Please read! Internet Marketing and Better Backlinks - Different? Better? Find out why.
News and information for online web publishing professionals!
Exact Match Data or Broad? Which is better?
Look at the keyword research data on this screen capture from the Google Adwords keyword tool and then tell me the answer to this question:
“Which is better to use for analyzing potential traffic when creating your primary target keyword – “exact match” or broad match???
Open keyword research data FULL SIZE in a new window
Do you see how the broad match search volume for medicine sports is so high – yet the “exact match” is so low???
If you asked me – this keyword – medicine sports – would be a BAD primary target keyword.
If you get a top ranking for medicine sports you are NOT going to get the broad match traffic.
The exact match traffic is what that backwards keyword really gets!
I personally feel that I would rather look at the EXACT data representing the exact search as it is typed in by the searcher.
That’s how Google looks at it.
Broad match does not reflect the volume that a keywords will get with a top ranking – only EXACT MATCH accurate indicates potential volume for a top ranking.
The “exact match” data more accurately reflects what the target is capable of in volume becase it does not include all the other possibilities that borat, er, I mean broad match includes.
If you only look at broad match data it is deceiving. Even if you get #1 for medicine sports it will NOT bring you very much traffic. In fact that 170 volume is pretty low.
You don’t get rich by getting number one for low volume keywords.
Take the above example and compare the Global Search Volume for
medicine sports
which has a broad match volume of 368,000 which seems very high.
If you change the Google Adwords setting to “exact match” however the gloabl monthly search volume drops to 170 – yikes!
And for me – 170 just isn’t enough to justify making it my primary target keyword.
The reason it dropped to 170 from 368,000 is that the PHRASE is backwards. No one searches for medicine sports – they’d search for sports medicine maybe – but even then – the BROAD MATCH data ALSO includes words BEFORE – AFTER – IN BETWEEN the keywords! You’d have to target EVERY POSSIBLE VARIATION which is next to impossible and would require too much time!
If you target the wrong BACKWARDS keyword – it will be more challenging to work it into articles because no one ever really says medicine sports naturally.
If you look at exact data you see that – 170 is actual usage! No one talks like that or writes like that – so why target a keyword like that?
I do not! You shouldn’t either.
Target keywords that make sense when you read them out loud!
Do not target keywords that are backwards or shuffled in different order!
It’s harder all around and the payout wil be less in the long run than if you qualified your target keyword with more accuarte search volume data such as that from “exact match”.
That’s why BROAD MATCH data is not good for picking your primary keyword targets.
It might be OK to look at to get an idea of ALL the possible searches including the keyword – but the truth is – we can’t adequately target that many keywords anyways.
I believe it’s much more effective to find several “exact match” keywords with more realistic volume data to target rather than go after broad data that is pie in the sky and unattainable – so why even look at it?
It’s deceptive.
If you want LOTS of traffic you need to focus your initial efforts on a PRIMARY Keyword Target which gets a good volume as shown in “exact match” data.
I say 200 minimum per month “Exact Match” Global Volume for beginners. If you have experience getting top rankings then over 1,000 ”exact match” at the low end for intermediate level web publishers.
Otherwise if you target under 200 exact match volume you’ll be building pages for keywords that won’t bring much traffic.
So – start checking that EXACT MATCH box when you do your keyword research! You’ll have an advantage over those who are fooling themselves with the BROAD MATCH data!
Now – if you find this interesting you should really go check out my full article on this at the Blogging Undegrround forum.
The post is called More Keyword Targets are Better!
Tags: Adwords, Google, google adwords, Keyword Research, top ranking, Traffic70 Comments »
RSS feed for comments on this post.
You make no mention of “Phrase match”. I tend to think the phrase match is the most useful indicator for a keyword to target.
Comment by Graeme Eastman — May 31, 2011 @ 1:09 am
Mike,
Great Post. Your exactly right, using anything but exact match is not only misleading it will cost you money.
Working to get a #1 Google rank for a keyword with low traffic and very low Commercial Value is to be avoided.
Thanks for the valuable info.
John
Comment by John Robberson — May 31, 2011 @ 1:16 am
Great post Mike and have to confirm that I only recently learned this lesson, to my cost. The advice on keywords is different from many product promoters but since doing SEO for companies, I have found this to be the very best method of ranking.
Above from an agreed reader and thank
Comment by AnneMarie Callan — May 31, 2011 @ 1:25 am
Very clear explanation Mike. This post will help a lot of people from targeting the wrong keyword. Unfortunately I had to learnb this lesson the hard way. I targeted a KW which had good ‘broad’ searches but didn’t pau any attention on the ‘exact’ count, which was less than 200 by the way. Short story, I got the website ranked on the first page out of 92M results but the site was a flop. Had to let it go eventually.
Comment by Carlo — May 31, 2011 @ 1:40 am
I also use just “Exact” match for keyword research, and I even find that on a lot of occasions those search volumes fail to materialise when I get the top rankings …
Comment by Richard — May 31, 2011 @ 1:45 am
Hi Mike,
Good article and very true. Just to add something about the competition analysis side of keyword research.
A lot of people use allintitle:”keyword(s)” as their absolute measure for going for a keyword or not. It’s actually quite vague. Sometimes you will find that a keyword with a high allintitle / url count etc will have quite a low average Page Rank of the top 3 positions, and it is those who are your true competitors, not the 100,000′s of other non ranked pages. SEO Quake is excellent for determining the competition strength, ie backlinks and Page Rank.
Thanks for the good info,
Andy
Comment by Andy — May 31, 2011 @ 2:01 am
Hi Mike, thanks for this. I have earlier used the broad match because the figures seem tempting. I initially thought that 100 should be a good number to start with and now you have just demystified the issue. I will give this a try.
Comment by Reuben — May 31, 2011 @ 2:13 am
Very informative post, but as Graeme Eastman has mention in his comment, you did not mention anything about the phrase match. I have used the phrase match for one of my sites and got second spot on page one out of 20 million sites. I must admit though even though it’s on page one I’m not getting as much traffic as I expected.
Thanks for the great tips Mike
Comment by Arnold — May 31, 2011 @ 2:17 am
Hi Mike,
So true – especially about the backwards phrases. The thinking was “well, all the words are there so it must work”. I have a few domains like that and to be honest, the traffic is not great even though I add original content all the time. I just checked for one of my domains and it is below 200 for exact match – what a dum dum. Oh well, we all live and learn.
Thank you for the great info.
Comment by Michele — May 31, 2011 @ 2:44 am
Hey Mike,
Thanks for that info, yeah have been researching broad and thinking looks okay but will definitely dig into the exact match more.
Cheers!
Comment by Rich Muir copywriter — May 31, 2011 @ 3:11 am
Thanks Mike for the interesting post. I am always experimenting weith keywords and this really helped put things in perspective. Much appreciated and have a super day!
Comment by Linda — May 31, 2011 @ 3:30 am
Hi Mike,
some useful tips in your post there
what I use some times is have the core 5 – 10 exact match keywords (KWs) for which I’d create the 5 – 10 unique articles targeting those individual KWs.
Then in my tags, article marketing, social marketing I’d use the broad match KWs including those exact KWs. This has resulted in more traffic over all as search engines start sending traffic for related KWs too.
I think they’re times when either type of KW should be used as long as one has as understanding of what the true worth is for a KW in as far as traffic generation or convertion is concern.
Thanks for a great post
Comment by Susan - PLR Niche Blogs — May 31, 2011 @ 3:58 am
why would you even go after 200 exact match a month how is it possible to monetize such low traffic
Comment by Arthur — May 31, 2011 @ 4:20 am
Hey Mike,
Thanks for the information.
But, as someone pointed out, I think phrase match is also a good indicator for determining the correct level of traffic.
What is your take on this?
Cheers
Raj
Comment by V K Rajagopalan — May 31, 2011 @ 4:24 am
Hello Mike,
Thanks for the informative approach to keyword research. I just wanted to add one point.
It depends on the intent of the keyword. If you are an affiliate marketer selling a product and Adsense is not your primary method of making money…then very low volume keywords with definate buyer intent can convert much better than the larger “research” type words that get more traffic.
I usually target words with trickles of traffic…but I have a pretty good idea that the researcher has their credit card in their hand when looking them up…rather than someone that is just looking around.
I find many of the larger volume words are just researchers or lookers…vs…buyers.
Comment by Deb@Wealthy Affiliate — May 31, 2011 @ 4:42 am
Hi Mike,
I wish I had read or learnt great information like this before I started!! This article has made me take a look at a few of my keywords…let’s just say it explains quite a few things! I have a lot of homework ahead of me!
Comment by Jennifer — May 31, 2011 @ 4:45 am
Thanks! Wish I would have known about wrong order keywords before because I recently did it. Built a new page, did a promotion and all that for http://all-about-anti-aging.com/anti-cream-wrinkle
OOPS! My new pages will be correct now that you schooled me.
Comment by Randall — May 31, 2011 @ 6:22 am
Hi Mike,
An excellent article! It explains why you should target exact match really well, I have always had trouble explaining it well myself. For me the most important idea here is that the keyword should be readily understandable english as that’s the way most people type in their searches. And as you said it’s easier to fit into an article.
Comment by Steve Dolan — May 31, 2011 @ 7:23 am
Hi Mike,
Well I must admit this is certainly an eye opener. I read lots about it, but I guess it is the way you put it across.
Thanks for sharing and looking forward for you webinar coming up.
Thanks for sharing
Comment by Cindy Williams — May 31, 2011 @ 7:59 am
Very interesting article. I am using exact match myself but I will not go after any keyword with less than 1,000 searches per month under exact match. To me it’s not worth it.
Thanks for the info in this article Mike.
Comment by Daniel K. Voyles — May 31, 2011 @ 8:17 am
Market Samurai has a setting where I can get rid of keywords that are backwards so I seldom see them. As you say, they are a waste of time.
When I’m using Market Samurai I set it to look for exact match first. Then I eliminate all of those KWs with to little traffic. That leaves the best of the exact match KWs. Then I switch to broad match and look for the highest amount of traffic and the highest cost per click.
While I may be content with 20 per day on exact match I want over 500 for broad match. If there is no traffic to ramp up to I don’t see the point. When there’s not going to be enough traffic in the end there’s no point in going after the KW.
Comment by Mster Mind — May 31, 2011 @ 9:52 am
Regarding “Phrase match” – it’s better than broad match – but the data includes keywords before and after the phrase. “exact match” does not include the data before ad after so the volume data is more accurate.
Comment by mike — May 31, 2011 @ 11:02 am
Hi Andy
Personally I rarely look at competition data such as allintitle: because all it indicates is HOW MANY pages there are with the phrase.
I know that with what I do for SEO I only need to worry about the top 10 or 20 – so I would rather EYEBALL what’s on the page and then make a decision based on my GUT FEELING
It’s far more accurate than any formulas for competition or profitiability etc.
Comment by mike — May 31, 2011 @ 11:04 am
Good Article again Mike, i use Keyword sniper pro 2 which AUTOMATICALLY displays the relavent EXACT MATCH you were talking about wich is so handy.
Comment by Den — May 31, 2011 @ 11:29 am
I usually do not – that’s my cut off point for consideration – lower volume is betterfor beginners or those who want to get atop ranking fast and easy.
And you are missing a big part of the equation – the RELATED keywords. You don’t build a site for ONE keyword and that’s it. It’s just a starting point.
While I agree 200 is a low number – you can make it work if your site has other keywords that get decent volume.
In the end your goal is hitting CRITICAL MASS where you will get traffic for 100′s and 1,000′s of keywords – long tail keywords – it’s your reward for building out your site!
Comment by mike — May 31, 2011 @ 11:51 am
Hi Deb
The wonderful thing about web publishing is you can make money doing this stuff in so many different ways.
I agree that keywords that have intent to buy are very good ones to target! Especially when the exact match data shows they get some traffic (ideally 200 or more) – but in the end you must get some real targeted relevant traffic to make money!
Comment by mike — May 31, 2011 @ 11:58 am
Hi Steve
To be honest – it wasn’t until I took some PPC training a few years ago that finally I got a good understanding of the match types and realized that broad and phrase match were not 100% accurate for the volume.
Most people do NOT understand the difference and that’s good for Google and their Adwords advertising revenue because they can get people to spend more money with them.
After all – if the advertisers saw a keyword had little or NO traffic would they run ads for it???
Remember – the match types used in the keyword research from Google are intended to indicate voulme available for PPC advertisers!
Most SEO training never mentions the different “MATCH TYPES” or they don’t explain it.
When I finally had that AHA moment everything changed for me.
As I looked for new keywords to target I was now able to use my mental filter to skip past the majority of garbage that is available and focus on the ones that have real value!
Readability is important and accurate volume data is critical!
Otherwise you’re just shooting in the dark!
Comment by mike — May 31, 2011 @ 12:07 pm
Hi MasterMind
That’s a smart feature! Is it accurate? How do they know which ones are backwards?
Comment by mike — May 31, 2011 @ 12:30 pm
I’ll start with exact match to get a realistic picture of the potential of a particular keyword phrase, then phrase match to see what additional long tail variations might yield.
Broad match is completely useless to gauge potential traffic, unless your site is of such high authority that it can rank just by virtue of having certain keywords scattered on the page. Besides which, the majority of the extra searches are for untargeted, quite often irrelevant phrases – - if you did somehow manage to get all that broad match traffic to your page, 90% would immediately hit the back button and Google would think your site is worthless.
Comment by Garth - web marketing consultant — May 31, 2011 @ 12:50 pm
Hi Mike, excellent explanation of how to do your targeted keyword research. As keyword research is probably the single most important aspect of on page search engine optimization, it is vital to do it correctly. It’s no good optimizing for “wrong” keywords as it will not generate targeted traffic.
Mick
Comment by Mick Robinson — May 31, 2011 @ 12:56 pm
Way to shed light on the situation.
(1) Beyond the fact that exact match snuffs out those bad keyword apples; (2) beyond the fact that those backwards keywords not making any sense — just think about WHO those visitors are that enter those backwards keywords. Is that exact visitor profile that you are looking for to fulfill?
I generally advise against even two (even three) word combinations, but I think it’s wholeheartedly healthy to make sure your long tail keywords DO involve and include your MAIN KEYWORD phrase that you will eventually rank for once you build a nicely layered link portfolio.
Just think of how many people out there are ignorantly using the Google Keyword tool and thinking pie in the sky only to be quite deflated with the results by not checking exact phrase.
The other thing to realize is that while Google may be 75-80% of all Global keyword searches, there is another 20-25% of searches taking place on Yahoo/Bing and other search engines.
The other shortsightedness with the Google keyword tool is that you cannot extract keyword questions and super REALLY long tail keyword phrases like you can with the WordTracker database.
That’s why any smart keyword researcher uses multiple databases — among those being competitor keyword research tools like SEM Rush — the same folks that put out SEO Quake mentioned above (a great Firefox add-on/plugin.
Although you can imagine that I am quite biased about by my own keyword analysis tool and it’s humble results:
http://nbx.s3.amazonaws.com/fixes/analysis-x-attributes.jpg
Peace to all,
Jim
Comment by Jim Morris — May 31, 2011 @ 4:36 pm
I always use exact match when I target a keyword, that way you know what you are going after in terms of traffic.
Comment by Brian — June 1, 2011 @ 7:48 am
very nice information, would come again and love it
Comment by best stocks for 2011 — June 1, 2011 @ 6:44 pm
very good info to have when planning your blueprint for keyword matches. I’m new to the internet marketing world, but receiving info such as this I’m beginning to understand more of the structure of internet marketing.
thanks for the info,
Rich Austin
Comment by Rich Austin — June 1, 2011 @ 11:58 pm
Thanks! Wish I would have known about wrong order keywords
Comment by hanfeng evergreen — June 2, 2011 @ 8:17 pm
Once again great information Mike!
Thanks for explaining so clearly a really important subject.
Beth.
Comment by Beth @ EarlyLupusSymptoms.com — June 3, 2011 @ 11:05 am
I have followed your work for a long time.
You was the first one who made me realize how important keywords are to the internet business owner.
Anyway, to your article.
I have found the same thing.
Keywords can have a huge broad match but a pitiful exact match.
The example you gave is extreme but certainly 170 searches a month is an, “I’ll pass” on that keyword phrase.
Comment by Gary J Martin — June 7, 2011 @ 10:22 pm
Love your stuff will be a member for sure feel like you your members and your team really get it done thanks.
Comment by Arik — June 9, 2011 @ 10:29 am
Simply awesome tools! Thanks for letting about this tool here!
–
Call Centres
Comment by Mike — June 11, 2011 @ 2:59 pm
Mike,
great post. I agree with you, lots of people when searching online, they type words exactly as they speak in the navigation bar. It’s another example why exact word search is better than broad search.
Comment by Gregory — June 12, 2011 @ 9:13 am
Mike,
Great Post. Your exactly right, using anything but exact match is not only misleading it will cost you money.
thanks for sharing nice post
Comment by robit — June 13, 2011 @ 5:31 am
Great article. Back when I had just done my first website, I had read/watched some videos by a well known internet marketer who would always use the broad match. So I followed his lead and chose my keyword domain that way. Afterwards after watching many other internet marketers use the exact match only, I realized that my keyword domain was only getting less than a 100 monthly in the exact match. I learned this the hard way I guess, but now I only do my keyword research in exact match. I just don’t understand why so many that teach, make it a point to use broad match. I guess they don’t want to reveal the real secret….
Comment by blog names — June 13, 2011 @ 5:06 pm
I have spent the last few hours rechecking some of my keywords…”Yikes” is right!
I have looked at broad match data for way too long.
Thanks Mike for explaining something that is right in our face, yet obviously so overlooked.
Comment by Charles — June 14, 2011 @ 8:46 pm
Thanks for the helpful words!
I took a look back over some of my keywords and was able to take a new angle on how I’m approacing my campaigns.
Comment by Nicole Locklear — June 23, 2011 @ 9:48 am
omg i can’t stress how important this post is. Alot of people teach broad or phrase and it drives me crazy and i have been so stupid in the past following along and have wasted so much time and money on kewyword domains that get hardly any traffic.
I have since changed and did a quick seach to get some confirmation on this to find this post. seriously this is an awesome post. Short and sweet but could save others heaps of money and time.
Comment by Sherone@Nimbus RO — June 25, 2011 @ 7:17 am
It sounds like if you use a broad keyword though, you may have too many competitors, is it not worth the trade-off?
Comment by Nicole Locklear — June 27, 2011 @ 2:52 pm
Instead of short keywords, long tail key phrases will be better, it will serve for high quality relevant traffic, low cost as well as it will organically bring your website forward towards the niche keyword.
Comment by BookMarking — July 5, 2011 @ 6:40 am
Hi Mike,
I have just finished posting short blogs to my BU. User friendly and easy site to use BTW.
I personally look at exact match as well as broad to get a better understanding of the market.
Regards
Catherine
Comment by Catherine @ Affiliate Marketers College — July 5, 2011 @ 10:34 pm
I have just finished posting short blogs to my BU. User friendly and easy site to use BTW.
Comment by شات الخيال — July 15, 2011 @ 2:28 am
I was really bad at keyword research for a long time for this exact reason. I think I knew that my terms weren’t going to get much traffic, but I liked seeing the volume of the broad match. I don’t even bother with broad match anymore.
I guess that’s why I struggled for so long. Now I pretty much know what I’m gonna get if I put in the hard work to rank for a term!
Comment by Ryan - Online Profits — July 15, 2011 @ 10:31 pm
SEO Quake is excellent for determining the competition strength, ie backlinks and Page Rank. It’s another example why exact word search is better than broad search.
Comment by text message marketing — July 19, 2011 @ 12:01 am
Ah wow, thanks for the clarification Mike!
I’ve always been looking at “Broad Match” during keyword research, and after following your advice, it’s apparent that I’ve been targeting the wrong keywords… (Doh!)
Thanks again!
- James Francis
Comment by James Francis — July 19, 2011 @ 7:49 am
Hi, Mike
Thanks for giving this information to me.
Comment by MarkDMC — July 25, 2011 @ 11:00 pm
I would guess “exact” would be the one.
Len
Comment by Len — July 30, 2011 @ 2:58 am
I use only the exact match keywords for the analyzing because it gives the best estimation of the amount of the revenue that can be generated by getting a certain amount of traffic.
Comment by free online fax — July 30, 2011 @ 12:01 pm
Thanks for sharing these useful tips
Comment by silverlight developer — August 8, 2011 @ 9:39 pm
I agree Mike, and since I switched from looking at broad to exact matches, I’ve seen more success with my efforts. In certain situations I’ll look at phrase, but rarely at broad.
Comment by Carl Weber — August 10, 2011 @ 7:28 pm
I’ve been using broad matches when looking for products to promote. I’ll use the exact from now on to see if it is really a product I want to promote or not. Thanks for the info.
Brenda S.
Comment by Brenda — August 12, 2011 @ 12:23 pm
I avoid broad match in all of my keyword research. Although it is interesting to see the potential for a given keyword I only want to target exact keywords that I know will bring me substantial traffic.
Comment by Mom Freebies — August 18, 2011 @ 12:07 pm
you say,
“Personally I rarely look at competition data such as allintitle: because all it indicates is HOW MANY pages there are with the phrase.
“I know that with what I do for SEO I only need to worry about the top 10 or 20 – so I would rather EYEBALL what’s on the page and then make a decision based on my GUT FEELING”
Can you explain this more? E.g. you are examining a keyword “cheap calls to India”. You get heavy duty calling card sites on the first page, such as Pingo, callingcards.com, raza, however these are not targeted to this keyword. How do you decide if it’s worth competing?
Comment by John Dime — October 18, 2011 @ 11:16 pm
Determining which one to use, either exact match or broad match, is very critical. Making a wrong decision on this could mean wasted time and effort on the campaign. Thanks for giving light on the subject. Valuable stuff here indeed.
Comment by prostate cancer pittsburgh — October 21, 2011 @ 4:43 pm
I’ve always thought that as long as there are keywords in it, they are all equally effective. This article will be really helpful to those who are starting up and to those who hasn’t known that exact match could turn out to be misleading.
Comment by Hendersonville real estate — October 22, 2011 @ 4:17 am
Getting into medical school is no easy task but the process has become a business. The price of medical education has increased steadily and admission has become more difficult. Being a doctor is an amazing thing but people are often times entering into programs that are based outside of the states and these programs fail to prepare students for what residency programs have in store for them. Unfortunately Politics has entered the arena and the practice of medicine isn’t what it use to be. Some of the most capable people in our society are leaving school jaded with huge loans and are subsequently significantly underpaid. The average graduate owes hundreds of thousands of dollars and is required to spend anywhere from 3-8 additional years in residency while making approximately 40K-50K all while paying heavy loans amounts back. All in all its an amazing life choice but the costs are heavy.
Comment by Buck Davern — October 27, 2011 @ 7:37 pm
Very clear explanation Mike. This post will help a lot of people from targeting the wrong keyword. Unfortunately I had to learnb this lesson the hard way. I targeted a KW which had good ‘broad’ searches but didn’t pau any attention on the ‘exact’ count, which was less than 200 by the way. Short story, I got the website ranked on the first page out of 92M results but the site was a flop. Had to let it go eventually.
Comment by Eimis — November 23, 2011 @ 3:03 pm
First of all clap for Mike!
Thanks Mike you shared an important aspect of Keyword Research.
Secondly I’d agree with ‘John Dime’, as he mentioned in his/her post quoting your ones “Personally I rarely look at competition data such as allintitle: because all it indicates is HOW MANY pages there are with the phrase.
“I know that with what I do for SEO I only need to worry about the top 10 or 20 – so I would rather EYEBALL what’s on the page and then make a decision based on my GUT FEELING”
I also want to know more about this from you Mike, please elaborate this for us.
And what I know about “allintitle:” means Google will show you those results which are using that particular keywords (i.e allintitle:cheap shoes) in their PAGE’S TITLE and not in PAGE or Content.
Thanks
Rizwan
Comment by Rizwan Maqsood — December 18, 2011 @ 1:49 pm
Figures reveal that nearly 80% of all serious site owners need linkbuilding solutions to launch his or her organization for you to better height. With all the improving number of …seo
Comment by Seo url — January 14, 2012 @ 12:52 am
Great article. Back when I had just done my first website, I had read and watched some videos by a well known internet marketer who would always use the broad match.
Comment by Natural Stone Jewelry — February 5, 2012 @ 5:28 am
A few years ago, I bought a clickbank product that taught me the wrong way. Lesson learned after I tried to build my first site they way they instructed. I use a combinations of exact and phrase, twitter searches to see how much interest there is in a particular keyword and market samuari. I tired Micro Niche Finder but found it to be inaccurate. Great info, thanks.
Comment by Frugalvoice — March 30, 2012 @ 8:18 pm
Broad match just gives you an unrealistic vision of the niche you want to target. Instead target multiple exact match keywords that you deem high value.
Comment by Baby Coupons — October 9, 2012 @ 8:12 am
I use exact match almost exclusively. Every once in a while I’ll switch to broad just to see the overall scope of the category I’m researching.
Comment by Baby Coupons — October 18, 2012 @ 2:57 pm