Can the number of Adwords help you discover a profitable niche? 15 April 2009
The late Gary Halbert used to ask the following question of his students at his copywriting seminars:
"If you and I both owned a hamburger stand and we were in a contest to see who could sell the most hamburgers, what advantages would you most like to have on your side to help you win?"
Answers from his students included: getting superior meat to make the hamburgers, a superior location, or even having the lowest price.
Once the students had exhausted themselves, Gary would offer them all the advantages they asked for if he could have just one--
When it comes to marketing, the most profitable habit you can adopt is to seek out groups of people who are hungry for a product or service; they are probably going to spend money to get it and if you supply it, you'll get some of that money.
Keyword research is a great way to discover that starving crowd; the more relevant keyword statistics you have, the better you'll be able to judge just how hungry and profitable the marketplace is. And to get a niche market to reveal how profitable it may be, you need answers to three fundamental questions:
- Are there enough hungry people_ easily reached?
- Are the people willing to spend money to satiate their hunger? Are there enough suppliers spending money to reach and feed the hungry people?
Let's look at how keyword research can help you get answers to these questions:
1. Are there enough hungry people easily reached?
The aim is to find enough people looking for products and services to fill their need; but not so many that they attract too much competition, pricing you out of the market.
One way to research market demand online is to use a keyword research tool like Wordtracker:
According to Wordtracker, there are roughly 3,685 searches every day by people who want to learn to train their dogs. Now that's a starving crowd!
2. Are the people willing to spend money to satiate their hunger?
Simply put: does your market have money to spend? Do potential customers have credit cards and/or disposable income and are they currently buying products and services in the niche you wish to target?
Students, for example, don't like to spend money. Nor, do I suspect, do people who search for 'debt reduction'.
It's worth noting that if advertisers aren't spending much money in a market, or there are too few of them, it could mean that the market does not spend much money either.
3. Are there enough suppliers spending money to reach and feed the hungry people?
Google is one place where buyers and sellers meet online. But it's not until a potential customer types in a keyword (also known as a search term) and clicks 'Search', that Google auctions off eleven positions on its results page; the sellers then all bid for that click.
When all eleven positions are filled with advertisers, this usually means there's plenty of competition for that keyword. A high number of ads also usually means a higher cost per click; the cost per click for the keyword 'dog training' in the above example is currently a whopping $3.22.
If we compare 'dog training' with a keyword that has just one advertiser e.g. 'dog training types', we can see the cost per click is now just $0.05:
More competitors usually means a higher paying niche, where customers are prepared to pay more to satisfy their hunger; this pushes up the cost per click as advertisers spend more of their resources trying to attract the attention of the searcher. As a marketer this is good to know. If there are many advertisers battling for the click then there must be money in the market--advertisers are not in the business of losing money!
Don't be afraid of competition--it's actually a good thing--it means many people are making money in that particular market. Think of when you find several Chinese restaurants in a row; the area gets known as the place to eat Chinese and attracts more customers. Every restaurant benefits.
The time to worry is when you don't find any advertisers. Don't celebrate when you discover a market with no competition; you're unlikely the first to try to enter the market for the single keyword 'cats' (two advertisers at time of writing). There are few advertisers because there's probably not much money to be made. So don't you then create a product to fill a need you've only guessed at--you'll likely go broke. Just remember that old Wild West saying: "the pioneers are the ones with arrows in their backs."
One way to discover Gary Halbert's "starving crowd" is by looking for a large number of competitors who supply that crowd. The more competition, the more likely that buyers exist in that market. But we at Wordtracker want to help you discover profitable niche markets, not necessarily the most competitive! This allows you, the marketer, to make an important choice: do you compete in a competitive yet profitable niche with many advertisers; or dominate a smaller and potentially less profitable niche with fewer advertisers?
According to AdGooRoo's Search Engine Advertising Update -Q109, approximately 42 percent of Google keywords contain no ads throughout the course of a given month; 13 percent have only one ad; 15 percent have ten or more ads. That should give you pause for thought. Those 15 percent of keywords are likely the largest online markets with the most competition; but that still leaves 30 percent of keywords with between two to nine ads--a hungry yet not starving crowd. But most likely still profitable.
So how can Wordtracker help you discover these profitable markets?
Well, for starters, we aim to count and display the number of advertisers for each keyword like this:
Notice how the adwords column reports both the number of adwords for each keyword in the list, and the total number of adwords (in this case 556), for all 1000 keywords:
(Note: This is a design mockup. The number of adwords feature will be implemented pending feedback.)
It's interesting to see that 'dotties weight loss zone' has no competition at all. An opportunity perhaps? (Note: That's zero advertisers in a niche that has already proven to be profitable online).
When many keywords in a competitive niche like 'weight loss' all report 11 advertisers, you can dig for more information by clicking on "More Sponsored Links" at the bottom of the search results page:
If you click on this link, you see the total number of advertisers bidding for the words 'weight' AND 'loss' in any order is 5,317:
We might also compare the total number of advertisers for broad match advertisers (who bid on the words "weight" "loss" in any order) and exact match advertisers (who bid on the words "weight loss" in that exact order), like this:
Do you see the keyword 'buy weight loss'? There are 139 broad match advertisers but just 25 exact match advertisers. That gives a ratio of broad advertisers/exact advertisers of 5.56. Do you find that a useful number? Please tell me.
Here are some other issues we'll take into account when reporting the number of adwords for each keyword:
- If you search the Wordtracker US keyword database then we will count adwords ads for the US only; similarly if you select the UK database, then we will count adwords ads for the UK only.
- We will count the ads when most relevant: for the US, this is during office working hours 9am - 8pm (to cover East & West coast). This is the time when you expect the maximum number of advertisers have their campaigns running and the time most buyers traditionally buy.
So now it's your turn.
Has this post been helpful? Do you find reporting the number of adwords for each keyword a useful feature? Do you agree that a large number of advertisers suggests a starving crowd and likely a profitable niche?
Please let me know your thoughts in a comment below (even if you disagree).
I really appreciate your feedback.
As always,
-Mike Mindel
Technorati Tags: number of adwords, profitable niche, gary halbert, hungry people, mike mindel, starving crowd, suppliers spending money, research market demand online, dog training research, wordtracker, profitable markets, dotties weight loss zone, weight loss, broad advertisers, exact advertisers, ppc, pay per click
32 comments
Very interesting and useful post. I were just finding these kind of posts to know the right process to find out useful and step by step information about Profitable Market Research and Profitable Keyword Research. But I do not wish to pay for keyword research. If you were used traffic travis for Adwords Ad Count rather than Wordtracker's Paid service, It could be more informative to me.
The climax of your post is not satisfactory and At the last I could not satisfied 100% because you have ended your discussion with a question mark for me. But thanx. Its a great post. I will cover up missed things with other websites of blog.
Another feature that would be cool to pull in would be click price. Particularly because then if you multiplied click price by search volume that would create an exceptionally strong signal of relative value of the keywords.
The dog training example above kills me...someone recently bought dogtraining.net for a great price, and the above example only furthers my internal discouragement for not having purchased that domain name ;)
This is a great tool to have. But in some niches it is erroneous to assume that most buyers traditionally buy between the hours of 9am and 8pm. One example is dating, where most of the crowd becomes active after the traditional so-called "prime hours".
Hi Mike,
This is a great article and an even better tutorial on what Internet marketers should be doing. Any tool that automates the process is in my opinion indispensable. You asked for input on several points and I will humbly interject my thoughts on some of them.
You asked if the broard advertisers/exact match ratio is useful. I declare that it is. That ratio actually helps me gauge what opportunities exist within a specific market segment. Please add that feature as soon as practical!
I agree with David that some buyers don't buy during normal business hours, but I also understand that you have to pick a time to actually count. My question to you is whether or not the counts will be performed in a live mode or is it possible to lag that count by 24 hours? In other words, report the highest adword activity for a specific phrase within the last 24 hours, instead of limiting it to normal business hours.
I do believe that Google adwords are here to stay and can be representative of an active market, so please add that feature soon. I like your analogy comparing active markets to that of restaurant rows. (places where large quantities of restaurants accumulate). I would suggest that while this analogy works with restaurants, it fails miserably with shopping malls. Yes, more ads in a specific market might indicate that more advertisers are vying for a piece of the pie, but whether or not that market is profitable remains to be seen. I know of several active industries where there tends to be a lot of advertising but little money flowing. So I would suggest that it can be used as an indicator for an active market, but not necessarily a profitable one.
Keep up the good work!
Best regards,
Joey
Interesting article but I think there needs to be more flexibility on choosing the reference times AND dates. In many hobby industries and sites that cater to kids, sales are made primarily outside of business hours - on weekends and from home. In addition, seasonality really needs to be considered. If I were looking for a niche biz to offset my spring/summer busy time, I might want to see which terms were most searched for in the last holiday season.
Also, it would be nice to overlay the adwords info with the rate of change on a keyword...in other words, is there a keyword that is rising in popularity but there is a lag on the adwords competition because companies have not yet identified that market.
Thanks for sharing. Never thought of the actual number of ads means more real estate or Google, hence more sales are higher competition.
As someone in the dog training niche myself, and a user of the Google External Keyword tool, I think knowing the number of Adword ads on a keyword phrase is interesting information but agree with the previous comment that a column showing cost per click would be cool.
Of course that cost per click is to appear on Google page #1, #2 etc. and is not to be confused with what you might hope to achieve if your were building a site to monetize with Adsense. One reason why I have reduced my Adsense profile on my dog training site and would rather chase those affiliate commissions.
WOW thanks for this keyword article. I learned lots reading this.
Thanks everyone for the great comments - just a couple of points here:
@Joey Lowe - we’re still looking at what will be the best/ most effective metrics to include in the finished version of the new UI, and over time (and with useful feedback like yours) we’ll add the most useful - keep watching.
@Aaron Wall - CPC would be a great metric to have available, but as I’m sure you’re aware, the CPC is variable according to a number of factors including ad relevance and landing page relevance - have you seen the Hal Varian video on this? You’re right, though, an average CPC would be useful.
Yeah, maybe there are other tools that will display adword ads but having it conveniently included within your initial research as an overview before digging further is a great time saver.
I also strongly agree with Aaron's post about including the click price if possible. The number of ads is good but knowing the money being spent is ideal.
But a third useful element would also be how long each ad has run. The longer an advertiser runs an ad an assumption could be made about the success rate. This would be particularly helpful with keywords displaying fewer ads. (Hungry but perhaps not starving crowds.)
Either way, my vote is to add it for what it's worth. Thanks.
@maldarwen, @aaron - even so, whatever the CPC happens to be just the fact that people are paying it is a good indicator. This may not be what you will pay but still a good indicator for research which is what we are really talking about.
Great ideas! Here are my comments on your plans:
(1) I really like the option of comparing the number of broad and exact match advertisers. You could even calculate those two numbers as a ratio, and use the ratio as an indicator of how sophisticated the niche's advertisers are. (A ratio of "1" would mean that all advertisers are using both broad and exact match, suggesting a sophisticated group of advertisers; a ratio of 0.25 would mean that only 1/4 of advertisers are using exact match, suggesting a largely unsophisticated group of advertisers).
(2) One thing you didn't mention about the ad-tracking feature: over what period of time will you track ads? In other words, if I search for "weight loss" and see that there are 13 advertisers, does that mean there are currently 13 advertisers? If yes, it would be even more helpful for WordTracker to track the number of advertisers over time for each keyword. I think a more accurate indicator of a profitable market is a persistently-high number of advertisers.
(3) I agree with Aaron that click price would be a killer additional feature.
(4) In my experience, the number of advertisers in a niche isn't a strong predictor of how lucrative that market is (nor is CPC, for that matter). I'd put the strength of the correlation at maybe 0.3
I research information gaps in niches (in other words, what information are people willing to pay for that isn't easily available on the web), and have found markets with $0.08 CPC and only one or two advertisers, but markets that are nonetheless very hungry.
I have noticed a decent level of activity for my products from the UK and Singapore, for instance. It would be great to be able to compare results with US searches. I am also interested in finding out why the rest of West. Europe is not showing up in my shopping cart, so would like to get results for more than one "region" e.g. US, UK, Germany, Italy, in the same chart ...
Thanks for the skilled information. I have a long ways to go to become a keyword maven!
How about reversing the search? Instead of trying to guess what might be a good keyword, maybe I could search for keywords that fit these criteria: keyword length 3 or more searches in excess of 1,000 Exact match competitors of less than 50,000 Broad match competitors of less than 750,000 plus additional possible criteria
????
Good post, thank you. I just want to add that Mark Bulleit's request is a good one, and I want to second it. Knowing how long these ads have run is very useful information. If a keyword keeps attracting new advertisers but loses the old ones, chances are, it's not a money-maker. But if several advertisers have each been running AdWords for the same keyword for a few months or more, then we know it's a winner. Great idea, Mark!
This is an interesting display that you have shown, but you've left out "phrase match" as a choice to compare versus "broad match" or "exact match" on Google searches. Google explains the differences in these search terms here:
http://adwords.google.com/support/bin/answer.py?hl=en&answer=6100
In order to increase their revenues, Google Ad Words' default setting is "Broad Match" which will quite likely increase the number of clicks that an ad receives using that setting. Of course, this will burn through the pay per click budget faster, which is what Google "likes". Without intervention by the customer or the account manager, ALL keywords in Google are set to "Broad Match". Unfortunately "broad match" may not be "relevant" to the search phrase.
Routinely I use "phrase match" for my clients as this setting is now looking for the phrase being searched for to include search terms that have other word(s) either before or after the "phrase" and seems to work well for my clients. I suspect that the more relevant ratio would be the comparison between advertisers using "phrase match" versus "exact match" and if there are any benefits to shift to "exact match" from "phrase match" on a given keyword search term. You might consider a drop down that allows the Wordtracker user to change the comparison that you've illustrated.
The current Google cost for click information for positions 1-3 would be helpful but how about the cost for last week or month to show trends.
I second the request on history of each ad. Many people jump into the market for a short time and then move on which can skew the data.
Garage Doors Pricing
Interesting article. First, some quick feedback. I would find the adword competitors and the broad/exact phrase match to be very useful. The better you know your customers the more important exact match becomes. I can give examples all day long, but the bottom line is that it is important.
I would also find it to be useful for you to break down the ad competitors by time of day. I am in an industry where a significant portion of my customer base is college students who happen to stay up late. I might miss out on a great opportunity b/c I only see the normal business competition.
This tool makes me look at wordtracker very differently. Up until now I have used/viewed it as a SEO tool more helpful for generating meta info. This makes it much more valuable as a SEM tool.
When is the full release with competitor info?
Jeremy
Hi Mike,
(we met briefly at Ken's System Intensive London last year).
A good post, and any tool that shows the difference in bidders on match types will be very useful. I manage and review many AdWords accounts and sole use of broad match only is a major problem I commonly see (along with no negatives).
Market territory is also key as there can be big variations. I had a great campaign in UK running for holiday planning - it tanked in USA until I used "vacation" keywords instead.
I think also you'd need to sample ads over a 24 hour period rather than just specific hours - again most advertisers are not sensitive to, or testing, scheduled ads unless they are experienced. And the longer you did so, the more solid the data becomes as busy markets have a lot of churn, with advertisers coming in and leaving all the time as they chase the latest affiliate "make money easy" dream.
All the best!
Has this feature already been added to wordtracker? also it would be good to see wordtracker adding more and more features for people like us who live by ppc/Adwords.
Great Post, though in my opinion, raised more questions than answered. For any body who wants to learn how to find profitable niches, there is book on lulu.com called Niche Goldmines Exposed. I learned all my niche research from it. It contains several methods in great detail
great post, nice point about if there is competition then people are probably making money from that keyword/area...
the only issue i had was that your data showed that currently there were no adwords running for some of those keywords, if that is something that google is choosing to do, no point targetting that keyword. There is no easy way other than checking manually if Google is no running any adwords for that search term.
Your post looks promising at the beginning. It starts to explain to me clearly and at a level that I can understand, how to form insights about key words and where to get the data from to inform my thinking. Then it suddenly changes and becomes complicated and confusing. Rather than leaving me wanting to know more or to purchase your service, it puts me off and makes me want to go and find another supplier who can explain the process clearly from beginning to end. Your question "Do you find that a useful number? Please tell me." is meaningless to me because at that point in your post i was finding none of it useful... just pure confusion. Did you write it at two different times? The style seems to change half way through. I hope this feedback helps.
This is a very interesting post. I think it is useful knowing the number of Adwords for various keywords. It's true that advertisers won't spend where there's likely to be no return. So if advertisers are spending money using certain keywords, those are probably effective words/phrases to use for optimizing a website. The strategy of looking for keywords that produce between 2 and 9 Google Ads is new for me, but it makes sense. It's like setting up a Chinese restaurant several blocks outside Chinatown rather than in its heart. It's still accessible to people who like Chinese food, but you're more likely to stand out.
I agree with most of the comments that as an indicator Adwords competition is certainly a metric to consider and that Wordtracker providing that information is a valuable addition to the service.
However, invariably, as other posters have highlighted, there are often more questions raised not answered directly eg demographics, seasonality etc.
No one tool can provide all the answers but there are others that can be used in conjunction such a Google Insights and Trends which show seasonal trends, breakout keyword terms and even seeks to differentiate between the sex of searchers.
Also, Google's new Search Keyword tool provides search volumes (over various periods) and projected click cost to achieve top positions and Microsoft AdCenter's Commercial Intent tool gives an indication of the buying intention of searchers.
The bottom line is; the more information you have available the better able you are to identify profitable keywords, phrases and niches to help make better marketing decisions.
Hi, nice post. I have been wondering about this issue,so thanks for sharing. I'll definitely be subscribing to your blog.
Just a quick thought, could the time frame that the google adwords feature ( you should employ it, it's a good feature) be a variable. Either selected by us or at least a choice of daylight listings and a choice of the less salubrious hours?
very interesting article. I wish the PPC actual cost was displayed though
Yeah, maybe there are other tools that will display adword ads but having it conveniently included within your initial research as an overview before digging further is a great time saver.
This is very interesting reading. I am always looking for ways to get better keywords which will be worth using but will not be overwhelmed with too many users.The most useful keywords I find nearly always have three or more words. In other words the much discussed long tail keywords.The challenge is to find the right combination of words which will bring a decent volume of searches but is not being heavily used by advertisers.This article is certainly helpful in that direction.JM.
Thanks for the wonderful read. Good that you have done a great deal of home work related to the topic. Keep up the Good Work.
All the best! Judy.
Looks as though this post is pretty old but I enjoyed all the same. One thing that has always served me well when researching niches (especially if I am monetizing with AdSense) is to keep my research "product centered".
In other words, I find the AdWords data to be far more useful when I am considering creating a site based on "things". These kinds of sites do better for me overall and with all the possible permutations, the number of keywords you can target can grow to be quite large.
Thanks again.