Another WTF am I doing wrong thread (w/ cliffnotes)

[TripleT]

First campaign, wanted to get started ASAP and not wait for domains for LPs to come through, so I decided to do direct linking.

Cliff Notes: get very little traffic even though I'm prepared to spend money. I thought it was spend $100 - wait an hour or two - see what ended up happening. Instead it's been a day and a half and I managed to burn through all of $15 worth of shitty 2nd tier traffic. Halp.

Journal so far:

Yesterday - Spent most of the day reading and setting up PPC accounts. I ended up launching one campaign, at MSN and Miva.

What I did was deposit $100 at miva, and commit to 100/day-100/mo-deplete budget ASAP at MSN. Let's say my offer pays $30. I just thought of a short list of keywords myself, like 15, and submitted them to both. Offered $0.35 at Miva, $1.00 at MSN. I created only one ad.

Results after 18 hrs or so: Like 30 clicks from Miva, no conversions. Prosper also shows a bunch of clicks from MSN, but they're from the MS site. MSN shows 200 impressions or so, no clicks.

So today I decided I needed more volume, and went to the Adwords keyword tool. Typed in a few main keywords, and got a list of 200 suggestions. I deleted a bunch that seemed bad straight off the bat, ended up with a 150.

Copied this list to Excel, added all the 'avg monthly searces' numbers, ended up with 2 million or so. I split this into percentiles, and divided them among 6 different ad groups on msn, roughly corresponding to the percentile chart, with the long tail keywords in smaller groups %-wise. I created 4 more ads, 5 total, and used those for all the adgroups.

It's been on for about two hours now, and NO IMPRESSIONS. Could this be because I didn't edit a target URL into each individual keyword? The reason I did this was because I had already set the target URLs for the adgroups. I did get a few 'fake' clicks again, from the MS server.

I want more actual traffic - I know, test test test, but how the f can I test if nothing bloody happens? I looked at some of the Miva referring sites, and they're utter crap. So I want traffic from MSN. Is it because..

- My bid is too low? But I reckon I can't afford much more if I'm already requiring 3% conversion just to break even..
- The niche is too small? I have literally no way of accurately judging that, other than the 'two million/mo' figure from Adwords keyword tool. Top one is 500k or so btw. Got the idea from my aff manager when I asked for 'good place to start for a noob'
- I should have launched like 50 campaigns for different products instead? Would like to, but I'd rather not put in the 5 grand until I feel I'm better at this...
- I didn't create landing pages and used Adwords like everyone else? It's not really laziness that prevented me, more 'ok either you have a LP up and running in two days or you start this right now and learn something'. But now that this is going so shitty, I'm hesitant to work on a LP if I already know my ads don't get any fn clicks.
- I otherwise fucked up?

I was expecting it to be like spend a hundo or so on clicks, wait an hour or two, see if I made a profit. Instead it went more like spend $200, wait a whole damn day, and see that I didn't manage to spend enough to even gauge how well I'm doing.

I'm willing to put a lot of work into this, but at this point I don't see what else I can do and it feels like I'm just wasting time. I mean it prolly took me longer than it should to get everything set up, but at least I learn from that and will be able to do it faster in the future.

Seriously this is more frustrating than if I had just lost the money...

 

[papajohn56]

miva and msn are low traffic, plus with that many keywords it depends on the amount of ads you run. if you're running 150 keywords on 4 ads that's a problem

 

[dewme5]

I've heard of people saying bad about Miva. I'd suggest searching that one.

If, Miva isn't good, you need to throw it out of your calculations.

If you require 3% conversion to break even, be prepared to pay for the education.

5 grand is a decent budget, don't blow it. Your knowledge will grow leaps and bounds, and in 2 weeks, you'll be much better then right now. It'd be nice to be better, and still have some money to experiment with.

Niche too small???? is that possible?

Landing page. You don't need one, but from your current results, It can only help. Plus, the practice will benefit you more then checking your stats every 2 minutes.

 

[TripleT]

miva and msn are low traffic, plus with that many keywords it depends on the amount of ads you run. if you're running 150 keywords on 4 ads that's a problem

could you clarify this a bit? How is this a problem?

Should I..
- create ads specifically for each set of 10-20 or so keywords? So 6 groups, times 5 ads each, where any ad only belongs to one group?
- have like 50 ads that are each assigned to all the keywords?

My rationale was few clicks -> few ads so I have a more accurate picture of which does well, and i don't know shit about ad writing -> ads are for all keywords, cause I personally can't tell in advance which ad should go with which kw

 

[TripleT]

I've heard of people saying bad about Miva. I'd suggest searching that one.

If, Miva isn't good, you need to throw it out of your calculations.

If you require 3% conversion to break even, be prepared to pay for the education.

so does this mean my bids are too high? (can't count on that high a conversion) I mean I can lower the bids but then I'll REALLY get no traffic..

5 grand is a decent budget, don't blow it. Your knowledge will grow leaps and bounds, and in 2 weeks, you'll be much better then right now. It'd be nice to be better, and still have some money to experiment with.

Niche too small???? is that possible?

umm.. I guess? I mean if $100 a month in ad spending is all it will sustain, and that generates $150 in revenue, that's still a fine investment obv.. but it does mean I can't really test and improve my campaign/skills in the short term. Are you saying no niche is that small, and it should easily be possible to test in a day or so, as I had originally planned to do?

Landing page. You don't need one, but from your current results, It can only help. Plus, the practice will benefit you more then checking your stats every 2 minutes.

will do.. at least i'll be able to run adwords that way. is there any way to gauge in advance how much it will cost me to get x traffic/day, and how much traffic there is to be had - BEFORE i have gone through the process of creating a LP and submitting that? and even then, provided it has T&C, about, copyright, etc.. is the 'traffic estimate' from the start the one it's gonna stay at - provided the CTR isn't horrible - or am I gonna still have to throw a bunch of money at it before I can get cheap clicks?

 

[TripleT]

btw, i realize all i'm doing here is ask questions, so thanks for the help. hopefully one day i'll be a big shot and know enough to be able to return the favor

 

[papajohn56]

could you clarify this a bit? How is this a problem?

Should I..
- create ads specifically for each set of 10-20 or so keywords? So 6 groups, times 5 ads each, where any ad only belongs to one group?
- have like 50 ads that are each assigned to all the keywords?

My rationale was few clicks -> few ads so I have a more accurate picture of which does well, and i don't know shit about ad writing -> ads are for all keywords, cause I personally can't tell in advance which ad should go with which kw

I've had a 270% ROI in 4 days with like, 8 keywords and 5 ads at first, and branching out from there after I get more data

 

[mcuk2000]

1- Why you use google keywords tool on adcenter campaigns? Use the adcenter tool.
2- on MIVA bid at different level for the same keywords, let's say you have the keyword free ringtone: bid at 0.05-0.20-0.40 . You will end up getting more volumes. ANd track the IDs!

 

[TripleT]

Update: I figured out one thing i was doing wrong.. I updated the geolocation database from tracker202 (even though it's not 100%, according to their forums it's broken) and it seems a large part of my miva clicks are from outside the us, while my offer is us only. That would explain the non-existent conversion rates.. couldn't figure out how to geo-target traffic, wrote my rep about it.

I used the google keywords tool because I found it easier to get a traffic estimate for individual keywords - figured the volume should be smaller, but proportional in terms of keywords, on msn. But sure, I'll use adcenter next time

So I read some article linked to at jonathan volk's blog, and learned that I probably discounted the importance of having the ads tailored to individual keywords. Feel stupid for not realizing this - even tho I never ever click on ads, when I look at serps I'm more likely to click the one that has most of my query in the title, no reason to assume it wouldn't work the same for ads.

Anyway I'll try more keyword-specific ads tomorrow, with a little less keywords. Then if I get zilch traffic again I'll add a bunch more and go from there. Will keep you guys updated cause I'm sure I'm gonna run into some problem again...

This actually illustrates an interesting point - just from reading blogs and articles, all this stuff seems like it's common sense, but when you actually start to run your own campaigns all that 'common sense' crap adds up and you run into a lot of problems you weren't expecting at all. So to who(m?)ever out there doing jack shit just 'studying', don't fool yourself you think you know but you don't. I mean based on my Internet knowledge and bluehatseo-reading I felt pretty confident too, but now I see there's sooo many little things you just need to actually do, and then when you fk up at least you know what questions to ask.

 

[DreamMachine]

No idea what the niche is, but if it's a $30 payout it's probably a weightloss or bizop rebill? If so, $0.35 and $1 are too low unless you have 4-5 word longtails in your list.

Also, always create 2 ads. Doesn't take much more time and the info can be invaluable.

 

[TripleT]

Yeah something like that..

So can I conclude that, if $1 is too low, generic keywords just aren't going to be profitable and the only way I can make it work is to have longtails, but an absolute shitload? Or be happy with the few clicks I do get and hope I'll manage to make ten bucks a month? Or is my 3% conversion guesstimate too pessimistic?

 

[dewme5]

With keywords, you can go wide, or you can go deep. With specific long tail keywords, your rates are going to be cheaper. You will also need to many more to get the same traffic. After all, how many people search for "pink leather and diamond iphone case" when compared against "iphone"?

iPhone is going to cost $6 while the long tail will cost $0.10. If you pick a niche you know, the idea is that you can come up with keywords that convert better. While you get less hits per keyword, you get a higher conversion and cheaper bid due to less competition. Instead of "alcohol", you could choose "Bacardi 151", or "Grey Goose Vodka".

test, test, test.

 

[dewme5]

On the niche to small topic. You aren't going to have one niche, you are going to have alot. It's probably a little late to own the "dating" niche. Maybe trying the "midgets dating millionaires" niche

 

[TripleT]

On the niche to small topic. You aren't going to have one niche, you are going to have alot. It's probably a little late to own the "dating" niche. Maybe trying the "midgets dating millionaires" niche

Just out of curiosity, how does this owning a niche work with PPC? I can see how with SEO it's gonna be near impossible to beat the amount of links/PR/regular visitor base/whatever else Google uses that the top contenders have.

But with PPC, SOMEBODY is paying the big bucks for the huge traffic keywords. Hypothetically, if I had the bankroll, what would prevent me from going to their site, copy their campaign as best I can, and doing the exact same thing? Is it their long standing history of well-performing ads that allows them to pay much less than anyone else could hope for?

 

[r1se.x]

could you clarify this a bit? How is this a problem?

Should I..
- create ads specifically for each set of 10-20 or so keywords? So 6 groups, times 5 ads each, where any ad only belongs to one group?
- have like 50 ads that are each assigned to all the keywords?

My rationale was few clicks -> few ads so I have a more accurate picture of which does well, and i don't know shit about ad writing -> ads are for all keywords, cause I personally can't tell in advance which ad should go with which kw

if you´re doing adwords, the less keywords you have per adgroup the better, because you can better control them and also your quality score will rise because the keywords are more relevant to the ads. You should also make at least 2 ads per adgroup and a/b split test them.