Welcome to the Q&A Forum

A CNAME is a DNS record that says that domainA.com lives where domainB.com is. That means you then do another lookup on domainB.com and get its A record. Somewhere down the chain you have to have an A record. An A record is what ties a domain to an IP.

The problem here is that a CNAME is not the same thing as 301. If you go to the CNAME as mentioned above, your browser will still say domainA.com. We use a CNAME because we have a load balancer with AWS. So our site resolves to a CNAME that resolves to the load balancer address but it still shows up as www.ourdomain.com. We have dozens of URLs like this pointed to the same hosting configuration and each domain is seen as the original TLD. This will cause duplicate content problems for your client.

The correct solution is to set them up with an A record pointed at a web server and then have your web server return a 301.

You can adjust campaign budgets by day. it's just not obvious where.

1. All online campaigns
2. Campaign tab
3. Automate button (to the right of the create Campaign button). Be sure to check only those campaigns you want to automate
4. Change Daily Budget When...
5. The real trick here is Frequency. To change your budget on a given day, you need use Weekly and then you can specify by day.

Remember to do the math. This method increases or decreases budgets, it does not have an option to directly say "On Monday i want to spend $x and Tuesday spend $y", you have to have it adjust the budget based on the budget WHEN IT HAPPENS. So on Tuesday your budget is $270. On Wed you will increase your budget by $54 and then you'll want another rule to remove that $54 on Thu.

If you elect to do percentages you can't just say "raise by 20% then lower by 20%" or you'll start creeping your budget. To make a 20% boost work you have to increase the amount you decrease by. For instance, we used to drop our Thu spend by 36%. That meant the boost had to be 56.25% (small rounding error but acceptable) on Fri to go back to where it was.

Why do you NEED a blog? More importantly, why do you need a blog for what is a highly technical (and probably arcane) topic?

Blogging is not for everyone and everything. Blogging is for where there is some form of dialog or constant change where you can generate fresh content people want to read on a regular basis. I have one website where a blog makes a lot of sense. It lends itself to lots of artsy-deco blogs and we have some talented young ladies who have found a niche in writing that. Their readership isn't stellar, but combined with social media efforts it works decently well. I have many other websites where we have no blog and never will, because attempting to blog about those topics would be pointless because nobody would ever read them. Instead, we put the information out and then market the sites as best as we can.

It sounds like you're a bit too niche here. If all you do is SEO, and don't offer, say, online marketing, you're really missing out because SEO as a job by itself is really hard to sustain. Offer to set up an Adwords campaign or offer to do some work on their social media. Maybe they need more local efforts. Do they show up on your phone if you search for what it is that they do? I find myself using Google Now a TON for local because Google makes it stupid easy to find. Websites, maps and phones numbers (with one click to open up my dialer and call).

Offering more services should help keep your clients happy. Especially if they don't have the time (or drive) to work with you on SEO.

Physically hosting your server in a target country does help, but that help is miniscule. In Geotargeting, the largest signals come from your link profile. So if your links come from the US primarily, you should rank better in the US. You also have a genertic TLD instead of a ccTLD, which helps.

Some possible actions to help your site be seen as US would be to do some marketing within the US. Another option is to see if you can get covers.ca and then work on geotargeting that site to Canadians. That would send a strong signal that covers.com is aimed at the US.

Have you journeyed to Seattle and tried meditating in the Moz break room while holding a Roger plushie until security escorts you out? (sorry, I couldn't resist)

First off, you need to better organize your site. It's not horrible, but the URLs are all over the place. Blogs are smashed together onto single pages. Content is disjointed.

Second... pervy quest? What? You need people to take your site seriously and dangling the prospect of NSFW imagery isn't going to help that (and might get you added to some domain filter lists). Don't play up the "naked" part. There are plenty of legitimate reasons to use that word.

Third... what is the site about? It's not obvious from a quick glance at the front page. Theory of Everything? That's cool but... what was I doing here again? You need a better design (too many stock images and cut out the rotating images as it distracts from the page) and you need some focus. All the content needs to pull in the same direction. What is it they want visitors to leave with? (besides a crystal bauble)

You need to understand that SEO is more than just about SEO scores. Scores are more about letting you know if you've screwed up. Only Google can tell you why you don't rank for your terms. Maybe you're competing with a large brand. Maybe you don't have enough signals coming from elsewhere (social, links, etc). There's a myriad of reasons.

My advice to you is to start a marketing campaign. Do something to get your site out there. Maybe its Adwords, maybe promoted Facbook posts, or maybe just some guy wearing a goofy sign on the street corner. Don't assume Google will come beating your door down just because you have a Moz A rank site.

The bulk of his argument could be summed up as "Fresh content usually wins". It has some merit. But the rest of the article could be taken as saying "Game what you can now and move on."

SEO is a subset of marketing. It always has been and always will be. SEO is where you market to the search engines.A major part of Google's problem is we don't have any idea what constitutes what they like anymore. So his guess about it all being about freshness sounds good until you find searches where that's not true. But you CAN produce quality content and still lose the SEO game. But you can lose the SEO game and still win the war.

The danger for this guy is that he seems to advocate link spam. He heavily caveats it but offers no real reasons not to engage in it. I have no doubt link spam still works. The question you have to ask yourself is "Do I want to build something on top of that which Google is trying to actively destroy?" If you can live with being a flash in the pan, go for it. Make your money and then watch as Google throws it in the garbage. And hope that the next time around they're just as gullible.

I would 301 the pages and get them out of your site's index. Even if you canonical all of them Google will still have to index 1000 pages instead of 1. The 301 will transfer most of your rank to the new page and you'll improve your crawl budget.

Why take the 301s out? Just leave them in there in case there are links pointed to them.

Not really. People are obsessed with calling duplicate content a penalty. It's not really a penalty as much as engines trying to sort the noise out. If pages A, B and C all have the same content, there's only room for a single #1 rank, so that means the other pages will have to lose out. They're not penalized, just not ranked as highly.

The place where duplicate content is harmful is if someone scrapes your site and outranks you. That's a different problem entirely. In your case, the concern is that you will harm your site. That's not the case. The engines will simply pick one review page in your case and devalue the rest. In my book that's not necessarily bad because people are still going to come to your site and read a review page.

Canonicals can help by letting you pick which page wins the duplicate race. You simply tell Google that B and C are duplicates of A and, thus, A wins #1.

While not having a 301 or even the original domain is problematic, I don't know that it is a bad as you think. Unless they took the previous site and set it up exactly as it was (which I doubt the new owner would do that), you should be able to set up your new domain and start over as it were. That's not as hard as it sounds and, believe it or not, Google can figure out domain migration like that without a 301.

Once it's up and running again you need to market the fact that it's moved. Nobody likes dead or incorrect links so go find all the old links you had and let them know that you've moved. That should get them to update and you can possibly rebound some of your original brand authority.

According to Matt Cutts, trailing slashes are automatically canonicalized. It shouldn't make any difference.

It used to be true that you had to have one IP per SSL certificate, but if you have a modern web server you can use something called Server Name Indication instead.

I'm not sure what you mean by "offered 2 types of certificate". There's a couple of ways to take that.

1. Your provider is offering you a SHA1 or a SHA2 (SHA256) certificate. If you get this option select SHA2, as SHA1 is now deprecated due to security issues and will be unsupported in modern browsing soon.
2. Your provider offered you domain validation vs extended validation. The latter is only for registered corporations and gives you the "green bar". It does not afford you any more security from an encryption standpoint (both use the same underlying systems), it just looks better in the browser.

Putting your whole site under SSL is not a bad move from an SEO standpoint. Google publicly said that SSL is a ranking factor (nobody knows how much of one it is, but the boost is minor if there is one) and robots have supported SSL for many years. What I would do, if I were you, is make sure you have a proper redirect to force users into SSL. Don't run a mixed encrypted/unencrypted site if possible.

Last but certainly not least, after you get your SSL set up you need to audit your SSL. This is a critical step many website owners miss. Just this weekend I found a site with incredibly outdated software and, as such, it supported every SSL flaw in the book. A site with SSL exploits still enabled might as well not run any SSL at all.

You could go rip the entire site to your local drive using a program like Black Widow. Then you could use something like Notepad++ to search the entire directory.

Duplicate titles or duplicate content?

Duplicate titles are a small problem while duplicate content is a larger one. You should avoid both in an ideal world. Duplicate content should be removed via 301 redirect or sorted using Canonical http://www.seomoz.org/blog/canonical-url-tag-the-most-important-advancement-in-seo-practices-since-sitemaps
Duplicate titles won't help you but Google doesn't discount pages with the same title tag, provided the content is different.

Duplicate titles can impact clickability of your result, tho. Let's say your titles are all "Blue Widgets!", but you also sell brown, black and teal widgets. Even if you get a page to rank well for "teal widgets", the page title says it's "Blue Widgets" and people might not click on it, thinking Google made a mistake. Yes, Google will highlight the content it thinks is relevant to your search, but your title tag is what they will click on.

That seems like a really lame solution to your traffic problem. So much traffic we're going to mirror your website? That sounds like a recipe for disaster.

What we do is we have a load balancer sitting in front of virtual machines. If we need more capacity, we spin up a new instance and link it to the load balancer (we use Linux and a setup program called Chef). This is 100% transparent to the end users, including bots. No URL changes are necessary because the domains are pointed to the load balancer. We can have more capacity for our websites in less than an hour. This is more commonly called "cloud computing". Amazon Web Services, Rackspace, etc, all offer this kind of setup. The word you need to know here is scalability. Azure is a Microsoft cloud (as Alan said), so it only does Windows. AWS can do Windows and Linux.

If you're interested, here's the AWS page about their load balancers

First and foremost do NOT ditch your domain name. At best I'd say launch a new site in parallel and see how that goes but you'd basically be admitting defeat and starting over from scratch.

Second, buy some PPC. You're a chain so leverage that and get some PPC going to raise your traffic. PPC is also a great way to identify long-tail keywords you're missing. But don't just buy them. Take time and groom them daily. Pair them with Analytics. Suck the marrow out of them. Live them for the next week.

Third, realize you've set the bar really, REALLY high. You're chasing a generic keyword, and a massively competitive one at that (506M results in Google for the word "fabric"). In the post-Panda world, I'd say that's unrealistic. And why are you so obsessed with that one keyword? I've found single word terms are some of the poorest to go after. Lots of traffic, but major bounce rates and poor conversions (I would assume you want to sell, fabric, not just have people come look at it). Fabric could mean cloth, but it could also mean reality. Do you sell camo fabric? Google wants relevance above all and if people are bouncing out of your site, it will fall for that term. Generic terms, as a business model, are horrific.

What you need is to understand who your audience is and what they're looking for. Are they looking for bridal fabric? Maybe woodchipper macadamia fabric? Before you laugh, consider that maybe there is such a thing and a rich person who is buying a lot of it right now... somewhere else. I've found that owning a dozen smaller, niche terms that convert well beats the generic term any day. See #2 for how to find these juicy terms.

Then what? Content! You have a pretty site. Lots of things to interest. But you need to diversify. Your front page alone is way too busy. You have 6 banners stacked like one of those crazy subs at Subway. Each has value, but crammed in they have no value. And your deal of the day... uhm, they're sold out. I can find more in 19 hours??? Do you think I'll come back tomorrow to see what 3 fabrics sold out? Stagger them. Let me know there's another deal coming in 4 hours, 32 mins 23 secs, not 19 hours.

And your titles... ugh! Your main page is titled "Best Selling Fabric" but it's a lie! Thank you for visiting, Mario, but the best selling fabrics are on another page (just click past the Mushroom Kingdom patterns from the 1980s at the bottom). And what site did I visit? Hmm... the title doesn't tell me. Is this Beverly's? Fabric.com? Hobby Lobby? They're descriptive but they still miss the mark. You need your name at the end. Even Amazon does this (and they sell fabric too!).

Last but not least, fire your moving company. They've packed you in a box. Get out of there. Watch my swinging spoon tied to a string (I don't own a pocketwatch. Work with me). You are no longer an SEO. You are a marketer. You must convince me why I should buy your fabric. Show me the money fabric! Put it out! Make your site scream it. Right now your site is struggling to sell fabric to naked people (why naked people would be using Google is your problem to solve).

Yes. It's better late than never. You might not get any rank but I consider 301s to be good policy beyond the SEO aspect. I hate going to a page where there's a link and I click it and I get a 404 or bounced to the front page. Perhaps I have a bookmark. Perhaps it's an old link. Whatever the case, do your visitors a courtesy and redirect them to the correct page.

It really depends on the source of your traffic. If the source of your traffic are old links or direct traffic (i.e. they typed the URL), I'd leave them forever (provided you can't get them updated). It's bad form to drop a redirect from an old page when new users can find old links that appear useful. A good example for me was weather.com, which recently updated their site URL structure and refused to 301 from the old to the new (old page would 404). So for me, who had bookmarked my local weather page, I now had to go back to their home page and search all over again and THEN I was back at my local weather. It was annoying when 301s are cheap and easy.

If all your traffic comes from search engines then I would say you can drop them after a year or so. That's more than enough time for the index to update and the old URLs to fade out.

Bots are to websites like ants are to sugar. They pick up content and gravitate towards it.

Gutting your site content is NOT a good idea. In fact, you'll find that removing your content pages will also remove the traffic! It sounds like you have a lot of naturally popular content but not a lot of conversions. So how do you capitalize?

1. Add links to your product pages within the content. I mean, if you make AND sell the product, let people know that you sell the product to make money. Good internal linking structure is a necessity for the functioning of a website, let alone SEO. Don't just link either. Get a small block embedded in your content that shows a picture of the product, its price, and maybe an "Add to Cart" button.
2. Beef up the content in your product pages. Naturally, product pages are anemic in content. Fix it!
3. Get some links to your product pages. Use Open Site Explorer and find out where your links are now, and see if you can get the natural linkers to maybe plug your product page as well.

"Autocomplete engagement" sounds like a terrible name for a reality TV show.

Haven't heard of it. Considering that it's Google's best guess as to what you're typing, and it's based on what other users have searched, I wouldn't say it's impossible. But the catch is you need to know exactly what your users are typing in. Let's say you sell dialysis equipment. You would need to know if your users are searching "dialysis machine", vs "how do dialysis machines work" (and those represent two very different searches). You then need to rank well for that term.

The only thing I can assume is they're going to somehow flood Google with clicks on a term. Done hundreds, or even thousands of times, I could see it working to some extent. But any successes would be transient (this is the reason why mechanisms like Panda, Penguin and Hummingbird exist in the first place). As such, I place little stock in it.

Rule #1 in SEO Outsourcing is this

There is no SEO Secret Sauce.

Put another way, there is no proprietary SEO knowledge outside of Google and they aren't selling it. This means that any company doing SEO work on your behalf should be providing you a report of what they are actually doing. If they balk at it, shut them down because it's the surest sign of a scam.

Rule #2 in SEO Outsourcing

NOBODY cares about your site like you do.

Even if I am paid by you to care, I still won't care like you do (or should). I am constantly amazed by people who are shocked by poor SEO going on under their nose by people they paid. There are multiple tools available (paid and free) to you to check on your website. And if you have a feeling that your website is being harmed, do your own research. While a good SEO will try to provide meaningful metrics about what their efforts have generated, they will always put the best spin on it (that's not an ethical lapse, just human nature). Look at your own metrics and make your own decisions. If you don't care about your site, don't be surprised when an SEO you paid runs it into the ground.

And they would have gotten away with it too if weren't for those meddling kids and their pesky subdomains

Just an educated guess on my part but I'd say that it's probably due to the indexing problems of late.

First, this index still contains data crawled up to 100 days ago.We try to make sure that what we've crawled recently is stuff that we believe has been updated/changed, but there may be sites and pages that have changed significantly in that period that we didn't update (due to issues I've described here previously with our crawlers & schedulers).

It's clear that Avvo is being crawled and indexed. You have a link on this page that shows up in OSE already

http://www.avvo.com/attorneys/34655-fl-daniel-dean-1267113.html

You might want to 301 redirect the version you don't want to the version you do (i.e. redirect non-www to www)

http://www.seomoz.org/learn-seo/redirection

You're actually pretty close to the box of swag reward. I got mine a couple of months ago and it was filled with lots of really neat stuff (I got both a vinyl AND plushie Roger).

The rewards are handed out manually so it's not some automatic process. I'd get your 2k and wait a month or so. You should get an email asking where to send your swag.

This is a bit trickier working with Wordpress and a plugin.

If you have your Analytics linked, then Google will add ?gclid=XXXXXXXXX to your URL when someone clicks through. As a result, you can flag that session as an Adwords session. Normally I would add a session flag but Wordpress doesn't run sessions by default. You will have to add some PHP programming to make that work and that's outside the scope of this forum.

If you've gotten them to like your page then anytime you post something on FB, that will show up in their feed (and this costs you nothing). Unfortunately FB came up with this long after the "Like" mechanism so they couldn't name it something less confusing, like Twitter's Follow option even though it works the same way. Liking a post is liking a post. Liking a page is to follow their feed (unless you edit your feed).

Paying for ads on FB means that, quite often, what other people in your orbit on FB like will influence what ads you see. So I'll see a Paid news feed item that notes that "Friend A, Friend B, and Friend C Like Brand X". And that's how FB extends your reach. Ads will also appear in your target demographic feeds as well (whether or not their friends have liked it), but that social element is smart because tying it to social activity increases responses.

Using FB for product comments is also another way to engage people socially.

Bounce rate, in regards to Panda, refers to how you bounce in relation to a given search, not how many people bounce from your page overall.

Let's say I search "blue widgets" and your site has a page on "government funded blue widgets". I might not be looking for that so I bounce back and go to the next page, which sells blue widgets. I never come back to Google so Google goes "Hmm, they didn't like my #1 choice for blue widgets, maybe I need to change that" and so your page loses relevance for that term. Remember, Google wants people to be happy with their SERPs and if people keep bouncing from page to page, that's not good for them. It's unclear as of yet whether or not this behavior, repeated on most of your pages, can affect your overall authority or rank, but I haven't seen anything conclusive to say it does. I think sites experiencing site-wide devaluation are either suffering from some other issue or they were ranking well for a lot of terms that they shouldn't have been.

My rule of thumb on content is that, unless it's duplicate, it shouldn't be removed.

I would just 404 the pages (provided they're not already) and move on. There's nothing wrong with having dead links to your site. Some people would recreate the pages to try and capture some of that link juice but if they're from abandoned Chinese sites I don't think they would provide the high quality you need.

If you want the best of both worlds, have the page return a 404 but return some content that would direct anyone who happened to click in to some other part of your site.

I helped my mother-in-law with her ecommerce. I tried Zen Cart (a fork of OS Commerce) but it's horrible in security and support, not to mention confusing. I've never gotten the impression that ZC is mature.

Magento is definitely much more mature and is now owned by eBay. I replaced ZC with Magento and it worked well until she shut her business down. My gripe with them is they went whole hog into Zend Framework 1, which is notoriously slow. It also has a bit of a learning curve. But it was stable and relatively secure (no major security flaws), just clunky to install.

The only way to know for sure is, after having linked your GA to Adwords, is to look at server logs and look for the following (in bold)

www.yourdomain.com/somepage?gclid=XXXXXXXXXXXX

This is how Adwords passes the session on to GA. If you click on your own ads you should see this show up when you finally arrive at your site (if not check configs).

the amount of sessions of Paid Search (in channels) and  AdWords sessions don't match up (The number of actual PPC clicks is substantially higher than the Paid Search sessions)

This is normal. Because GA relies on JS code there will always be missed sessions. Some people have JS turned off, while others have plugins that block GA from loading and sending data back. Server logs are the only real way to know true traffic.

First off, chicago.yoursite.com is going to be seen differently from atlanta.yoursite.com, unless they are substantially similar (i.e. they are carbon copies). Remember that this means that each subdomain will have its own SEO. yoursite.com/chicago is going to be viewed as the same site as yoursite.com/atlanta because they share the same TLD.

Unless you're going to build a ton of content around each, I would suggest one site with a subdirectory. It's less work for you. Then make each localized page unique. Take pictures, discuss the features of each location, etc. Add the address and maybe a map widget.

Different goals for the two tools.

WMT is designed to give you some idea of what Google sees. It's a way for Google to give you feedback but Google is not in the business of telling you what to do with SEO. Google wants you to offer your best search face for their sake (i.e. if you have a lot of duplicate title tags, Google has to guess what to feed the end user). They offer useful hints and that's about it. If WMT hints something, you should pay attention.

Moz, on the other hand, is an SEO tool. It's looking to tell you everything you could want to know from an SEO perspective. So Moz over-reports things just in case you want that information (i.e. how many canonicals it found). Not everything in Moz should be heeded. it's up to you to figure out what to do with the information.

Depends what you're aiming for. The largest difference between mobile and any other browsing experience is real estate. You're extremely limited on a normal mobile browser vs any other venue. In this realm, the iPad is generally not considered mobile. You tend to get the same browsing experience you would get on a full computer.

Another difference is that no iOS devices support Flash (that includes the iPad). Adobe has also announced that they are dropping Flash support for mobile devices which leaves support for Android tablets in the future in doubt.

There are a lot of reasons why a site might drop in rank. The number is higher for newer sites because a site can often rank well after launch and then become weighted down to a "real" number. Another possibility is a temporary shift in your site rank. I've seen sites inexplicably drop (i.e. no recent SEO activity) and then come right back in a weeks time.

Your site's not been de-indexed. Keep doing proper SEO and you should come back up.

I've found that many sites do this, primarily for tracking reasons. Typically they'll do the tracking and then bounce you to a clean URL.

As long as the page is on the same domain and the end URL is relevant, I can't see how it would impact quality score.

A US address would help, but only in the sense of latency. If you're hosting in India and I'm in the US, that means the data has to literally travel halfway around the world and back again. With the cheap cost of virtualization (cloud) hosting, that just doesn't make sense if you're trying to target the US market. So i would physically host in the US for that reason. If you still need to target an Indian market, I would get a ccTLD .in domain (i.e. purplecola.in) and host locally for that as well.

Beyond that, breaking into the US market should be simple. Your two things to keep in mind are

1. Proper localization. This is more crucial since you're in India. There's a tendency for Indian companies (especially in the tech support industry) to try and pretend they're American. I've had Indian phone tech support tell me their name is Bob and it's just not believable (tell me your real name or it looks like you're trying to hide something in plain sight). This goes double for your SEO content because there are LOTS of Indian companies trying to sell American content that is not really American localized. You need your content to look and read American because you don't want bad localization.
2. Once your content is good, go get some American links. This will send a strong geolocation signal that you're a US focused site. You should already have Webmaster Tools set to the US anyways.

Never underestimate the power of legacy content. Yes, those shoes are no longer for sale but what if someone wants to consign a pair and needs details? Google might favor your site and send traffic your way. Mark it out of stock, add some products they might enjoy instead and move on. I've noticed that Amazon, Newegg, etc. don't delete their old URLs. Stuff I bought years ago is still out there.

The simple answer is no. Let's say you have page A and 301 it to page B. As soon as Googlebot gets the 301 (might take a bit longer but for argument's sake we'll say it's instant) it drops page A and indexes page B. As the ranking process moves along (which is slower than indexation) most, if not all, of the PR that A had is now moved to B.

Google can't read your htaccess file (Apache will not show any page with a period as the first character in the filename for security reasons) and I've not heard of any long term penalties from changing URLs. Just realize that there is a gap (days or weeks) between the time that Google notes A moved and B is ranked like A was.

It doesn't look like that site is even indexed by Google. Even if it were, disavowing an entire site is pretty easy so a one domain spam attack wouldn't be very effective.

I wouldn't worry about it unless you experience a massive rankings drop.

The major downside to Google Analytics is that it relies solely on Javascript. In other words, your clients' browsers have to load the JS, parse it and send the appropriate data back to Analytics. While this works for most visitors, it certainly doesn't work for all.

Log parsing is always the best solution because your server, by definition, gets all the traffic and can tell you with 100% accuracy your actual traffic. Most of the log parsers don't require you to keep your logs. Instead, they parse the logs and store relevant data in a database. The downside here is that you will have to have direct access to your server logs, which not all people have.

Some open source solutions include GoAccess and AWStats. There's also Cloudflare, which is more of a proxy but it can also feed you true numbers (although with less granular data, much like GA). There are numerous paid solutions as well.

You can but you don't have to. There is a small benefit to having a ccTLD on an IP that has a reverse DNS to that country. The main reason to have a different IP, however, is different demographics. If your .com is aimed at the US market then it needs to be hosted in the US because it lowers latency. Same with your UK market, your Austrailian market, etc. Lower latency is better. That doesn't mean you HAVE to do it. You can host all of it in the US and still rank well. It just doesn't have those extra signals so it's a bit more work, but not greatly more-so than setting up these different IPs.

Link audits are always helpful (hence why Moz specializes in them) but you want to identify bad or poor links. Relevancy is something that Google will determine for itself. I mean if a tire shop is linking to a florist Google might not give that link much weight but what if the tire shop has a blog and they want to thank said florist for sending flowers after a coworker dies? Google might not give it a lot of relevance to it, but it's also not necessarily spam.

Unless you can clearly identify that a given link is actively hurting you, I wouldn't remove/disavow it. Remember that link disavow is a flamethrower in that it can kill weeds and burn your house down. Which it does depends a lot on how you use it.

I think you're right. it doesn't sound like there's a solid case here for subdomains. Consolidation would make for easier marketing as well.

The correct answer is... as many as you can afford the time to build correctly

From the Moz link building guide

There are lots and lots of ways to get links. The right tactics for you depend on the resources you have at your disposal as well as the industry that you're in. Industries that are more established and competitive often require you to be quite aggressive with link building, and you might find earning those links more difficult. Other industries, often the newer industries that are quickly growing, are full of opportunities to engage with bloggers and build a community.

But the key here is to BUILD VALUE. From Rand's recent Whiteboard Friday

I would urge you to go the opposite direction. Narrow your funnel. Worry less about the number of people you're targeting and more about the success rate, because once you get the success rate high, you can turn up the volume really fast. But if your success rate is low and there's a limited market of influencers in your field, you can quickly burn all of them with your outreach before you ever have a chance to get good at it.

The simple fact is that your question is, more or less, "How many low quality links can I safely build?" and the answer there is none. I could easily go get 80 links... and then those links would get devalued or draw a Penguin penalty. Instead, you need to build quality links, and that means you have to take time to build some value with what you're offering to them. To sum up that Whiteboard Friday "There are no shortcuts to build quality links". You have to do it low and slow. If someone is willing to give you a link for little or nothing then that link isn't work building.

Can you elaborate on your "unique but useless" (my paraphrase) content? Normally I'd say to get as much content as you can out there but if it's not useful it might drive another signal you don't want: bounce rate.

I'm not sure there's anything else to be worried about. Matt Cutts made a video and it's pretty short on the matter.

Google should recognize the difference but, just to be safe, I would add a canonical to your page so you don't have anything to worry about.

Are we talking about something focused on a different country or just a different language group? If it's a different country, I would duplicate your site and then buy the appropriate ccTLD. Matt Cutts has indicated this is the preferred method for doing international and there is not penalty for duplication. This also helps greatly with geotargeting.

If all you want to do is add a different language to the same site, I would go with a subdirectory as opposed to a query string entry. Most sites tend to gravitate towards this anyways (domain.com/en, domain.com/es, domain.com/fr, etc.). Based on what Google has said in the past, this is their preference as well.

If neither is an option, Google will still recognize the page a different based on the query string.