I'm looking for software where I could paste in keywords and it creates seperate ad groups, dynamic titles/urls etc to save bundles of time. I don't suppose there's any out there is there? Pete
Damn, that kinda sucks for those who can't code and gives those who can a huge advantage. Not fair.. ah well. Pete
I don't really think so. Write out a spec for what you want, pay $500 and get somebody on GAF.. Remember you also need to pay for API Calls.
Woah, I never knew that existed. (AdWords editor) Awesome - thanks! By the way oziman, you have no idea what an advantage coders have over none-coders on Adwords. Huge huge huge.. too huge for words... This will do me in the meantime though. Pete
Why do you think so (I'm a coder btw). I've helped manage some pretty big campaigns, and we never, ever used the API, and we did just fine.
Aqua (http://google-apility.sourceforge.net/agua.html) uses the Adwords API. I'm not sure if it will do what you want, but it's open source, so you could add features you want.
IMO you can use the API to automate some bidding and other stuff (setup is a biggy) but unless you (you being the person building or using the app) know AdWords inside and out you'll automate yourself into the red When it comes to AdWords I consider API/Automation to be a phase two activity...and the majority of people I've dealt with are still in the early parts of phase 1...figuring out what works.
I think that unless you're managing massive campaigns where the advertiser doesn't really care so much about the spend it's a good idea. I'm not so hot on Automating AdWords in General.. I might have to do it soon for Yahoo! because they don't have the MCC for Google.
There are pros and cons either way...I tend to not introduce much automation until a lot of testing has been done and I'm only automating what has been proven to work. Here's a pretty good related post on bid management technology.
He has a good point. I'm just against using automation because you happen to have a Bid Tool already. I'm still a big fan of DIY, tinkering, etc.
Guys, I agree with you wholeheartedly that if, in the best of all possible worlds, every advertiser had the intellectual capacity and needed personnel bandwidth to build an A-Z Tinker To-Do List, then doing everything manually would almost always be an ideal solution. The problem is, we're not in the best of all possible worlds (well, actually I think we are in general, but that's another topic). In the marketing world I live in, 95% of PPC advertisers have chronic personnel, training and focus shortages, and they spend so much of those scant resources doing everything by hand (including basic ROI analysis) that they only get a fraction of the marketing & merchandising work done that they need to do. Properly instrumenting a PPC campaign first - to start collecting accurate keyword-level impression/click/cost/revenue/margin data - is in my view the most important thing to do, followed closely by everything else. Getting a system in place that can automate the process of collecting, modeling and then acting on the data based on the advertiser's goals/constraints frees up the advertiser to focus on the tinkering you speak of - ad copy, landing pages, sales funnel analysis, multivariate testing, etc. Testing everything (largely by hand), and automating things only after you've done all your testing is not, IMO, a scalable way to go about it. If scale's not an issue, then fine. But for most folks it is.
Shorebreak - I think the problem is two fold. One, I think a lot of people who do search marketing don't necessarily have a good education as to what they're doing. Two, I think there is too much of an obsession of statistics. I agree with you that keyword research and proper campaign setup is very important. I think the problem is that people go in without really thinking about how much they want to pay per conversion, how they are going to implement analytics, how they will do split testing etc. There's a perception that PPC is either very intensive or fire and forget. It's neither, in my opinion. You need to go in with a plan like any other business - research, budget, execution and analysis. I think because of it's nature people forget that. My $.02.
Great post & points shorebreak. In response to the above, since I posted earlier that this is a method I use, I assure you it's very, very scalable. Even when I'm doing things "by hand" there's a level of automation involved, especially in the data analysis aspect of things. Simply put I figure out what's working on smaller than it could be scale, then ramp those areas up and implement rules (based on the early data collected) and conitnue to monotor them and make adjustments when needed. Unless you're walking into an established situation (systems and data in place already) I don't see any other way to approach things.