|I have always been reluctant to show Google my conversions and results because I feel they can take advantage of knowing how profitable my campaigns are. Not so much that they would/could raise CPC prices even higher, but when it comes down to getting credits for "suspect" clicks. Historically I have been successful getting credits for broad match clicks that are ridiculous. I feel going forward that if Google can see that for every dollar I spend in Adwords I make $5. I fear they will automatically reject credit requests and think that I am just being greedy. |
That has not been my experience through all my clients, all my accounts and all my years with AdWords and Analytics.
Google can't ever know how profitable your campaigns are. You could be purchasing clicks at ten cents for items that sell for $1995; Google doesn't know what your *other* costs are. Maybe it costs you $1700 to produce, plus $100 to process an invoice.
Moreover, it's in their best interest for you to do well; accounts that do well tend to expand.
I'm eagerly trying to persuade all of my clients to use analytics and consider conversion tracking; I can't honnestly think of a single reason not to; even with very expensive tracking software in place we are implementing this as if Google can show the same or better results we could save a fortune.
I agree with the others. Google tools are just too useful for commerce sites.
Combine the results from GA with Google's A/B testing tools and you are in a very powerful position. My bounce rates dropped from 60% to 26%, my sales up by 30% and every other aspect up (pages viewed, time on site etc).
I could not cope without the tools these days!
Actually, I won't take on a client who isn't doing analytics of some kind anymore. Didn't realize until I got into it how blind I'd been flying those years without it.
Some people believe that by using all the tools provided by Google AdWords, they actually reveal too much data from industry to a keyword specific level which causes the cost to vary (goes up only).
One of the facts that backs such belief is the fact that you pay $0.75 per click for being on #1 in Premium (yellow) Area while there is nobody else there.
How Google derives that specific price for this keyword? Based on the history I guess. If you add the conversion on the other side, there you find how they could tune the algorithm against your profit (which BTW is self-tunable these days for sure).
This is all guess, obviously.
On the other side, all the folks that state "do use it" have their own reasons and beliefs for saying so - they simply think that way, or they have a commercial reason to say so.
Finally, the fear about revealing specific stuff too much so that you could end up by paying more (having your profit shrinking down) probably has no much realistic base (theoretical and personal only), as the rest of the people have already revealed too much anyway.
Finally, you have the technical side of it - some people make calls from time to time about discrepancies between their own and conversion data coming from Google. It's probably something on their side, but just you're aware of that, too.
To round up - the idea of having a tool that shows all data in one line for each keyword (or however) - that can only be positive I would say.
[Disclosure: I work at Omniture, the largest of the not-free vendors. I've been in the SEM space for 7 years, during which time I've seen hundreds of top 5000 SEM campaigns.]
Here is an article [sitening.com] written by Kate O'Neill of Sitening describing why in analytics and online marketing, you typically don't get what you don't pay for. Please excuse the linking to an external article, but I really think the article speaks *specifically* to avalon37's question.
To speak directly to the op's question, my company has been using Google conversion tracking for about 8 months now. We use it to track sales across two separate domains and it works very well. In fact it works just as well as shorebreak's product for everything we need.
The reason not to use it? Support. Tracking down problems and getting answers from the big G was just short of a nightmare. But, I got it to work, and it is still free.
Just to respond to Shorebreak and the article that he linked; we have clients using one of your biggest competitors, 9 times out of 10 we will turn to Google analytics to answer a question (where it's available), purely because most questions are about the basics, it's not every day you need to go beyond that.
[Disclosure: MMT does provide "industrial strength" analytics as part of its' offerings - usually for large organisations - but work with GA for companies who don't need highly complex analytical requirements]
It is "horses for courses", I suppose. For the vast majority of sites GA is absolutely all that is needed - but let me give you a "real-life" example of where something with a bit more flexibility is required.
A household name multi-national that sells in more than 100 countries wishes to not only report on nearly 1,500 possible "events" (downloads, purchases, registrations etc.) but also wishes to group reports into things such as "visits to European sites who have downloaded x and then registered for y" or "visits to German language sites" or "visits to Benelux sites" or "visits to sites for product group A in the Middle East - excluding Libya and Algeria and any paid search referrals" and need this reported in hundreds of possible combinations of event choices and/or country/language/regional groupings - such reporting to be able to be accessed by hundreds of different staff in different countries - all of whom wish to see reports that are relevant to them - while HQ wants to be able to see how different areas are performing in relationship to each other - and in real-time!
Setting that sort of thing up in GA is (at the very least)rather difficult!
Using a good enterprise analytics system can do this, often by providing the technical advice promptly to assist in the set up and (possibly) adapting the methods of displaying data to suit the client's needs.
Obviously, this comes at a cost - but can be done quickly and efficiently by a good provider - whereas GA, whilst very good, is a package where you are left to see if it can do what you want.
So in answer to the OP's original question, I guess it comes down to this, if you can show what information you can provide using a higher level product, that it will be commercially valuable to those with the buying power, and convince them to buy into it, you may be able to avoid using Google ;)
Usually, in these situations, there is not a lot of convincing to be done - the need has already been identified and (quite often) attempts have been made to do what is needed by alternative (free or internal analysis of logs) methods.
It is often easier and quicker to go out to the market and ask providers to quote for x in y timeframe! This can also be done for companies who can install GA for clients.
It also helps that very large multi-national corporates really don't like the idea that another multi-national corporate can eyeball their conversion data!