get my campaigns up on Facebook without getting banned.

---

I had beyond hosting set my tracking up for me back in the day before I knew how to do it myself and they handled all my tech issues and were vary patient with my clueless self. It's well worth the money to have reliable tracking and not have to deal with any tech headaches when you're first getting started. In the day(s) that it's taking you to deal with this, you could be well on your way to earning back the monthly cost and more of a service like beyond or someone else who offers comparable services. Time is the most valuable commodity here, don't spend days tinkering with some tech issues that could be resolved for $40-60 / month imo.

---

What hosting provider are you using? Shared hosting is going to give you issues. VPS is the way to go if you're starting out.

Also, what did the hosting provider say about the internal server errors? There is a server log that they should have access to

---

I am on Namecheap - had them for a couple of years now. But I will definitely look into the BH VPS again ( had an account with them when I really did not need it and I cancelled their VPS service).

Meanwhile I started already a mobile campaign on buzzcity, drilled it down after a 40% loss over a few days, now it is very targeted, but it does not get enough clicks; I funded a few more traffic sources. Should I start this campaign on those with the targeting I have (of course, after I get done with my forking tracking)?

I suck at POF as I started a campaign there too - just to try the interface, the uploader, etc. Lost about 10 bucks, but that's not the issue. The issue is that I am used to PPC and not CPM; there it goes I was paying $1/click, with a CTR on the creative of 0.033%.

Anyway, thank you for your advice (everyone) and I believe I will get another hosting account with beyond hosting and have this done...

On the other hand, is Mobaff tracker a working tracking solution? After all, I could just use the custom offers option on it to track...I mean, is any of you guys using it?

---

Mobaff tracker has a pretty good reputation, although I haven't used it myself.

May be more complex to get started with than Prosper, though, and there's less documentation out there.

From what you describe, I'd expect that the problems you were seeing were related to your hosting. As Jasper mentions, Prosper 202 really doesn't like shared hosting - attempting to run Prosper on shared hosting is pretty much always throwing money away.

Get good tracking set up, then start running your campaigns: definitely the best way to go.

---

Thanks George for the contact.
The guy installed STM for me, and configured the VPS server for another $5 bucks - which geez, saved me about 2 days of reading and trying to figure it out by myself.
Now off to setting up STM and loving it. LOL

Now I will get on another thread and bother you with my tracking questions.

---

That's the lesson here, outsource what you can't do yourself instead of wasting your time to figure it out.

You need to focus solely on coming up with good angles for campaigns and testing as much as you can.

That's your number 1 priority. It's more important even than tracking ( ask Jordan lol ).

Now keep us updated with your progress.

---

Thank you for all your awesome answers.
Now going down to business to get actually started...

Does this syntax look good for POF tracking with STM tracker?

http://MYDOMAIN.net/tracking202/redirect/tracker.php?t202id=822&c1={age:}&c2={gender:}&c3={ state:}&t202kw=creativenumber

...and for mobile tracking with STM tracker , (that would that be Mobpartner variables):

http://MYDOMAIN.net/tracking202/redi...t202id=243&c1=[=brand=]&c2=[=model=]&c3=[=os=]&c4=[=operator=]&t202kw=creativeorsomething

Is this good? I know you can do this with your eyes closed...

---

Your POF tracking URL looks good to go, except for the space before state:, which I assume is a typo!

---

All right, so I got that down (the POF tracking - yes, it was a typo).
I still have a question about the ads on pof:
Is it correct the following usage of the tokens?

"I am looking to meet single men {age:default} y.o in {state:default}..."

I am just not sure...because I have many ads using the tokens and they do NOT get
any clicks. I thought it would work the other way, these ads getting MORE clicks!

Now, here is what I did yesterday.
I started working on a dating offer, started campaigns,
direct linking, with this targeting:

Campaign #1
Country =United States
Age Between 30, 35
Gender =Male
Has children =Yes
Marital Status = Single, Divorced, Separated
Browser/Mobile ≠ Android Tablet, iPad, Android Phone, iPhone
Frequency Cap: 4
* No logins targeting

Campaign #1
Country =United States
Age Between 24, 29
Gender =Male
Has children =Yes
Marital Status = Single, Divorced, Separated
Browser/Mobile ≠ Android Tablet, iPad, Android Phone, iPhone
Frequency Cap: 4
* No logins targeting

I loaded up the MrGreen's uploader and it did about 122 ads (combinations) for each of the 2 campaigns.
I ran about 150-200 impressions for each one of them (each ad).
Campaign #1: 13 ads with CTR from 0.80% to 1.94%. out of 122
Campaign #2 : 10 ads with CTR ranging from 0.54% to 0.67% out of 122

Meanwhile I got 1 conversion, but that's not my focus right now.

Now, my questions are:

- Are 200 impressions enough to make a decision before stopping the ads without clicks?

- As soon as I started the campaigns, the mobile users started clicking. Is it okay that I took them out or I should keep the
mobile devices? What is your experience with this? My offer description does not say anything about mobile traffic...
Maybe I should ask my aff network AM??

- In my offer description it says that I should send back "subid groupings" info. What a heck is that?
"Please DO NOT pass back a unique subid for every click/lead, you will put a heavy strain on the advertisers tracking/server and will be asked to pause and or risk non payment. Please use SubID GROUPINGS only. "

- Should I make a clone campaign only with the few ads that show good ctr and take it from there while I lower the bids for the remaining
non-performing ads left in the original campaigns (#1 and #2) to give them some more clicks?

- Targeting - is it too tight? At a 0.53 CPM I barely get any traffic...I have a 15 bucks/day budget for each campaign. Could that be the problem?

- If I want this to go faster, I could use the "Distribute ASAP" option in POF, but is that going to give a fair chance to these campaigns to testing?
How many guys between 30 and 35 are right now on dating sites in USA? People are working...or...it does not matter?

What I am thinking to do next:
- 2 Landers for the ads that perform well and do a split test.
- Based on results, I will target a smaller age range ( I think it is too large right now - or isn't it?)

---

I think you know what errors you are getting. I had the same problems (loads of errors on the screen right after installation).

Ok, here's what you do (I think this will be the same for you):

Browse to 202-cronjobs/index.php and add the following line of code:

date_default_timezone_set('UTC');

Restart the Apache server and you should be good to go.

NB - if that doesn't work then look at the error message, browse to that .php file and add the above code. This is a known issue with P202.

---

Haha, cyber, you're about 2 days late with your answer. I already got it done with a VPS, now started advertising already. But I do appreciate your advice. Maybe it will be useful to someone else having the same problem.

---

Hey, nice work on getting started so quickly!

First thing I'd say is that you're probably testing waaaay too many ads. I'd never test over 100 in a single batch - more like 10!

I'm guessing you're testing a combination of images and text? If so, pick a set of text, and pause all ads with anything other than that text. Test your images - then, when you have decently-performing images, move on to testing all your text options with them.

Defintiely not. Read my summary post on ad cutting to get an idea of when to cut. You need 1,000 impressions as an absolute minimum - at 200 impressions, your cuts will essentially be random.

Different people have different opinions on this one. Personally, I'd say keep the mobile users in, but track whether they're converting. If they don't convert as well as Web users after a decent sample size (a few hundred clicks) exclude them then.

Hmm, you might want to ask your affiliate manager for more specificity on this. I suspect it's just a Cake-standard "don't use a unique subid in s1" thing (see the tracking guide in my signature) but it's best to be sure.

Nope, take your hands off the wheel and let it run for a while! It's too soon to start making changes - I know it's alarming to watch the money ticking away, but you've gotta get decent data!

Unless it's vital to your angle, I'd remove the marital targeting. I'm assuming the "kids" targeting is important? Other than that, looks good - although not targeting logins will mean that the campaign burns out faster. Gets you plenty of data in the beginning, though.

Always use "Distribute ASAP", in my opinion. There's no real reason to use even distribution. Just grab the volume.

Absolutely tons of them. USA/male is POF's biggest demographic.

Definitely do this. Split-test offers too.

A 5-year range is pretty solid, I'd say - it's about what I use.

Hope that lot helps! Looking forward to hearing more about your progress.

---

Love it, Caurmen. Thank you, you're a gold ball.

So yes, after I posted I did some research and I figured out I was running too many ads, and I did what you actually said I should have done.
I had about 20 ads running in each campaign yesterday, over the whole day.

The 30-35 yo tanked to about 0.10-0.15% CTR on those 20 - with 5 ads floating in this range after about 2500 clicks each. Testing another set of 10 ads today.

The 24-29 yo sits with 15 ads out of 20 at a 0.15-0.25 % CTR, but they got only about 1000 clicks each. They click, but no conversions. Testing another set of 10 ads today.

Is it only me or it is that this range rarely converts on dating offers?
I have been reading that 30 and up convert better than under 30 men.
I am thinking - if I don't want to throw in too much cash - I should maybe go for the 30-35-40 age range, and
do other targeting on THEM rather than wasting money on the younger guys (sorry younger guys reading thisLOL) . Just thinking loud here...

I have modified the targeting as you suggested, allowing mobile traffic and not targeting by marital status.

After watching these campaigns yesterday, and reading , and Reading, and READING around, I am pretty sure that
it can be considered a broad campaign...

One mistake that probably I am making is that, even though I call out the age and the location in my ads, I should go even more targeted than that.

As in "Bold guy needed for hairy woman that loves shaved heads in {state:default}!"...or something like that? It sounds gross, I know...LOL

So far I am down about $50.00. With 1 conversion of a few bucks! Yohooo! LOL

In fact I need to amek a PLAN ( the whole image) of how I want to test, run and scale these campaigns.
I like setting up goals and so far, it looks like I am getting stuck all the time with little things...

Too much talk and nothing to show for it.
I don't make too much sense, I know, but I bet a lot of newbies to this do not make sense when they are just starting out, right?
I will update this after I set up my "really" targeted campaigns for the 30+ age range and after I make a few landers. And then I will ask again...

---

How's this all coming on?

---

Home > Technical & Creative Skills > Tracking Campaigns