Google Search Top 1-2-3

[R. C.]

Friends,

I know this has been discussed, ad nauseam, and I DID read a number of threads re: CPC, CTR, etc.  I think this question is a new twist on the topic.

If you are not aware, the Googles have a new aggressive program for advertising. The new program is based in an AI they have developed for improving AD performance.

I have been experimenting with the concept and the sales results range from poor all the way up to the remarkable... moderate.

However, the increase in traffic to my site is good to pretty-darned good.

Because of this data, I know my AD copy is working but my "close" to get the sale is not working.  The Googles rep, yes, I have a rep a no extra charge, called me to make a suggestion.  The suggestion, if I up my price per bid to between $.78 USD and $2.01 USD, I would probably land in the top 1, 2, 3 of related searches.

Yeah, I know, but that is actually what he said. For those who don't know, making the first page in a Google search is outstanding. Making the top 3 is phenomenal.  Of course, raising the per bid amount would lower the number of available displays, and stop displaying when the daily budget is reached.

Question: Assuming I can get generate pull-through/follow-on sales, what is your opinion of a per bid rate of $.78 USD to $2.01 USD?

Cheers,
R.C.

[LilyBLily]

Can you make a profit if selling a book costs you multiples of $2.01? How many clicks gets you a sale from your Google ad?

Do you sell books on your site? Have you tried these same words with Amazon, BookBub, and Facebook ads so you know they deliver sales? Not just clicks, but sales?

I'm scratching my head trying to imagine a word or phrase that someone would search for on Google that would bring up one of my books AND convince someone to buy the book, let alone go to a strange website to consider buying it.

Of course the Google employee wants you to spend more money. That's a given. Does Google offer any comps (anonymized) to show you how a $2 bid would make a difference in sales?

[R. C.]

All very good questions, many I asked and I will try to respond.

Can you make a profit if selling a book costs you multiples of $2.01? How many clicks gets you a sale from your Google ad? 

Do you sell books on your site? Have you tried these same words with Amazon, BookBub, and Facebook ads so you know they deliver sales? Not just clicks, but sales?

I'm scratching my head trying to imagine a word or phrase that someone would search for on Google that would bring up one of my books AND convince someone to buy the book, let alone go to a strange website to consider buying it.

Of course the Google employee wants you to spend more money. That's a given. Does Google offer any comps (anonymized) to show you how a $2 bid would make a difference in sales?

Can you make a profit if selling a book costs... --- The answer here is surprising. I have more sales from Google search linking through my site than I do from other advertising. Am I making money? Not hardly but I am creating a following.  Not sure $2.01 is something I would consider but $.78 might be feasible in the short term.

Do you sell books on your site? Have you tried... --- I do not sell books on my site, only KDP.  I have analyzed and refined the keyword words but I stopped FB to experiment with Google, I am going to use the "successful"  words from Google in new FB ADs.  I have no presence on BookBub... Yet.

Scratching my head... --- This is the thing that might be broken in my marketing model experiment. The theory is my site gives them an overview of the series and direct links to KDP. BUT, and this is the real question: Why is the significant increase in traffic not resulting in sales?  Bad, blurbs on my site and KDP?  Probably.  Wrong demographics? Probably. 
--- I have refined the blurbs, again. This weekend I am narrowing the demographics. Hopefully it will translate into something positive.

All about the money... --- Of course it is all about the spend, and I am a huge skeptic. The thing they are pushing is the newness of their AI to refine the placements. I think there is something to it but... only time and money will confirm. 

Cheers,
R.C.

[RBC]

I'd be very careful and skeptic about it.

Of course the reps will tell to raise the bid. Doesn't mean you will get charged that price necessarily but it will increase your ad spend highly.

There is little chance that 100 clicks from Goggle ads will end up in 1 sale.  That's huge cost for 1 sale.

Also see which keywords they recommend to go for... 'thriller books' might be a big keyword but it's not really a 'buying intent' keyword. Like for example, 'book cover designer' for me is a good keyword, but better one is 'book cover designer for hire'. The intent of the search is more clear and targeted.

So be careful of splurging on general keywords. That's a thing to test and at $2 a click that's crazy budget need for testing.

I hope it works out, but only if that new AI is really good (and I've heard advice to avoid it from Media Buyers, people who specialise in buying traffic).

[R. C.]

I'd be very careful and skeptic about it.

... only if that new AI is really good ...

Too true: Call me skeptic.

I am running an experiment with the singular goal of increasing traffic to my site. I want to see if I can can get clicks and follow-on threading on my site. Early results are interesting.

• Small data sample, less than 24 hours
• Using the new "Smart Campaign"
• Very low bid amount (This is calculated on the formula: monthly budget / days eligible to run = max daily budget)
• Narrow, US biased time, and English speaking demographics

Results:

• CTR = 7.43%
• CPC = $.27
• Avg. Time on Page: Up 143%
• Number of pages viewed: Up 64% 
• Sales: TBD

Granted: The numbers are small, the AI says it is "learning".  I'll give it a couple of days, collect the statistics and report back.

If the trend continues, I will point the engine directly to a "purchase" link.

Cheers,
Roger

[RBC]

You're not even sending people to where they can buy something? Scary!


You should advertise with results in mind, not general goal of ''improving traffic''. More traffic is useless if it's irrelevant traffic and its' easy to get traffic. Padding numbers won't help you. A philosophy that's great for PPC ads is Direct Response Marketing. AKA there should always be one thing, one tangible result aimed for. There should be at least email sign up forms and then you can count how much you pay per subscriber in this case.

Chasing general goal is easiest way to waste money.

Study up on Google ads from NON-Google specialists like Perry Marshall (but they are certified at the same time, just less bias):

https://www.perrymarshall.com/google/

Get a book or two on it. His book is great.

Or website that actually targets to help people who buy traffic for living (called Media Buyers):

https://www.adskills.com/library/

I've wasted many a dollar with Google and Facebook sending traffic. Padded numbers look great but doesn't add anything to bottom line. :(

[R. C.]

You're not even sending people to where they can buy something? Scary!
...
Chasing general goal is easiest way to waste money.

Study up on Google ads from NON-Google specialists...
...
I've wasted many a dollar with Google and Facebook sending traffic. Padded numbers look great but doesn't add anything to bottom line. :(

Thank you for the input.  Yep, scary.

It is however, a simple little experiment to see if traffic has value based on the "new AI".  Past may be prologue: Of course, traffic does not equate to sales.

But, and here's the rub, I can already tell the increased traffic has led to secondary and tertiary page views.

This weekend will be the determining factor. I lowered the "per bid" amount. I will know Sunday evening if this was just another boondoggle.

Cheers,
R.C.

[RBC]

You're not even sending people to where they can buy something? Scary!
...
Chasing general goal is easiest way to waste money.

Study up on Google ads from NON-Google specialists...
...
I've wasted many a dollar with Google and Facebook sending traffic. Padded numbers look great but doesn't add anything to bottom line. :(

Thank you for the input.  Yep, scary.

It is however, a simple little experiment to see if traffic has value based on the "new AI".  Past may be prologue: Of course, traffic does not equate to sales.

But, and here's the rub, I can already tell the increased traffic has led to secondary and tertiary page views.

This weekend will be the determining factor. I lowered the "per bid" amount. I will know Sunday evening if this was just another boondoggle.

Cheers,
R.C.

Page views or page reads? Page views are again, same as traffic, useless. Page reads increasing would mean basically 'a sale' happened. So that would be awesome.

Good luck!!!

[R. C.]

Page views or page reads? Page views are again, same as traffic, useless. Page reads increasing would mean basically 'a sale' happened. So that would be awesome.

Good luck!!!

Apologies for not being clear. Page reads: The average time on page has increased by 3.56...  Which, in retrospect was the goal of the experiment. 

Lots of my short/micro stories are being read. 

Cheers,
R.C.

[R. C.]

Summary Results: The experiment was successful in proving the hypothesis.

Hypothesis: Does the AD copy and the new AI result in new traffic/clicks?

Goal: Using the new AI, drive traffic to site.

Conclusion: The new advertising AI will drive traffic to a site.  My site has had more page reads, time on page, in the past two weekends than the prior six months combined.

But... There is always a but...

I ran the experiment over three days, Fri, Sat, & Sun.  The results were, unfortunately, as expected.

Friday: Good traffic, medium to medium-high cost per click (CPC).

Adjusted formula to drive a lower CPC. (CPC aligned to KDP and FB)

Saturday: Zero traffic. Not low traffic, zero traffic.

Reversed formula to suggested pricing, above KDP and FB CPCs.

Sunday: Good traffic, medium to medium-high cost per click (CPC).

If I were selling quilted hiney wipe with free shipping, the new AI has value.

Value clicks that result in paying for an eBook... Not so much.

Next steps, refine KDP and FB targeting to increase the probability of a sale.

Happy to answer any questions.

Cheers,
R.C.