Rank Checking?

[Rupert]

Ah, yes.. conversion rates...   for each source.

and lifetime value of the customer....  how to get them back to spend again. It used to be so easy when I did not have to worry about that, but I have to say, I was starting to get bored.

One day Dogboy, you will have to explain that to me over a beer.   I am missing a key piece to understand what you were doing there.

[Gurtie]

>> if all you do is watch your ranks

yeah, but you'd be a lunatic if ALL you did was watch your ranks.  That doesn't meant there isn't value in watching your rankings though, even if you're only answerable to yourself.

[Drastic]

If all you do is watch your ranks, and you have no way to determine which clicks actually make you money (you do not track the click from the listing to your bank account) despite what you think, you are most likely wrong, unless you only really rank for a few big terms and nothing else.

Totally agree, I generally watch the revenue, and if that changes (pos or neg), then I look into the ranks/logs. This is after the site is humming of course, and I'd probably be better off to do both.

Market Samurai is supposed to be pretty good at tracking ranks over time, but I've not really used it so I don't know what scale it can handle.

[BoL]

OK, so a summary...

- Can't compare to leads & sales. Bottom line matters most, rank checking can be an indicator as to why income is changing.

- Ranks not indicative of conversions

- Rank checking coverage an issue for sites attracting longer tail terms going into 10s of thousands of keyword variations (unless able to stretch resources to track these too)

- Personalisation of results decrease meaning of rank results (check from more than one place? Have bots with cookie profiles?)

- Against ToS of engines

- Only useful for search engine traffic

========

- Clients like them as a 'progress metric', sparkly charts make you look busy and gives them something pretty to look at, if that's what keeps them happy

- Useful for big ticket keywords more so than long-tail

- Can check for a site or page being penalised e.g. I ranked #1 for X with my home page but now its replaced with /innerpage.htm at rank #54.

- Discover movers and shakers. See who's new on the block. Gives an opportunity to investigate backlink profiles and tactics.

- Some kind of grouping of niche/importance required for keywords for better understanding/ordering of results (See Adam C's post). Alongside categories I think using PPC * traffic is a good rough guide to keyword importance (with PPC indicative of keyword converting?)

=========

I think having rank check data and a backlink tool can have advantages, but the priority of looking and learning from this info is a little into the deck if your main concern is increasing income straight away.

That sistrix data pointed out that overstock had a wide range of ranks for terms demoted. Not only is it worth knowing if you're competing, but it's good to know who took their place and who went down with them.

[Rooftop]

Danny Sullivan tweeted a few things from SMX that I thought were relevant to this:

@dannysullivan

• 20% of Google queries not seen before in past 3 months
• 64% of Google searches have pages without exact matches to all query terms
• 44% searches on google have more than 3 words

So, chances are anything you are tracking SERP wise is only ever going to be a small part of the picture.  However I'd still rather have that part of the picture than none.  The trend it provides is useful even if the specifics are limited.

[dogboy]

Funny, but this thread bothers me a little. There are several people talking about several different scenarios and I'm not sure we are all really clearly outlining the issues, so we are all kinda talking about the same things, but we aren't precise enough in our language to really clarify our positions or shed any new light.

I can say that there are several intense learning experiences that deal with issues, like this one, that I spent a ridiculous time looking at theoretically as well as in practice.  And this one repeatedly shocked me until I got a handle on how things really work in practice.

Here is an example. You have an overall 1:8 lead:click ratio. your kw strategy revolves around a 3 word phrase that you can break down nicely into two words, and with a little effort, one word, although it's a little generic. But the one word kw brings tons of traffic, the 2 word brings alot, and then the three word phrase tapers way off. And if you have no way of connecting the end points you assume that the number of sales is somehow proportional to the amount of traffic...

You shouldn't do that. Don't assume anything here. Because you may be right. Or you may not. Which means you don't know so it's stupid to guess.

...now is where if i wasnt on my phone, and I had a bunch of time, I would come up with a bunch of examples of 3 keyword strings where I show you that you are totally right, and others where you would be totally wrong.

And then throw something out there like:

'car', 'used car', 'used red car' and ask you how you sold your car and you think you can come back to me and tell me you have a 60% chance of it being car and a 90% chance of it being either 'car' or 'used car' and then I say "wrong! It was 'Toyota Prius clearfield NY' and they found you using google shopping!  It only got one hit'

...and then you would say, 'how the he'll would I know that?'. And I would say what ever made you think you could predict anything with 90% certainty? And basically, eventhough I'm making a totally bullshit argument, you can't tell me I'm wrong. Because you don't know. And that's the problem. People are always sure they know something when in fact they don't.

[Gurtie]

but the thing is dogboy, you seem to be assuming that

1) we don't know what keyword actually resulted in the conversion
and
2) that the keyword that resulted in the conversion is the one keyword which matters

and thats not an assumption which should be made

[dogboy]

>you seem to be assuming that we don't know what keyword actually resulted in the conversion
First you will need to define 'we', as there are a few conversations here as I noted, but if you think I'm speaking to anyone in this thread in particular, I'm not....

But in many cases (and especially if you are sending money into an affiliate link) you have no idea which term actually generated the sale. This leads to watching your ranks with more intensity, but with no more data, which leads to speculation and assumptions. 

If you happen to be one of the people that control everything 'from SERP to checkbook' then you are lucky, but even lucky, many people have 3 stats programs, and log files, and can tell you where they are getting clicks, and they can show you a bank account... but the systems aren't really connected.  There's the traffic and there's the sales, but no one really knows how they are connected, just that they must be. Correlation? Causality? Both? I maintain the answer isn't one way or the other all the time. It depends on your unique situation.

>that the keyword that resulted in the conversion is the one keyword which matters
'what matters'... that's tricky. If I'm sending money through an affiliate program, I care but I'll never know.  If I am receiving traffic from an affiliate, I don't know AND I don't really care but it might be nice for my own advertising efforts.  There are also lots of companies out there that DO care and don't know. And vice-versa.  If you are an SEO that is trying to justify your job by showing ranks to your boss then it is important, especially if you aren't converting.  (it's either traffic quality or your site sucks, but at least you are 'ranking', which means at least there is opportunity for a sale.) If you are an SEO that gets paid on flat clicks, then you don't care about sales, but the company buying them, does.

...see what I mean?  'What matters' depends on your perspective.  But as far as which is more important, 'Ranking' or 'Sales' ....unless you sell rankings... 'sales' are ultimately the only thing that matters to a business.... assuming the business is in business to make money. 

[PaulH]

In one industry i work in, i look at rankings fairly often - is only part of jigsaw but is a nice guide for targeting our link building. We go after the major terms, because we can, they convert, and the long tail follows. The rankings are also used in industry reports to report on our performance which effects deals we do. So for this industry rankings are fairly important. We convert pretty well, helps the other results are mostly us too. Fortunately we can track that keyword to an exact cash value

For another sector, i don't think i've ever looked at rankings, or the traffic we get. We just do what we know works.

For another the top broad terms have so much volume our link building is just brute force and 100% focused on these terms. Ranking for big brands is the dogs So we check these rankings now and then to see how we're doing.

Tool, i would love is something that could pull in my rankings for 10,000's of terms i target - so i could better assign my link building.

For us rank checking really depends on type of site, spend, profit etc

[Adam C]

Tool, i would love is something that could pull in my rankings for 10,000's of terms i target - so i could better assign my link building.

Do you use AWR?  At what point does this break for you?

[Drastic]

• 20% of Google queries not seen before in past 3 months
• 64% of Google searches have pages without exact matches to all query terms
• 44% searches on google have more than 3 words

Great stats, but one of them really, really annoys me as a surfer.
64% of Google searches have pages without exact matches to all query terms

This really pisses me off when I'm searching. I type the terms in for a reason, and often have to quote them to get what I want. This is a huge fail IMO.

[Rooftop]

even worse now: Google sometimes insists it is right even when you put quotes around it, + it etc. 

[Cowley]

I check some of my sites and some of my competitors using an automated tool. I have a big site with 0000's of products so it's useful to see where the ranking/traffic gaps are.

[BoL]

• 20% of Google queries not seen before in past 3 months
• 64% of Google searches have pages without exact matches to all query terms
• 44% searches on google have more than 3 words

Yep those are interesting, a bit surprising at first and lack a little details?

>20% of Google queries not seen before in past 3 months

I'm guessing a lot of these are celebrities, entertainment, other names and brands. news too.

>64% of Google searches have pages without exact matches to all query terms

yep it's bugging how google assumes the words you mean. I'm not sure if this figure includes typos but it's a high number.

>44% searches on google have more than 3 words

the biggest problem with comprehensive rank checking. as said i think it depends on the niche. a q&a site will have 10's of thousands of kw variations. a mortgage site will have more predictable queries + locality searches.

[Gurtie]

even worse now: Google sometimes insists it is right even when you put quotes around it, + it etc. 

they just managed to confuse me very badly. I searched for a 2 word niche term including the word syndicate.  >50,000 results. I didn't click on anything on page 1, on page 2 every result was about the local nightclub called syndicate.  I had to click back and then forward again to get the results I actually wanted to see.

Somehow think they might be testing some click tracking variances