Horribly inaccurate! Google hasn't provided much in the way of explanation. It use to be worse because the numbers were based on the combined volume of both the search and all their networks. They changed this some time in 2010. Here are some other considerations that affect your impressions volume:

• Is your bid and quality score so high that your ad is literally running at #1 ALL the time?
• Are your campaigns capped or cut off by budget or scheduled time to show?
• Do you have any negative keywords applied to the campaign?
• Are using exact, phrase or broad match? Google supposedly bases the volume on exact.

It is not wise to completely rely on the Keyword tool local search volume. The other metrics like total search results, and competition help craft a better keyword list for your campaigns. I also like to export the entire data set, and look to see how the local volume is changing over time. Is it trending up or down? Is it a consistent average?

I can think of several things you could do, and depending on your campaigns, one or a combination of these could work for you:

1. Change the Ad Schedule settings and use "Bid Adjustment" to be more/less for desired days.
2. Change your Ad Delivery settings to Optimize for conversions. Let Google naturally automate this for you.
3. To create a hard daily budget for each set of days, run two separate campaigns with specific days in the settings. (I don't actually recommend doing this unless the other settings don't help you accomplish what you're looking for.)

I've read in the past that geo-targeting is about 85% to 95% accurate. That's the short ball park answer.

While search engines are getting better at geo-targeting, I'd say that right now it depends on how specific of a market you're targeting. Done on a national, state or (major) city level and it's going to be pretty accurate to about 95%. If you're narrowing down on specific zip codes, or using custom shapes you will see clicks from non targeted areas.

Brand + Product, in that order for Title Tag is definitely recommended not only for your SERP, but also to influence the quality of your Product Feed. Usually we do Brand + Product - SiteName. And if there is room, a USP at the end of that too. so: Brand + Product - SiteName + USP is our usual Title tag pattern.

It's also good to have some variation on the On Page Title (H1) and make it different from the Title Tag. Usually the H1 is just the Product Name itself, and H2, H3's will reiterate the Brand/Category/Offer, etc.

You'll probably get some different opinions on this. Your site link architecture and the design of your archive/search/category needs to be assessed to figure out the best solution.

If I had to pick one or the other for the nofollow attribution, I'd pick the Archives over Categories because they have less keyword value. Generally, I'd say no need to use it. Sure, you might say that those pages don't have much value since all they contain is a repetition of content on your other pages. And some may try to use the "nofollow" tag as a way to prevent any link juice from flowing to those pages. But PageRank sculpting using this tactic isn't really effective.

It would be more effective for your SEO to redesign your blog/website sidebar/footer links so that you have "less" links pointing to your Archives and/or categories.

Using Wayback would be strange.It's just a quick WHOIS lookup to determine age.