Not So Common Adwords Tips
Hey, and thanks for joining me on this edition of the podcast. I'm your host, Jason Brown, the co-founder of SERPWoo. I want to give a quick shout out to Chris [Dyson 00:00:20], Sean [Marquee 00:00:21] and Barry Adams. I know that they support and help spread the word of SERPWoo and I know they listen to the podcast, so I just want to give a quick shout out to those guys. Also, before we get into today's topic of paid advertising especially with AdWords, I want to go over what I did in the last episode where I shared three immutable laws of marketing from the book, The 22 Immutable Laws of Marketing from Al Ries and Jack Trout.
If you remember from the last podcast, I shared three laws and explained why those are immutable laws of marketing and help explain each law. I want to real quickly go into the next three. If you remember, I talked a lot about being first in the market is better than being better and I've talked about a few other laws and how all that connects. The fourth law is called the law of perception and it says that marketing is not a battle of products, it's a battle of perception. When you really think about it, truth is relative. Everything to your mind or the mind of another human being, that's relative. Are German cars really that much better than Japanese cars?
Did you know that Honda, Toyota and Lexus and Nissan even are perceived differently in United States than they are in Japan? Honda is nowhere perceived as well in Japan as it is in America even though they're exactly the same cars and pretty much for the most part, currently the same current marketing but the perception is a lot different to the Japanese than to the Americans. Really when it comes to thinking about marketing and thinking about your brand, it's not always a battle of what your product is but the perception of your brand, the perception of your product and the perception of the people who actually use it.
Now, the fifth law is the law of focus and it says the most powerful concept in marketing is owning a word in the prospect's mind. When this book was written, the word "computer" was owned by IBM. The word "copier" was owned by Xerox, "chocolate" by Hershey's and "cola" by Coke. It's very true today, there's going to be a word like say the word "overnight delivery," that's owned by FedEx in most people's mind. That's actually a perception, and FedEx plays on that, so does Hershey, so does Coke, so does Xerox because they were first two market, they became a leader and now, that word is owned to that company when somebody says copier or says overnight, there's even words that are owned by like a car company, the word driving is owned by BMW.
That is relative though to whoever the prospect is, so the word "overnight" for FedEx might be owned by an American who perceives that and has seen that marketing and has finally believed in that marketing, but the word "overnight" might mean something totally different to somebody in an Asian country who is used to dealing with DHL, and DHL is a major player within their country and has a lot more marketing and is more relative to that particular person in Asia. The most powerful concept in marketing is owning a word in the prospect's mind. That word could be just a perception to that individual person or group of people.
Now, the sixth law is the law of exclusivity, and that says that two companies cannot own the same word in the prospect's mind. When somebody like a competitor owns a word in your market, you're not going to unhinge them from that word. When you say "chocolate," you're not going to dismantle Hershey out of my mind from my perception. When you say, "overnight," you're not going dismantle FedEx from that particular word and it goes on with the same as cola, Pepsi has tried to unsee Coke for the word "cola" or just generically Coke, soda, pop. There are a lot of people though who love Pepsi and they think of Pepsi when they think of those words, but Pepsi actually targets a younger new generation, whereas Coke targets an older generation. After some time, that will probably flip flop.
You're going to see differences in locations within a country of people who recognize Coke more than Pepsi or even the word soda more than Coke or pop even. It definitely comes down on perceptions and it comes down to the individual. Winning the war in marketing really comes down to the individual level and what that person's perception is of your company and when you finally own a word in that prospect's mind, it is almost impossible to unhinge that word if you're a competitor and replace your brand with that word. Once you actually get instilled into a prospect's mind, you're more likely to be here forever with hardly anymore additional marketing once you win them over and that really goes back to the first three also. If you've missed the previous podcast, I highly recommend you go back and listen to the first 15 minutes, listen to the first three laws. Then, listen to this again and see how this all works and interconnects.
Now, we're going to go to podcast. I got asked a ton of questions about AdWords and paid advertising in general. Most of you all know that I handle a lot of paid advertising for universities and colleges, especially those who have online degree programs that they're looking to get adult, students in the road. They're looking to get online students in the road. They're basically looking for leads for their online degree programs, sometimes even their traditional ground programs but I've worked with many universities and colleges on their programs. I've also worked with a lot of drug rehab facilities and I've worked with a lot of people in general from business-to-consumer to business-to-business from pharmaceutical and supplement all the way to working with colleges and lead generation.
I've worked with many, many millions of dollars of advertising over the course of close to 20 years now within paid advertising space and a lot of people come to me and ask me the same questions over and over again, especially AdWords in the Facebook, but with AdWords in particular, they're always trying to either get started or more commonly, they've gotten started and now they want to scale up and ramp up and really find where their profit is at and what they can tweak into and help them get to the next level.
I thought for this podcast, I would throw out some random tips, some random suggestions over common questions, over things that I found from managing millions of dollars worth of paid advertising especially within AdWords and probably the first tip that I'll give you is that when it comes to geolocations, it's pretty straightforward that people like to target their location and then exclude another set of locations, and that's fairly straightforward, fairly simple but when you're a true pay-per-click, paid advertising analyst, something that people don't think about is trying to reverse even if it sounds a little odd, even if it sounds like, "Well, that should work anyway," you really need to dig in and ask these questions, come up with these possible theories, hypotheses and truly test them out because a lot of times, you find some gold nuggets that might only be 2 or 3% increase, but that might be 2 or 3% increase within your geolocation.
Then, you do a 2 or 3% increase in maybe your day parting than you do a 2 or 3% increase, say within your demographic if you're doing display targeting and that's all going to add up. For geolocations, I'll give you this scenario. Let's assume you're a plumber in Chicago and you want to advertise your business in that local area, you don't want to advertise nationally of course, so you narrow down the geographic reach just to the Chicago area, and you might even narrow it down further to just the section of Chicago because maybe you're not able to service the west side of Chicago which is too far for you. What a lot of people will try to do is they'll target maybe the south side of Chicago and don't even go as far as doing zip codes, so they know that they can service these particular zip codes because they're close to them and they don't want to drive out of that range because it starts costing them more money.
What they end up doing is within AdWords, they look at their geolocations and they start saying, "I want to advertise in ..." Then, they list 10 hospital zip codes. Then, they're going to exclude the rest of Chicago, the rest of those zip codes, the rest of the areas just because they don't want their ads to show in those other zip codes. That's a straightforward way, but I want to tell you this, have you ever tried the reverse? What the reverse would be is selecting all of Chicago first saying, "This is where I want to show my ad all of Chicago," and then excluding just the zip codes you do not want to advertise in.
Instead of selecting the zip codes first and then doing an exclusion even if you don't do an exclusion, just select all of Chicago and just exclude the zip codes you don't want. A lot of people might think, "Well, that's reinventing the wheel. That's exactly what I did." No, that's not actually true. Many times I've done this test and I found out that I actually have received more traffic doing a reverse and it was traffic that was qualified and traffic that was actually in the zip codes than if I would have selected the zip codes first and excluded everything else, or not excluded anything. If I did the reverse, selected all of Chicago, and then just excluded the zip codes I didn't want to be in, I found that I've gotten more traffic, I've gotten more volume, I've gotten more leads actually and they were the correct leads, they weren't leads from areas I did not want.
Another common question I get is about budgets and about bids, and usually the questions are how much budget do I need? What should I bid on my keywords? I've only got a $1,000. Is that enough to even try AdWords? Really, that's a black hole. It's a question that really comes down to what can you personally tolerate and how have you setup your campaign in the website? Because at the end of the day, the answer that I want to give is you're going to lose money upfront, that's the expectation I want people to have because if you're not making money out the gate, that's great. More likely, you're going to lose money because this game of AdWords and paid advertising really boils down to buying data. When we're buying data, the key thing that we have to remember is not all key words will perform the same at the exact same time, not all of the ads will perform the same at the exact same time, not all of your prospects or leads are going to act in the same way at all of the same time.
People, they're going to act differently, your ads are going to influence people differently and how AdWords reach your keywords in your account is going to be differently throughout the day too for a lot of different variables. What I try to tell people is that at the very core, you need to find out how much a customer is worth to you? A lot of people get this wrong and they get it tricked up and they really think that customer's worth maybe less than the initial transaction or the initial value of their first conversion for them, which might be a lead form where there's no actual transaction value but that lead might turn the customer later and there's value there. They might assign a small value to that lead, but that's actually incorrect. The initial transaction, the initial conversion is not what that customer's work to you unless you got a really poor sales funnel and sales cycle because actually, the customer is worth whatever their lifetime value is to you.
If the AdWords' lifetime value of a particular customer is $1,000, then you've got at least $1,000 that you can spend before you get customer if you plan things out right. Now, you might not actually want to spend $1,000 because you might have product cost involve or you might have other cost involved, but a lot of people get this wrong and they say, "Well, the customer bought a sample of some essential oil that I sell and I sell that oil for $20, the cost of goods are $4 on that. Then, I've got shipping and then I've got a few other expenses added in as it goes out the door." They might actually say that the customer that buys that product to them is only worth maybe $10, but if you have things set up correctly in your business, and you know that on AdWords, people who buy that product buy that product plus spend another $200 while they're a customer of yours, that $10 is not actually correct. You want to work off that $200 mark.
Then, that's where you actually start and work backwards from, it's that lifetime value. Now that we know how much we can spend to acquire a customer, a lead, a prospect, we start breaking down how many assets does our AdWords camp has? How many keywords? How many ads? How many landing pages? How many different veritables might be in play as far as desktop, mobile, tablet, as far as geolocations because these assets do not leave in our backing. They're all going to impact and effect each other at different times of the day and different locations with different ad messaging. If you've got a campaign that has 20 keywords and it has three ads, and you're displaying to all of the United States, a certain group that have 20 keywords might actually perform better when one of the three ads is being shown for compared to when a different ad is being shown for that side of keywords.
You're going to have instances where certain words do better in maybe the state of Georgia than they do in California. You've got a lot of this compounding going on where certain things are working correctly and certain things do not work correctly when you put them together and you don't always see those things upfront when you're in your AdWords account. What I like to tell people knowing that is that you want to give each keyword three to four times of the value of what you can pay to acquire a customer or a lead.
I know in our earlier example, I used $200 as the lifetime value for our customer. If I was running an e-commerce store, well that was true. I would want to allow a keyword to spend possibly $600 to $800 without conversion before I shut that off because that's how you're truly going to test, and get the right information on the keyword. Now, I know that's probably blowing a lot of your minds out there but this is how you buy data to actually then go back and look at that data and say, "We gave it enough statistical data and relevance to look at it to say this keyword is definitely a waste," because what will sometimes happen is you'll be running a keyword and especially if you're an expensive, a very high cost vertical, you might be paying $25 keyword.
There's times that I paid over $100 a keyword, you might only be paying for this sense of keyword but what ends up happening is especially in an area like a higher education or in drug rehab, even in some of your poor professional services, you're going to pay a lot of money just for the click. If my lifetime value is $200 and I'll allow this keyword to spend up to 600 or $800, I'm giving it enough time and enough attention to gather the right data because there's been very many times where a keyword has hit the upper level of the goal that I wanted to allow to have. Let's go back to the $200. Let's say the upper goal was $600, that's three times its goal. I might be at 590 and I'm thinking at myself, "Tomorrow, I'm going to have to shut this off. It's just waste of money and spending money," and I go to bed, I wake up, two conversions have came in, two leads, two sales. Then ultimately now, I'm at $600 but I've got two conversions, two leads that happen to come in back-to-back ,maybe even within minutes apart or hours apart and now my cost per conversion is $300.
I've seen that happen tons and tons of times and then you can take that and start maybe bidding that keyword into maybe second or third or maybe you look at the data and you notice that both of those sells came from California, so you tell yourself, "You know what? I'm going to advertise just the California now," or you find out that the conversion's coming in late at night more frequently than they do throughout the rest of that day, because maybe the people you're targeting, they're nurses and there are ins and they're looking to get their bachelors of science in nursing, their BSN and they're more responsive late at night maybe when they're working at the hospital and they're working the third shift. You don't always know what's happening but you've got to spend the money to find it out. If you start making those tweaks maybe just in the state of California, maybe just late at night you bring the bid down to bid in an average of third spot, you might find out that now you're getting five conversions a day and your cost per conversion is now down to $67.
If you wouldn't have allowed that first almost $600 of spend, you would have missed all of that. You would've shut it out the door. You would've called it a fail, a waste of time, and you would've never have really been able to allow the spend and allow the data to come in for you to find out how to correctly advertise that keyword. Now, that's just one keyword. When your campaign has 20 keywords, you need to allow all those keywords the exact same thing and when you think about that, you're going to spend a lot of money in a campaign if you've got 20 keywords that technically need $600 each in spend before you start cutting things out and you really need to look at your ads the same way until your ad should have to spend a certain amount of money before you cut an ad.
That's usually going to happen but what I find out is a lot of people build their campaigns, they have 20 keywords. They want all 20 keywords to get $600 in spend before they start optimizing but then when you look at their ads, they've got like 20 ads. That spend is spread across those 20 ads and you really can't say that this ad has spend enough money relative to the keywords to say it has enough data to start cutting it. Sometimes you can see those things quickly, some ads are just complete duds and you'll spot that but truthfully, your campaign or at least your ad group should not have more than three ads in it because once you start getting into an ad group that's got 10 ads, and you're spending all this money, well now when you go look at your ads, it's so spread thin. You don't know what the cut, at least not with statistical data, so I recommend three ads in your ad group.
When I'm looking at geolocations, it's the same way. There's throw advertising in the United States, "Hey, there's 50 states. There's thousands of cities. There's thousands of counties." When you break it into Canada, you compound that. When you break across the world, you compound it even more. The last thing you want to do is go into your geolocations and cut a state that only has $80 spend because there hasn't been a conversion yet and if we're going back to the $200 lifetime value and a $600 spend for keyword, you're going to want to let that state spend more than $200 and with 50 states, you really start compounding the impact of how you built your campaign, how you got built your ad groups out, what your lifetime value is, and ultimately even how your landing pages and sales funnels perform because if you got a really weak sales funnel, you're going to blow through a lot of money.
If your campaign structure has a ton of assets in it like a ton of keywords or a ton of text ads, a ton of image ads, a ton of geolocations, you're just going to compound that spend at least to be statistically relevant to a very high number and you're going to have a lot of risk associated. Bottom line is when it comes to your budget in your bid, the correct answer is you're probably going to have to have a lot more money upfront than you ever realized at least to get correct data and data that you can actually work with. The way that you mitigate that risk is is you start your campaigns very, very small. If I've got somebody who's highly price sensitive and I'm not sure how their sales funnel is going to work and things seemed sketchy [over in price 00:26:02], what I'm going to do is this, I'm going to set up one campaign, one ad group. It's probably only going to have two or three keywords in it which are going to be the keywords that are the core keyword of that service or product.
Then, I'm going to put three ads in. I want to advertise to maybe three or four states and the way that I got those three or four states was talking to the customer or looking through my data and realizing that we get a lot of customers who are paying customers or we got a lot of data that's good data for leads within these three or four states. In the higher education market, it really breaks down to a ton of the people who ended up being actual students live within 80 miles of the home campus. When I'm making these campaigns and I'm really worried about price or losing money or risk, the campaigns are going to be very, very small. Two or three keywords, three ads, three or four states, I'll probably leave the time for the day parting all the way open. I'll probably leave mobile and tablet open as long as well as desktop.
When I'm now working with this campaign and maybe I need to spend $200 per keyword or $600 per keyword because my goal was $200 for a lead or a conversion, now the most I'm going to be out at least on my keywords if I'm only running two or three keywords is going to be 1,200 or $1,800 and because I focused it within a few states and I focused it within a few ads, I'm going to find out real quickly which ads work with which keywords and then, I'm going to look at the data and find out how that compares in each state and I'm going to [pur 00:27:57] my state down. Then also, I'm going to look at more data to find out, "Well, which position am I really getting the best ROI in? Is it really first position, second position, third position, fourth?" When we have right hand ads, I loved betting for almost every industry, third and fourth ad placement.
Now that we don't have right hand ads, the data is a little bit different because if we're third or fourth on average, that probably means we're spending a good amount of time below the fold. You really need to look at that data and if you're trying to keep the cost low and your budget low, and your bids acceptable, you're just going to have to build a very small campaign that has very few options and focus that money on those options and then start to build out because if you focus that money upfront on two or three keywords, three ads, four states, you might spend less than $2,000. At the end of that run, you realize, "You know what? All of my sales came from California. At least in an ROI perspective, they all came from essentially one ad and that one ad worked very, very well for the two or three keywords that I'm running."
By the way, I found out that most of the sales come before noon between 8 am and noon. Then what you do is this, you basically rebuild that campaign, you tweak it to just California, just from 8 to noon, you tweak it to where you just run that one ad. You might want to make a second version of that ad as a B test and then what you do is you relaunch it and you relaunch it with a few more keywords, keywords that you learned matched to your original two or three keywords. You might have learned that there's five more keywords they got matched that are bringing within sales too, so now you add those in. They're only running in California and you're only running on that one ad and maybe also your B copy of that ad and you just start to expand out that way because now you know that these core keywords are what's working, this core ad is working and this core state is working.
When you went to start to scale now that you're profitable, you slowly add things in like you slowly add back in one of the states that failed because the question that you have is did the state of Nevada fail because maybe one of the fail ads shown more times in Nevada than in California, or is the timezone different in Nevada than it is in California? Your timezone is going to be based on what you said in an AdWords but maybe because your AdWords are set to Eastern Time, maybe Nevada or another state that you had that had failed was at timezone behind, and maybe that impacted why that state failed.
You start adding things back in one by one, you start testing things out one by one but only when you have that core success. In the beginning, you don't know what that core success is even though you've got a great, good instinct and you might have some prior data, you just don't know where that's at. Then as you build out and you start seeing other problems in other areas for advancement, you start bringing in another campaign that just focuses on that and that's how you build out. That's how you set your budget. That's how you set your bid, but you don't know that until you have the data. If you're price sensitive, cost sensitive, the only way that you're going to be successful is you're going to have to wait, you're going to have to be patient, you're going to have to build a small campaign.
You're not going to go gangbusters overnight and be rich and have this successful campaign and you need to explain that to your clients, if you're a freelance or a contractor, an agency because a lot of these [inaudible 00:31:47] think that they're going to come to you and get success within 30 days. Then, you're going to triple the revenue in 30 days, when in reality, what you probably have to do is go into their account, tweak it and optimize down to a core, get it really, really profitable and then start building out and look at all those failures and see if those failures now work based on your core ideas.
I know that the 200 cost per lead or cost per sale is high for some of you guys, when you're at $600, they probably blew your mind even 800 without a conversion, it probably blew your mind. Luckily that's the experience for me because I work in a vertical that's very, very competitive, very, very expensive. You, however, might be in a different situation where your cost per lead or your cost conversion has to be at $10, has to be at maybe $35 or $30. At that point, the role is still the same, three or four times that which at $30, three times, that would be 90, four times, that would be 120. That's what you're going to want that keyword to go before you cut it.
Then even after you cut it, and you find other successful keywords in your account and you start cutting out states and start cutting out ads, you probably want to actually retest that word because maybe that word or that term works better in a different state or works better with the different ad because listen, I could bid on five keywords all day and I could bid or create an ad that has the word, "Free iPhone," in it and I'll have thousands of people clicking on that ad and never get a sale if my keywords don't match up to that.
Even your failures, you're going to need to retest but don't get freaked out about spending 200 or 600. That was just for my example, you need to find out what your lifetime value is, what you can pay for a lead if you don't know the lifetime value. Then, times that by three or four. Then, look at keeping your list of geolocation small, your ad small or your keyword list small. When you get failures, have a plan to cut those out but to retest them later and once you get to your core, that is actually profitable for you.
When I'm looking at an account, and I need to start optimizing, a lot of people don't know where they should start. Believe it or not, this is actually very important where you start cutting because if you've got 20 keywords in an ad group or campaign and some of those are bad keywords, keywords that you don't want to cut for whatever reason because you believe that that's part of your brand, or that's an important keyword or somebody else told you to keep it but it's not giving you ROI, and you instead go in and start cutting ads, cutting geolocations, cutting mobile out, you're really doing some unforeseen damage to your account that you'll probably never realize until later if you give it the thought.
Where I like the optimize and cut first will always be with my keywords. The keywords are pretty much for most people what makes their campaign profitable or what makes it a failure. Again, I'm excluding the fact that you maybe wrote an ad that's misleading that says, "Hey, free iPhones." I'm ignoring that aspect of it but for most people, they live and die by their keywords and that's why keywords are so important. The match type is so important. The bid that they're in is so important and the negative keyword list is so important. If you get those things correct, it almost doesn't matter how you're day parting, what your ads are saying as long as they're relevant and they have some good copy. It almost doesn't matter where you advertise as long as where you advertise is a place that's valid for you.
If you cannot take sales from Canada or ship to Canada, you shouldn't be advertising in Canada. Before all of the things, you live and die by these keywords pretty much and these keywords will impact all of those other areas whereas if you've got some bad keywords that are dragging on your quality score, if you've got some bad keywords that are getting a lot of click throughs but they're not converting, you're going to do much more damage going in and trying to adjust the day parting, adjust the platforms, adjust the ads before you do those keywords. Whenever you're optimizing, you need to look at the keywords first.
Once all of those keywords start having statistically relevant data and they've spent enough money to have that data and now you're looking at your goals and you're hitting your goals like three or four times over the value that you need, you start cutting back those keywords. You're adding them those negatives, you're looking at maybe I should bid this keyword in the third instead of first. Once you have that set, once you have those core keywords and spend enough money to finally have those core keywords set and those core keywords are running, that's when I then go back and I start cutting or changing or looking at my ads.
That is the time I start looking at my platforms like mobile, desktop and tablet. That's when I start looking at my geolocations to tell that, "You know what? I'm doing really, really horrible in New Mexico, I'm going to cut the whole state." But wait, maybe it's not the whole state. I've got a couple of conversions. They're just not profitable and then you look into the city level of that state and you realize, "You know what? Really the only place I'm losing money is Albuquerque. Let's cut just Albuquerque and run the rest of New Mexico," and now you've made that profitable.
I'll do all of that once and after I've paired down my keywords and work with my keyword data because then once I've done that, I'm not getting match for the irrelevant keywords because I've got a great negative list. My keywords are not being impacted by competitor so much or [inaudible 00:38:33] so much because I've been able to bid them into the proper placement now like maybe third. I've looked at this keyword, if I've got it hooked up to Google Analytics, I can tell the bounce rate so maybe this keyword shouldn't go with this particular landing page after work. Maybe I need to make a new landing page and send that particular keyword to this new landing page before I start cutting out states, platforms, day parting and such because you're going to take out possibly good platforms, good states and cities, good times of the day, good ads when in reality, the problem is still your keywords.
Another tip I have is that you can judge how healthy your account is running especially if you're in direct marketing and you're tying your paid advertising to the results that it brings in. You can really judge the health by looking at things, at cost per thousand or cost per milli. The way that I'm looking at that is yes, I'm probably bidding cost per click. I might be doing an enhanced, I could've picked one of the other options like bidding my CPA and letting Google handle it, but I'm going to pull down all my reports, and I'm going to work in Excel and I'm going to look at how much I've spent and how many impressions I've gotten even down to a keyword level, down to the state level, down to an ad level and I'm going to look at what is my cost per thousand, per thousand people that I've shown my ad to based on this keyword, based on this ad, based on this state, based on even this platform like mobile, tablet, desktop, how much money have I spent? How many impressions that I get from it? None on my cost per milli, cost per thousand, that's real easy to get.
Then, in another column in Excel, I'm going to look at my conversion value or if you've done it right, if you're an e-commerce store, it's going to be your revenue and I'm going to look at, "This is my revenue. Now, let's divide that by the same number of impressions and I'm going to get my revenue per milli or revenue per thousand impressions, and I'm going to look at that and I'm going to make choices based on those numbers," because when you really want to look at, "Is everything effective?" It's much larger than this keywords profitable and individually, this ad is profitable and then individually, mobile is profitable.
You really start tying these things together and combining them at the cost per milli and breaking it down to even the keyword level and you can start seeing this is the overall health because if I've got a keyword, and I know the cost per thousand and I know the revenue per thousand, if the revenue is higher and depending on the amount higher it is, the increase, I know that that keyword in general is matching up to the right ad if I've got multiple ads and is showing on the right platform, same mobile, desktop, tablet, and that's showing in the right states at the right times, that the health of that, the health of my account is good and that the messaging that I'm sending on that keyword in those states on those platforms is good and healthy and I'm going to be able to judge really quick the overall health of my account based on those numbers.
I'm going to be able to base it a lot better than just quality score. I'm going to base it a lot better than just what's the conversion value for just this keyword. When I look at it at the per thousand level, I know immediately that I've got the right core keywords, the right core ads, the right core platforms, the right core geolocations and the right core day parting schedule that I'm profitable for that keyword or for that ad or for that platform. Then, I look at the next keyword. Then once I have all of that data on my sheet, I can look at it as a grand total and I can tell myself the right messaging is matching to the right keywords in the right places at the right times, in the right level and the right platforms like mobile and desktop, the way people surf and they're hitting the right messaging on my landing page too.
When you look at it as cost per thousand, you're genuinely looking at the entire flow of your system from the times when I type something into Google until the time they hit your thank you page after they've ordered or at least signed up for, "Fill you booking form," as a lead. Most people don't break things down as cost and revenue per thousand, they rely on just looking at quality score. They rely on just looking at the conversion value for just that one element like say, for one ad or just for mobile or maybe they look at click through rate and try to gauge, "Well, was this the right ad for this right keyword?"
You really need to look at it for the entire cycle that that searcher has done to type in your name, see your ad, click on it, then lay in your landing page and then the side to fill out a form or buy the product. When you break it down as cost per thousand and revenue per thousand, you get a much better look at that entire flow. For the last generalized tip that I'm going to give you for AdWords is I got a lot of questions about campaign, ad group setups, how to setup the keywords, what's a general strategy for that? Really, it just depends on your business. It depends on how you're going to market, but if I break it down to a really, really high level ... I already told you about possibly starting small.
Another thing that I like to do to help control cost and really find really good data is you really want to work with exact match keywords. There are possibilities with phrase match, modified, broad and just road keywords that but if you're just starting out, you really want to focus on exact match keywords, and I've got a strategy to find out what those exact match keywords are. The first part of the strategy is getting a tool that can tell you what your competitors are being shown up for for their keywords and if you use a tool like SEMrush or SpyFu, you can literally put in the domain name of one of your competitors and find out all of the keywords, find out all of the ads that they're using and being shown for, and you can go through those keywords and filter them out and copy paste them into your account and you're just going to make them exact matches.
Now, wait and just do this on a wholesale level, you really want to find out the keywords are going to be most relevant for you so you've got a competitor that sells blue widgets, yellow widgets, green widgets but you only sell green, you're going to want to filter that out to just the green widgets and you might even want to take it one step further and get really highly relevant and just get the, "Buy green widgets," keywords or green widget review type keywords instead of getting all of the green widget keywords, the ones are probably going to lead into a purchase or at least gets you to the middle of the funnel and not just the very top of the funnel where people are just trying to browse and then, being a [tire picker 00:47:34].
You could filter down that way and the second option that I have which I recommend doing this one as well is I like to set up a campaign in my account and it's a discovery campaign and by discovery, I mean I'm going to pick probably one or two really relevant keywords. If I go with green widgets, let's just say that one of those keywords is green widget itself. In this campaign, I'm going to run that keyword as modified broad, that's with the plus symbols and in front of green and the plus symbol, in front of widget, and I'm going to bid that into the number one spot, so I'm going to pay as much is I can for that number one spot and I'm probably going to set up a negative list ahead of time because if I'm selling a product, I don't want anybody looking for a free green widget. I don't want anybody looking for a green widget stamp and I'm going to put those negative keywords in as best of my ability, and run it as modified broad, and I'm going to pay as much money as I can to be number one.
What this discovery campaign does is it's spending money, it's probably going to lose money but it's buying me all of this data about the green widget and whenever I get a conversion, I'm going to look in my search term report and I'm going to look at the term that matched to green widget and got me a conversion and I'm going to copy that keyword and put it into a second campaign. The second campaign is just going to be exact matches that I know work and I know gave me a conversion. I'm going to take that keyword from the search term report and I'm going to copy it, and I'm going to paste it as an exact match keyword into the second campaign, and I'm going to bid as much money as I can for two, because I know that if somebody types in this exact term whether it's a four-word phrase or a five-word phrase, whatever matched and I got a conversion on, I am going to put that in as an exact match and a new campaign, I'm going to bid as much money as I can for it.
Somebody is going to have to type this in exactly and then, my ad will show and the chances of me getting the conversion were probably going to be pretty good since it's already got in conversion before. Now, what I'm going to do is the keyword or key term, I'm going to now negative match that in my discovery campaign, the first campaign, the campaign I found it in, because I do not want this modified broad term matching for this again if somebody types in exactly. Generally what I'm going to do is I'm going to just go into the negative list for that campaign or ad group, I had no preference. I usually do that at the campaign level and I'm going to go on my discovery campaign and paste that in as an exact match negative so that that discovery keyword that modified broad keyword no longer matches for that on an exact match but my second campaign will and hopefully I'll get another conversion at a good rate.
Now, I will tell you this. I do this a lot with campaigns and you'll get an exact match that gets you a conversion, then you'll go through this process and you put it in its second campaign. There's no guarantee that that keyword will ever bring you another conversion again because possibly, it's a very low volume and you just happen to match for the one person who search for it in the last three months and convert it. Also, it could be good volume but maybe you don't get another conversion ever again and how you spent $300 on it and you've only had that one original conversion. I've seen that happen, so there's no guarantee that this is a highly profitable set up but what it does do is it helps me start filtering down what is exact match? What is converting? I put that into its own campaign, I give it its own budget and I let that run.
After several months depending on the volume of the keywords that you're bidding on, after a few months that discovery campaign almost becomes worthless and it starts losing money, if it's not already lost money, because you've basically dug all of the gold nuggets out of it, so you need to be mindful that, to watch that the discovery campaign over time as you add more into the exact match setting campaign and you're negative matching those terms back into the discovery campaign. You want to watch out how the performance impact is set over time so that you can cut that off when it's not losing money and you'll know at the right time when you dug all the good gold nuggets out of it.
Someone give you a bonus tip actually, you need to think of things when you're at looking at your metrics as blended totals. To give an example, let's say that you have decided you do not want to pay more than $80 for a lead or conversion or sell and you're attaching this to keywords. You might have five keywords in your account or for your ad group, for your campaign, and for those keywords, they're all at a $50 cost per conversion, so you think, "Great." The fifth keyword might be $100 for its conversion and you think, "I want to cut this keyword out because it's obviously over my goal. It's costing me money." But when you really look at the total, you notice that this $100 keyword is bringing you in five sales a day, five conversions a day, five leads a day.
The other four keywords, they're at $50 per conversion and together combined all four of those keywords are bringing you a conversion of only five a day total combined. You're getting two conversions a day, five from the four keywords that perform at the $50 cost per conversion and then you're getting five conversions from the one keyword that is at $100 conversion. When you look at that as a total for your ad group or your campaign or possibly even your account, when you blend that and you look at it as a total, you're getting 10 conversions a day at possibly $75 of conversion. You're still under your $80 goal, but if you want to take that fifth keyword away, you're going to lose half of your leads, half of your conversions. It would make much more sense to try to figure out how can I earn more from these leads from these sales possibly with an up sale, cross sale, possibly looking at the future sales and sales funnel, possibly maybe moving that one keyword that's at $100 per conversion into its own campaign and trying to tweak the ad just for that one keyword or maybe the geolocations and day parting.`
It would make much more sense to keep that keyword and figure out how to make it more profitable than to cut it, because you're going to lose half of your leads even though it's over your cost per conversion goal when you look at it in a blended sense, it makes up half of your leads and in a blended sense, you're still under your goal as well. That's the last tip I want to give you. That was a bonus because I see a lot of people who get really eager or really trigger-happy when it cuts stuff out and this is also a reason why you want keywords to go three times over their goal, or four times over their goal because even if they're over their goal by three or by four, they could be helping you bring in a lot more volume possibly and in a blended sense you're still meeting your goals at the account level or at the campaign [inaudible 00:56:47], you're still meeting your goal that you need in a blended sense and then you try to work to make that keyword more profitable for you, or less expensive within its own campaign maybe down the road.
For the business idea on this podcast, the business idea is this, and this is related to pay-per-click advertising especially with AdWords. Somebody needs to come out with a service selling negative keyword list and it's a little bit more difficult than it sounds but I definitely know it can be done. I know everybody's business is different. Listen, I've handled tons, tons of campaigns from AdWords to Facebook, to Bing, to a whole bunch of third parties. I've worked with tons of different companies and tons of different industries handling millions and millions of dollars to spend a year. I know that all of these accounts are different, but I can tell you if somebody were to come out on a negative keyword list per vertical, there's definitely some money there and now especially if you're looking at the very high-end verticals, there's definitely money to be made.
There might not be sales every single day but what you can chart for it would definitely more than a make up for it. I'll give you an example. I've worked a lot in the higher education vertical and by this time, helping more than 50 universities and colleges, I've got a pretty good sense of what keywords do not convert in general. Now, the keyword list that I would have probably wouldn't apply to every single degree program or every single school but I would know that going into it, so I would have this master list that I know that it generally does not work or I know that I can put this keyword list then, and then alter it for this core or for the program.
That is very, very valuable because these negative keywords cut a lot of cost out of these accounts. Now, when you're dealing with keywords that cost between $23 a click to over $100 a click, it's massive savings and it's massive time that saved for the person who is managing the account. I know also on the drug rehab space that this negative keyword list are worth more than gold, I actually work with somebody who paid a lot of money to two different people to get a good negative keyword list. In the end, the keyword list were shit. Then I had to go in and basically do the discovery campaign idea and going through and building up our own negative keyword list but now that we have this, we could easily launch drug rehab campaigns for all kinds of facilities all across the United States and have a very good, good percentage going into that it's going to be a profitable campaign because we've got this negative keyword list rolled out.
To give you an example, in the drug rehab space, you're going to keywords that are keywords for physical rehab like somebody is getting rehab on their knee or it's going to be rehab like a television show name where celebrities go and they're doing rehab on a talk show. There's lots of keywords like that. In the higher education space, there's people who are looking for courses and classes that aren't necessarily university or college space, stuff like [inaudible 01:00:50], stuff that might be free classes online or classes by an internet marketer.
Knowing those keywords and negative matching them out is very valuable for somebody that's in higher education or drug rehab or any of these other super competitive hot click price verticals that roll out on paid advertising and if you were the go-to person and you actually had correct and good data and not just some list that you combined from all of these resources online because there's free websites online, there's blogs online were these people have combined negative keyword list and it's very, very generic. It's stuff for like the word free for e-commerce. It's just very generic words. There's usually about 200 or so but if you were to be able to come and say, "I have a negative keyword list for higher education. We spent millions of dollars. We've got a good list. Now, we want to sell this list to other people or to drug rehab, we spent millions of dollars in drug rehab. Specifically we have this list. We want to sell it to other people." You would make out with a ton of money per sale.
You probably would not get a sale every day because there's only so many drug rehab facilities or so many colleges and universities, or so many opportunities for that to even come up within those industries but if you were to look at all of the industries and then just have this, it would be a very simple digital delivery. It would be pretty much hands off. There would be very little of any customer support other than how to use the list and to know that the list still possibly needs to be filtered out for your specific use. Maybe the list was built knowing that in the drug rehab space that the rehab campaigns that these were run on were all in South Florida, so there might be states in there that are in the negative list like the state of Montana.
Somebody who happens to buy this list, if they ever rehab in Montana, they're going to want to have to go through there and filter that out, but that is very, very little work compared to the cost savings in both time and in profit to do that task than to do it the opposite way which is throwing a lot of money in your campaign, and find the keywords are aren't profitable and then try to negative match those keywords. This is probably a very good idea for people who are already pay-per-click marketers who already work at agencies, or maybe people who work at companies and your company is in a highly competitive vertical and you've got access.
This would be a cash cow for somebody who could set this up, do a digital delivery, have very little customer support and charge a lot of money for it because I know in some verticals, having a list like this is worth thousands and thousands of dollars just on its own because of what it can do and what it can help. That's the business idea. If you all use these ideas, if you like them, if you don't like them, let me know, let me know in the comments, let me know in the direct message. Get back to me. Let me know any other questions or concerns you have below and I'll see you next time.
Love this content? Create a free account and you will automatically be signed up to our newsletter and emailed each new blogpost when it's released. You can unsubscribe anytime, and we don't spam! We hate spam!