Online Advertising Campaigns Are Built, Not Bought
Online Advertising Campaigns Are Built, Not Bought
Hello fellow advertisers! Today I will be discussing the three foundations of successful online advertising campaigns: search engine marketing (SEM), search engine optimization (SEO), and landing page optimization (LPO). Each of these “pillars” of line advertising directly affect each other, so I’ll be reviewing how to properly identify where you could be further optimization your online advertising strategies. Enjoy!
Hello everybody and welcome to another episode of my video blog series. I am your host, Nick Rennard, and today we will be going over video blog … This actually got stemmed from a quote that I use a lot with my clients, or one liner that I use a lot with my clients where I say that advertising campaigns are built, not bought. I’m going to elaborate more. I actually mentioned this to my boss on Friday and he said that he really liked that line, and so I decided, “You know what? I’ll just write a little bit about what I mean by that,” and yeah, go over the foundations of advertising campaigns and some of the variables that if you’re running advertising campaigns yourself, you may be missing, and yeah, let’s just go ahead and jump right in.
Most of the focus on this vlog is going to be on search engine marketing or pay-per-click, because that’s what I do, but I’m also going to be talking about the other aspects, what I call the “three pillars of online advertising,” which would be LPO, SEO, and SEM. LPO stands for Landing Page Optimization, and it’s pretty much how good your website is. There’s always something that we could be improving on websites. People are constantly doing A/B split tests on websites through programs like Optimizely and stuff like that. Having a better website, pretty straightforward. If it’s a really nice website, people will get a good first impression, conversion rates will be higher, user engagement will be better, and the inverse is true if your website is not so good.
The second pillar of online advertising is SEO, Search Engine Optimization, which to put it simply, is how well you rank in the search engine results page, so Google, Bing, Yahoo, I mean even Amazon, whatever. Any kind of a SERP page, your ranking, I mean, that is the core of SEO. Any kind of SEO optimization you can do is similar to landing page optimization where if you clean up the coding on your site or something like that, or add meta tags when you didn’t have meta tags, whatever, all that kind of stuff makes it so that you will rank higher and get more business out of it. I’m trying to keep the terms here pretty simple. Also, SEO isn’t really my field so I don’t want to go too deep on there without not knowing what I’m talking about.
Then, the third pillar, and the one I am most involved with would be search engine marketing, or SEM. People also refer to this as PPC, which stands for pay-per-click. Things like adwords and Bing Ads are probably the 2 most well known pay-per-click platforms, but essentially you put money into these things to bid on search queries that are typed into Google or Bing or wherever, and you’re paying to have your ads show up for particular keywords that you want to be showing for. The important thing about these 3 pillars of online advertising is they directly affect each other, and the reason that I use the line of … I’ll go back a slide here. Of “Advertising campaigns are built, not bought,” is I get a lot of clients where … I work in SEM, so a lot of times we start building a campaign and when we’re pulling out bottlenecks as to what may be preventing a campaign from doing better, identifying the biggest red flags in an account, often times it isn’t always SEM.
It always the quality of the campaign, but rather, a common thing that comes up is, the website is not good, or outdated, and they’re getting bad quality scores because their landing page isn’t good, their conversion rates are bad because people are landing on the page and it doesn’t look good so they just bounce back, getting a bad first impression is not good. That’s a case where even if we setup a perfect campaign for SEM or PPC, it could still fail because their landing page is terrible. The inverse is also true. We could also have a not so good campaign, but an insanely good website and the advertising campaigns might actually perform well just because the website is so good, but you know, the campaign isn’t necessarily optimized to the best that it could be.
Having these pillars built up, the problem is that they’re going to start bottle necking each other, so when you have a really bad website, it’s going to make your SEO bad, it’s going to make your SEM bad, so that’s what I mean by how they directly affect each other. It’s important, about this quote, saying that campaigns are built not bought, is that we have to address all aspects of these 3 pillars of online advertising, and that’s why I’m going to be going into some of these a little more in depth here. The line I have here it says, “You need to build a foundation of all of these to be successful. A perfect SEM campaign still won’t generate profitable revenue for your company if your SEO/LPO are subpar.”
Anyway, pretty straightforward. Next here, now this is something that I always try to clarify with clients. One of the main reasons is for expectation management, and it’s also true, which is probably the main reason, that nothing is guaranteed in SEM advertising. You could put 5000, $10,000 into SEM advertising, and often times I am handed accounts that were built by previous companies, or sometimes they’re built in-house, and I can see that these people are spending tens of thousands of dollars a month on online advertising, and they were getting practically nothing out of it.
They were essentially just taking that money and lighting it on fire which yeah, not good, but at the same time, nothing is going to be guaranteed in SEM advertising, so we need to be aware of the risks that can happen, and make sure that if we can identify that enough problems are going on with, let’s say our SEM campaigns are perfect, but the landing page sucks and our SEO sucks, we need to make sure that we’re allocating our resources properly and readjusting to focus on the pillars of that foundation that need the most help. Maybe throwing $20,000 a month at … My numbers keep getting bigger here. Maybe throwing thousands of dollars a month at SEM advertising isn’t a good idea if you could be allocating that money towards a new website, and then all the money that you invest after you have a good website is now so much proportionally better than it was when you were sending people to the old website, that is more profitable to make that decision than it is to keep throwing your face at the same thing and expecting different results, which is the definition of insanity.
The point I want to make here is that every industry is different. What works for your competitors or what doesn’t work for your competitors, it may have different results for you. Just because you’ve heard, “Oh, my competitors are doing remarketing,” or, “My competitors are showing up number 1, number 2,” it doesn’t mean that they’re making money off of it. You have to analyze what works for you. If remarketing is fantastic for you, then invest money in it. I mean, it’s pretty straightforward. I always say that we let the data speak for us, so if we see that we’re getting good results from these things and we have statistical significance proving it one way or the other, then adjust your resources appropriately.
Some of the things here about every industry being different, your campaigns that you have versus the competitors, they may have different quality scores, different keyword strategies, they may approach … I mean, we have ways of spying on what competitors are doing and trying to mirror strategies that they’re doing to see if they work for us, but the truth is is that if let’s say your competitors have a quality score of 9 because their SEO and their LPO are awesome, and let’s say our SEO and LPO aren’t … Maybe they’re not 9 out of 10, but they’re like 6 out of 10, so we’re still working on it, but it’s not perfect. It’s also not terrible. That makes a big difference because something that might work with a quality score 9 out of 10, which lowers the cost per click of the clicks that we’re paying for, it might work for the competitor, but it might not work for us.
Also depending on our budget, we might want to allocate those resources differently. Another thing again, just reiterating here, the website, that can also affect quality scores, conversion rates, et cetera, so that can change how we analyze our results, as well. Now, SEM advertising, I draw a lot of SEM advertising and gambling, because we’re putting money in. We’re essentially sitting down at the blackjack table and we’re hoping to walk away with more money. Now, the good thing about SEM advertising, the odds, if we’re doing it correctly, the odds are significantly more in our favor. Now, the reason that we emphasize testing is because, and this is a big huge metaphor for gambling, but different tables, like blackjack or poker or craps or whatever, different tables in SEM advertising are going to have different odds at which, or different rates of profitability.
If remarketing is really effective for you, then you would want to sit down at that table as much as possible, i.e spend as much money as possible at that table because it has the best return for you. Now, there’s always diminishing returns. We can’t just invest a trillion dollars into remarketing with a 4:1 ROI and expect to get 4 trillion dollars out. As you invest more and more and more money into something, you are going to get less and less ROI out of it. That’s just the law of diminishing returns. Anyway, it’s a metaphor that I use for gambling, but the point is that again, nothing is guaranteed in this business. You could throw $1000 in, get nothing out. You could throw $1000 in, get $10,000 out. That’s why I draw those parallels between gambling.
The most important thing is that we’re analyzing our data, which I why I made this slide called “All About That Data.” Testing is so, so, so important, and I want to go over some of the core things that we want to be testing in our SEM campaigns to make sure you’re analyzing the correct variables when trying to identify what issues within the pillars of online advertising could be bottle necking the success of your campaigns. The first one here is platforms. We always want to test different platforms. There’s going to be different demographics on things like AdWords or Bing Ads or LinkedIn or Facebook. Again, Bing Ads is generally for example, an older demographic. They generally have more money.
There’s not as many people on there, so the impression share is lower. AdWords is kind of the inverse of those. It’s usually younger demographic. The impression per share is much higher, there’s a lot more people using Google. Linkedin is more business oriented. Actually, it’s a lot more business oriented, but yeah, depending on what your industry is, if you’re selling apparel, that’s going to be a very different demographic that you’re trying to target than if you’re trying to sell cloud software. It’s very different. Cloud software, you might want to be more oriented towards LinkedIn, because they’re more business oriented people, maybe those who are looking for cloud software more, whereas apparel, maybe you want to be looking on Google, because you’re trying to target younger teens who are interested in fashion.
Again, that’s all theoretical. It sounds like it makes sense, but the truth is we would need to test it to know for sure. The next thing we would want to test is campaign types. Search Network and Display Network are going to be the 2 biggest ones that you’ve heard of. Going further into Display Network, the most common form and I would say also the most effective form of display advertising would be remarketing or retargeting, they’re the same thing. I see them used interchangeably a lot, but yes, remarketing and retargeting are the same thing. Another type of campaign would be a shopping campaign. Again, going back to the apparel example. If you have individual items like that that you’re trying to sell, you can setup shopping campaigns, so setting up these campaigns, testing if they work, gathering statistically significant data to tell us if they work or don’t work, and then adjusting our resources appropriately.
That’s just the process of testing, and we need to be doing that to develop these, again, to build these online advertising campaigns. We don’t just throw a trillion dollars at SEM advertising and all of a sudden, just magically know that shopping is great for us or not great for us. We have to test that stuff and find out which takes time, and also takes money. Next thing here, A/B split testing. The main thing that we A/B split test is ad copy. The thing that we’re looking for the most in ad copy is higher click through rates. If we can find the line of text in our ad that works better than other lines of text, and gets us a higher CTR, keep in mind that click through rate is Google’s most dominant variable within their algorithm for determining quality scores.
Favoring ads with higher click through rate will not only get your ads clicked on more, but it’s going to give you higher quality scores which is going to give you a lower cost per click, which is going to give you a better cost per acquisition. The simple point of that is you’re going to make more money if you favor click through rate when you’re doing A/B split testing on your ad copy. Some other things we A/B split test are landing pages, different calls to actions, yeah, I’ll move on here. The last thing we want to talk about with testing is statistical significance and making good decisions. Ask yourself is the data that you’re reviewing accurate or even correct?
You need to have statistical significance. If you’re looking at an ad that’s been running for a month and let’s say you’re A/B split testing the ads so you’re looking at 2 ads and there’s only 20 clicks on both of the ads, that may not be enough statistical significance to tell us that one ad is better than the other. Even if the click through rates and the conversion rates are drastically different, 20 clicks is not enough. I’m not going to go in depth about where those thresholds lie in terms of statistical significance, but it is something that depending on … Because it’s going to be a case by case basis. It’s dependent on a lot of variables so that’s why I’m not going into it, but if you don’t have statistical significance and the data that you’re looking at is not accurate, is not correct, then obviously making decisions off of that data is not a great idea.
The second point here, or the question to ask yourself is are you measuring the correct variables? This is really important. Vanity metrics versus performance metrics. If you’re looking at the wrong data, for example, if you think that clicks is more important than revenue, or I mean, if that’s the variable that you’re looking at, like if you’re saying, “Oh, well we got 30 clicks this month, and we only got 25 clicks last month,” that doesn’t necessarily mean that you made more money. I actually wrote a blog called “Understanding Your KPIs,” that goes more in depth about this but you need to make sure that the metrics that you’re looking at are the ones that are actually affecting the revenue that’s coming into your company so, “Is this campaign generating revenue?,” is the question that you should be asking yourself when you’re trying to analyze the performance of it.
Then, last here, are you analyzing/using your data effectively? On the same note, so let’s go ahead and move on here. The next thing I want to talk about is budget. The reason I want to talk about budget is that budget accelerates time. When I say that these things are not bought, they’re built, or that campaigns are built, not bought, did I say that correctly? I think I did. If we have more money to allocate towards a campaign, we can obviously accelerate the rate at which we gather data, which allows us to make decisions more quickly. The first note that I have here is that the very high majority of companies will have limited advertising budgets.
Now, that means that you need to understand your limitations, and it also means that you need to understand that if you throw a certain amount of money at an advertising campaign, that we’re only going to be analyzing, or be able to make decisions as fast as we spending money to gather data to make those decisions. If we spend twice as much, and let’s say that you spending twice as much gets us twice as much data, so twice as many clicks, twice as many impressions, et cetera, then we can make decisions twice as fast.
Now, the problem is that there’s diminishing returns and we throw more and more money at something, the trillion dollar example that I gave earlier, so when you throw twice as much money at something, usually you don’t get twice as much out, because you are going to have those diminishing returns. I have a little graph here that shows that the more time we wait to gather data, the more statistically significant our data is. The third variable that isn’t in this graph because it’s not a 3D graph would be money. If we can have more money, it essentially accelerates time because we can get data faster, with that in mind, diminishing returns.
The final variable here that goes into budgeting is cost per click. This is going to be dependent on your industry. I also say that it’s pillar dependent here. If your website is terrible and you’re getting bad quality scores because of it, then your cost per click is gig to be a lot higher. Again, that’s another variable you need to take into consideration. If you have a really high cost per click, let’s say $10 versus $5, you’re going to be gathering data half as fast because your cost per click is really high, your quality scores are really bad, et cetera. All right. Then, the last point I want to make here is analyzing your SEM management.
One of the frustrating things about managing SEM campaigns is you can, as a technician for these SEM accounts, we can setup a perfect account and still not have it be profitable. What I mean by that is we can follow all of the best practices and we can be doing all of the best things we possibly can with the budget and limitations that we have. It doesn’t necessarily mean that it’s going to be profitable. I want to go back to the first example I gave about the foundational pillars, about your landing page optimization, your SEO. If the coding on your website is terrible, and the website itself just aesthetically looks awful, it doesn’t matter if me or any other AdWords technician sets you up a perfect campaign or a terrible campaign, it’s not going to do well if you don’t set it up for success.
If you haven’t considered those other foundational pillars of online advertising, then your accounts may be suffering for reasons that are outside of just how good is your SEM rep. It’s good to consider those other variables. Another thing to ask yourself is, are you investing the correct amount into search engine marketing? Some people are investing too much. They’re spending too much on it because they should be investing that money into one of the other pillars, so LPO or SEO. They could be investing too little. They might be investing let’s say like $1000 a month, but if their cost per click is $25, what’s the math here? Kind of embarrassed I can’t do that off the top of my head.
1000 divided by 25. That’s only 40 clicks a month, which I mean, that alone is not statistically significant, let alone if you’re doing A/B split testing, you’re splitting that amongst multiple campaigns and ad groups. Too much, too little. It’s good to ask yourself if resources should be reallocated more appropriately to gather that data at a rate that is healthy for your company, and then the second note I have here is regarding diminishing returns. That goes into the “Are we investing too much?” If your impressions is already capped out and you’re just spending tonnes and tonnes of money to show up in top positions but those top positions aren’t necessarily more profitable, then you might be running into an issue with diminishing returns where you’re paying too much for what you could be doing.
The next thing, a question to ask yourself is, “Have you tested or analyzed everything?” I went over this a little bit already with these 3 points of is your data accurate or correct? Are you measuring the correct variables? Are you analyzing your data effectively? Again, I just want to reiterate that what you’re doing with your data is extremely important because a lot of people are running these campaigns, and then 6 months later they’re like, “Well, why is it not working?” A lot of the times the reason is because you’re not drawing the correct conclusions from the data that you have in front of you. Other times, I mean, it can be other things, too. It can be that you didn’t have conversion tracking setup, in which case, you essentially don’t have any data. I guess that kind of is the same thing, but yeah.
The last final piece of advice here, because this video is dragging out here, the final piece of advice that I have is don’t get ahead of yourself. Educate yourself, understand your limitations, in terms of with the budget statistical significance, how fast you can gather, at what rates are healthy to gather that data for how good your other pillars of online advertising are. Test your advertising campaigns to see what works best for your business, and then adjust appropriately, and remember that these advertising campaigns are built, they are not bought. Obviously we can throw in more money at them and it goes faster, but we need to be testing and analyzing this stuff properly, which isn’t something that can just be solved with throwing a tonne of money at it. You need to make sure that you’re executing these strategies properly. That’s all I have for you today. I hope you guys like the video, and thanks for watching, and I’ll see you guys in my next vlog.