Well I seemed to have glanced over my initial start to this challenge so here goes:
Day T-minus 2:
I know what merchant I want to promote and know what industry I want to use, so it's time for research and putting together a landing site. I say landing site because adwords hates landing pages now. If you use them and they are not part of a larger site, you will get slapped with high minimum bids. I know, I've been through it in the past. I had one site that made me a lot of money and google decided one day they didn't like it, so they upped my bids to $11 minimum on any keyword I had or entered. They're basically saying, "F you, if you want to use us, pay us a fortune". It works because people either quit using them or figure out how to use them the way they want. The way they want does not always make sense, but it is what it is. (I love that saying.)
So, for my merchant in the uk debt management sector, they will allow me to collect all information on my own site using my own form and then forward that info onto them. This is beneficial for several reasons.
1.) You get to see the data and make sure it's legit before submitting to the merchant.
2.) Because of that, you know exactly how many quality leads you are sending and information is power.
3.) No merchant phone number begging the person to call them. Usually if they call, you don't get paid. Although some sites have phone tracking too, I just prefer to collect my own info.
4.) Branding. If you've ever used white label sites, you know how good they are. You can compete with the merchant on a level playing field.
5.) Tracking. You get to track your keywords all the way through. You can use your own or use adwords/overture/msn's tracking, but at least you will know the keyword that converted. This is difficult with most merchants. (Although there are some that will work with you on this, but you need to be sending some decent business first).
So, my landing site is going to have several pages, lots of useful, related content and I like to show my form on every page, (ie make it part of the overall site design). I had a buddy of mine make up a template for me because I do lots of things, but I stink at graphics. My site cost $50 total but that was for a one page template. I filled in everything including the form and back end code. (If you can't do that, you can pay someone to do it too). Basically, the person fills in my form, I get an email with the information, I get an entry into my own database, the person gets a thank you email and then the information is checked for valid data. If the data is valid, my scrip auto submits it to the merchant using another script I wrote. IF the data is not complete or faulty, (fake app or just funky looking phone, email or postal code), then I get an email telling me it wasn't submitted to the merchant. That lead is not dead, I can contact the person and get them to correct any faulty data. IF it's a fake app, my script will catch it, so the merchant doesn't recieve crappola.
Now from a ppc perspective, the site must reflect and be relevant to the keywords, ad copy and overall theme of the industry. So I use a simple echo statement to show the keyword in the meta tags and at least twice on the page body. (All my destination urls send the keyword to the landing page.) I've found this increases my quality score on adwords and gets me cheaper bids. It will also help on msn and overture.
The site is half the battle, once complete, you can tweak it to suit your needs or improve conversions, but make one change at a time and document all your changes and their results. The hardest part of ppc marketing is finding those golden nugget keywords. Golden nuggets are ones that convert to a lead at a cheap price. In competitive areas, they are harder to find, but they are out there. You have to be creative. I use the search engines to find my initial keywords. For example, I can type in "uk debt advice" into google.co.uk and browswe through the results jotting down any phrases that stick out. Then you just keep on entering new phrases into google and find more and more keywords. You can also have a look at newspapers, magazines and even the television for ideas. I've even gone so far as to ASK the person what other things they might type into a search engine to find what they were looking for. (After they have completed my form). You can also use the hundreds of software packages out there like wordtracker or keyword discovery, but remember, everyone uses those, so chances of finding something new are pretty slim. Another good technique is to build a list of adjectives, nouns and verbs then use a "jumble" script to put them togther in various combinations, then exact match them in adwords. Yet another great technique is to find misspellings of high volume keywords, you can even combine those misspellings using a jumble script.
Anyhow, keywords are the biggest part of your initial and ongoing tasks.
Day T-minus 1:
So today I continued working on my site. Filled it with content and a privacy policy, (a must now for ppc engines). I also tested my scripts to make sure they can handle everything I want them to. All is looking good on the site front.
On the keyword front, I now need to take that list of keywords and put it into adwords, overture and msn. Each system is slightly different from the other with adwords being the easiest to use and hardest to master from a profitability point of view. But adwords has the most traffic so I've got to use it. I just make sure I approach it with a slow and steady, lean and mean approach. I will make one campaign and have hundreds of adgroups in that campaign. I will never dump 2,000 keywords into an adgroup because that is just nuts and won't work. Instead I break my list up into groups of 1 to 25 words, all very much related and laser targetted to ONE thing. I will then make an adgroup for those words and use only phrase and exact match to start. Plus I have to make sure I turn off the content network as it is just junk. I make 2 ads per adgroup and make sure I've got my destination urls entered correctly. I also put my conversion code on my thank you page so I know what worked. I only entered one adgroup today as it was getting late. It has 10 keywords in it. For bidding, I like to bid high to start, (get on the first page), then trim the results as things convert or don't convert. I use my target ROI for this. If I want 100% roi, I have to pay half the payout for a lead. So if one of my lead pays $140, (one type does), then I can pay up to $70 for a keyword before I reduce the bid/leave it alone or delete it. That $70 is total spent, not the actual bid. The bid will be whatever it takes to get me on the first page of search results initially.
For overture, things go a bit faster. I can dump all those keywords in OR I can group them like I do with adwords, just not as narrow OR I can test batches at a time. I like to test large batches or related keywords at a time. This way my ad hopefully be more relevent and convert better. I make use of the "category" feature on overture, dumping batches into their own category using the date and them as the category name. (That way I know how long things have been in the system when going back to prune things later). So I may dump 500 to 1,000 keywords into one category and bid to first page results again. I have written a simple php script that takes a list of keywords and spits out the excel bulk upload sheet for me, with all the url keyword tracking 100% accurate, so I can see which keyword converted. THis is because OV doesn't have the {keyword} feature.
For msn, I take advantage of their bulk import feature for keywords, I am still trying to figure out how to only have phrase and exact match without having to bid $0.05 on broad. But I'll figure that out eventually. Both overture and msn use human editors, so getting keywords approved can take time. MSN has only a fraction of the traffic of google, but it converts very very nicely.
So that basically brings me up to the first day. I will continue to enter adgroups into adwords and dump keywords into overture and msn.