|Product Ads won't spend my daily budget|
| 5:22 pm on Jan 7, 2013 (gmt 0)|
SO, i've got products ads set up for my feed of 4500+ ecommerce items that we sell. I first just had everything set up under one ad group that had all the products. I told google to spend a CPC of 70 cents, and I have set to enhanced (it tracks conversions and adjusts accordingly). Set the daily spending limit at $250.
During december all was going well and we would spend 250 every day and get a really good ROI.
Now we are still getting the same good ROI, but it won't spend the whole 250. Now it spends around 150.
We have it set to show ads for US and Canada.
I called adwords support and they suggested segmenting our products via auto targets. So i have done that for our top keyword matches, and upped the cpc for them. Still it won't spend as much as I would like it to.
So the questions is what am i doing wrong? I can't image that in the USA and Canada there just aren't enough people searching for our products for us to spend $250. Hell we want to be able to spend $1000+ since we are getting such a great ROI.
So what's the problem here? Is there anything I can do?
Thanks for the help.
| 5:25 pm on Jan 7, 2013 (gmt 0)|
PS: i just checked, it's not that we are getting a smaller percentage of the impressions. We are getting less impressions as well. Again i can't image the US population is just not searching enough for our products. It's not a niche product.
| 12:08 am on Jan 8, 2013 (gmt 0)|
Have you checked your feed for any errors?
I know with ours, we'll have it (PLAs) running great, and then it would drop off - only to check the feed, and there were errors.
Merchant center can be a pain sometimes.
| 10:03 am on Jan 8, 2013 (gmt 0)|
i don't use that system ... however you do know that sales drop right off in january, especially if you sell anything that is gift related.
| 6:13 pm on Jan 8, 2013 (gmt 0)|
To answer your questions:
1. Yes i checked the feed. The feed is perfect, no problems.
2. I would understand if conversion rate went down. If we received the same number of clicks, paid $250 but didn't sell as much that would make perfect sense to me. But we are not getting the same number of clicks. Clicks have gone down. Impressions have gone down by the same amount. Goggle is not showing our ads as much.
Now, I understand that certain search have an up and down cycle, and Christmas time it going to be a huge time for anyone selling anything. Our sales go up big time over the holidays in general.
That being said this is google. Google probably gets 10 million searchers during peak and say only 1 million during no peak. They should be able to give me $250 worth.
Is it the bid? After the holidays do you need to bid more? Maybe there are like i said 1 million impression slots, just that there are the same number of people competing for them, thus the bids need to go up?
I suppose i should test.
| 8:06 pm on Jan 8, 2013 (gmt 0)|
New bidders, position, bid, price competition, auto-targets, negative keywords, promotional ad text, image quality, trusted store status... you need someone to look at all the details here, the relatively superficial inquiry here, imo, leaves too many things out.
Google's morphing the relevance over time, including bid and brand strength and more. Take a high volume, high CTR Ad Group and do some bid experiments, to see what the bid-volume control ratios looks like, then wrap lessons learned into other auto-targeted Ad Groups.
You should be, over time, lowering the bid for your "all products" Ad Group, as you gain bid / volume / reach / ROAS control over all of your other (not all-products) segmented Ad Groups. If the targeting and bid overlap between your "all products" and specific Ad Groups is significant, your bid controls can be squishy. Make sure your Auto Targeting method of choice has relatively little overlap (where you need bid segmentation), or squishification will fungibilify your efforts.
I'm thinking what I just said would make sense to like 7 people, and 2 of them are my biz partners. Sorry. :-)
| 8:41 pm on Jan 8, 2013 (gmt 0)|
I have auto targeting on the 3 most popular categories with a higher bid now. Should i put the negative auto target in the all products ad to take them out? So there is no overlap?
| 7:12 pm on Jan 9, 2013 (gmt 0)|
use bids, not negative auto targets.
you want to end up here:
Ad Group A / Auto Target A / $0.42
Ad Group B / Auto Target B / $0.49
Ad Group C / Auto Target C / $0.72
Ad Group D / Auto Target D / $0.37
Ad Group All / Auto Target All / $0.15
Anything good percolates up from "All", target it in a new Ad Group / Auto Target combo, raise it's bid above the $0.15.
Think of it like people - they all start off equal, pull the better performers out of the group, deal with them individually. Depending on your overall size, you'll figure out when to stop yanking from the general population. But ideally, as you lift more out of "All", it empties out. If your size permits it, keep doing that until everything has it's own gas pedal, leaving the "All" really an empty bottom feeder (your datafeed guy will mess up your adwords_labels or adwords_grouping or some other targeting mechanism's data, and you'll still have your "All" catchall there to keep her flowing.
| 7:46 pm on Jan 9, 2013 (gmt 0)|
and how do you determine what the right bid it? I've been putting bids high, but the average bid is smaller. That being the case should that be my bid? Or if the average bid is the same as my bid I should increase it.
Or does having your do to high cause issues?
| 5:04 pm on Jan 10, 2013 (gmt 0)|
monitor conversions, bid for ROAS.
(i'm assuming you're a profit / ROI centric advertiser, which may not be the case for your goals - sorry, it's my nature. branding centric PPC is for large enterprises or dreamers, i prefer small and profitable.)
| 5:35 am on Jan 11, 2013 (gmt 0)|
|squishification will fungibilify your efforts |
Seems reasonable to me! :)
| 4:48 pm on Jan 11, 2013 (gmt 0)|
| 7:35 pm on Feb 2, 2013 (gmt 0)|
Fantastic advise from @RhinoFish - this is the way to work it. Temper with negative keywords from search queries.
Also so much funny stuff at the end of this thread, love it.