Episode 225 | How to Run a Paid Advertising Campaign

Show Notes


[00:00] Mike: In this episode of Startups For The Rest Of Us, Rob and I are going to be talking about how to run a paid advertising campaign. This is Startups For The Rest Of Us, episode 225.

[00:07] Music

[00:15] Mike: Welcome to Startups For The Rest Of Us, the podcast that helps developers, designers, and entrepreneurs be awesome at launching software products. Whether you’re launching your first product, or you’re just thinking about it. I’m Mike.

[00:23] Rob: And I’m Rob.

[00:24] Mike: And we’re here to share our experiences to help you avoid the same mistakes we’ve made. What’s going on this week, Rob?

[00:27] Rob: We had a nice tip from Josh Ayernov, and it’s about using Todoist, and email. This ties into a discussion you and I have had a few times, where I mentioned that I would love to be able to, basically, reorder my Gmail email, and make it a “to do” list. And he says, “Hey guys. Love the show. I’ve been using Todoist lately, and they have a Chrome extension, which lets you add a task from email, and append the email to the task as a url that goes back to that email. So if I have an email I need to respond to – that I need to postpone – I can just make a task, and click on it. So it essentially integrates the two. I figure if anyone is using Todoist, this would be a good tip, and if you’re interested in, kind of, being able to reorder your Gmail inbox, it could also be helpful.”

[01:08] Mike: That sounds really cool. It sounds very similar to what you’re, kind of, doing manually right now, with Trello, where you, kind of, send things over –

[01:13] Rob: Exactly. Forward it over, and then come back and search. Yup. Which is a little bit cumbersome – it works for me – but I can see, if I was not using Trello, I would consider using something like this. We had discussed the new EU VAT laws a couple of weeks ago, and Ryan Delk from GumRoad dropped me a link to an email. We’ll link it up in the show notes. It’s actually a blog post, and it says, “How we’re handling VAT at GumRoad.” The nice part is if you are using Gumroad, they basically handle it all for you, so you don’t have to worry about it. If you want to hear more about that, you can check it out in the show notes.

[01:41] Mike: Very cool. So this past week I went to a local photography studio, and had some new photos done. I took the time to update my social profiles – clean shaven, a little bit lighter, and no glasses.

[01:52] Rob: Yeah.

[01:53] Mike: It was like $130 for the sitting fee, and then $5 for each photo that I wanted. There were 10 that looked good, so I just grabbed 10, in different poses and outfits, and stuff like that.

[02:03] Rob: Yeah, every couple of years it’s nice to get a couple of head shots done. I’ve actually done it – the last couple of head shots I’ve gotten, one was when we did a big family photo shoot. We hired a photographer to come out to our property and do different location shootings here, indoors and outdoors. And I said, “Hey, can you just do a head shot while you’re here?” Then the other time was when I recorded that video course on how to hire a VA for your startup, which is startupvacourse.com. So, it’s nice to have a couple of different ones you can rotate.

[02:29] Mike: I figured I’d have something professionally done, that I can actually use in more of a professional manner.

[02:34] Rob: Right. So what are we talking about this week?

[02:36] Mike: Well, today we’re going to be talking about how to run a paid advertising campaign. But I don’t think we’ve ever actually done an episode on walking through what it is you should be doing, how do you compare different platforms, what sorts of things you should look at when you’re going through that, what things you should document. Things to pay attention to. Things to avoid. And I thought that what we should do is we should walk through a lot of those things so that people understand, kind of, the process that we go through, and learn from that, and maybe offer some insights back onto the comments of the podcast. Share the things that we’ve gone through, and what things we pay attention to because we thing they’re important.

[03:11] Rob: I think our intent is to make this, A, a little bit entertaining – so that you actually want to listen to it. And B, keep it high level enough that it’s something that is like “evergreen” content, so that you could come back to it in the future if you wanted. But it also gives you, like the mindset – if you’re going to start embarking on driving paid traffic – it gives you the mindset of how to, kind of, conceptualize all of this.

[03:31] Mike: So I think the first step in running a paid advertising campaign, is to know what your target cost of acquisition is. If you don’t know what your cost of acquisition is, you kind of need to guess what it is that you’re going to aim for, and use that as a guideline. Because when you start advertising on these different platforms, the cost of acquisition is going to vary dramatically between the different platforms. So, know what your target is, before you even start running the advertisements.

[03:57] Rob: I’m going to give you a couple of rules of thumb about how I would calculate – or ballpark – what I would use for my cost to acquire a customer. The first I’d start with is to figure out what is the average lifetime value of a customer. I’ll throw out three scenarios. One is that you have a one-time purchase product. Let’s say you have an ebook, or you have a WordPress plugin, or some downloadable software you charge “X” dollars one time for it. Now, some people may buy that, and then buy other things from you, but if you don’t have a bunch of data – that are easily at your fingertips – I would just ballpark it, and say, “My lifetime value is going to tend to be just the dollar amount someone would pay me for that.” Right? So if I sell four different pieces of software, for $40 each, I would say the lifetime value is going to be $40. Because I’m going to just say – unless I know for a fact “X” amount of people go and buy multiple items – I’m going to say, “My one-time purchase is $40.” And I’m going to start with that. If you have an e-commerce web site, I would just look at my average order value. Again, if you have the data, and you can go and say,

[04:55] “Well, this many people re-order, on average.” And if you can calculate it, that’s fine. But if I was going to ballpark it, what is your average order value. And if you’re a subscription service – so let’s say your SAAS, or your membership web site – I would tend to ball park – membership web sites tend to between four and six month, lifetime. So you can – to be on the low side – you might want to say, “Hey. It’s four times my monthly fee.” And SAAS tends to be between five and ten, until you’ve really hit product market fit. So again, if you want to be on the low side, and be more conservative, you can say five, and ten is not unrealistic for a reasonable service. So once you have that number – and if you’re selling a $19 ebook, then your lifetime value is $19. And if you have a SAAS app that’s $30/month, then maybe it’s $300 for that lifetime value. In general, you want to turn a profit, right? You don’t want to break even unless you have a back-end. And so, if I was going to be trying to scale up a SAAS app eventually, I would probably say, “To acquire a customer, I don’t want to spend any more than one-third of my lifetime value. And if I was going to try to acquire someone for a one-time purchase, you know, maybe you could creep up to 50% or 60%. But you don’t want to 80% or 90%, because then your profit margin is really, really low – unless you are trying to build that list and sell more things down the line. That’s a quick summary. There’s obviously a very large topic, to figure out what your target cost per acquisition should be. But that’s how I think about it conceptually.

[06:17] Mike: Yeah. I mean, you want your cost of acquisition to be as low as possible, but still at a point where you’re going to be able to make money. I think the other point to drive home here is, just because you have a target cost of acquisition doesn’t mean that over time you’re not going to be able to reduce that by optimizing some of your different ads. Because when you first start out, you don’t know what you’re doing. And that’s part of why you’re testing some of these different platforms and strategies. And when you do that, over time you will naturally get better at it. Or, at least, hopefully you will get better at it. Sometimes the platform makes changes, and they basically screw you over. Or there’s a lot of competition that comes in, and there’s really not a lot that you can do about it at that point. Your long-term goal is to get better at it, and thus drive your own cost down.

[06:58] Rob: Yup. The rule of thumb I use is, if I run ads and send them to a landing page, and do what I’m trying to do, and I can get within 3X of my target cost for acquisition, then I continue, right? Then I try split-testing, and optimizing ads, and doing other things. So that means if I can pay $50 to acquire a customer, and the first time I run an un-optimized campaign, it’s $150 or less, then I’m within target. But if I run it, and it’s $500 to acquire – meaning it’s 10X, it’s obviously possible that you could optimize into that, but that’s where I tend to just back out and say, “Boy, this channel is not going to work for me.”

[07:36] And I think this actually begs the question of like, “Well, what type of lifetime value do you need, in order to be able to support paid acquisition?” And it depends – B2C, B2B, and all of that stuff – but, in general, what I’ve seen is I don’t know anyone who’s making paid acquisition work, profitably, that doesn’t have a lifetime value of around $100. And so, if you are selling a $20 or $30 ebook, that’s hard. If it’s out of an ecosystem, you know? If it’s not a loss leader. If you’re actually trying to turn a profit on it, it’s pretty tough with a $20 or $30 lifetime value. There are exceptions. Patrick McKenzie with “Bingo Card Creator” has done it. He’s also very, very good, and he’s spent a lot of time – he’s one of the best people at AdWords that I know. If you are in a niche where the clicks are at all expensive, then you are going to be paying, easily, $100 or more to get someone to come and try your app, or turn into a customer. And so that’s where low lifetime value stuff doesn’t tend to be ideal for paid advertising. And you need to look at things more like SEO, content marketing by rally or word-of-mouth – kind of, the other marketing approaches.

[08:41] Mike: So once you know what your target customer acquisition cost is, you have to start identifying the platforms you’re going to use. And there’s a lot of different platforms out there, and it also depends on the specifics of whether you’re going after desktop users or mobile users. But, you know, there’s also like the standard ones that, kind of, cross into both mobile and desktop – like Twitter, Facebook, AdWords, Buy-Sell Ads, and a number of other ones. I mean, Rob, I’m sure there’s a list that you have taken a look at, and have leveraged as well.

[09:11] Rob: Yeah. Boy, I had a list that was like a mile long, when I was really hammering on this stuff with Hittail. And I ran on all types of crappy B2C networks. It’s like, there’s one called Chitika. To be honest, I don’t even remember the names of the other ones, because the ROI was so bad. I mean, I would pay $50 in ads, and you’d get all these 10 cent clicks, and no one would stay on your site more than three seconds. Like none of them. And so, instantly, that was just a non-starter, because I knew that if I spent $500 no one was going to stay on the site. There are a limited number of ad channels that you can think about if you are in the B2B space. You mentioned several of them. I would add LinkedIn. I would add AdWords, even though it’s very expensive – it can be expensive. The Bing ad network is reasonable. YouTube is actually something I’m doing right now, and I’m experimenting with. And then the other channel is re-targeting, which kind of spreads across all of these channels. But re-targeting is, while different from what we’re talking about – we’re talking about driving brand new traffic right now – but re-targeting is something I would want to have in place before I ran these paid ads. Because re-targeting gives you the chance to then advertise to them again, and your conversion rate on those is going to be a lot better.

[10:22] Mike: Yeah, that’s a good point. But the one thing you do have to be careful about with re-targeting is sometimes they have policies around, you have to have a product that people can reasonably buy. And sometimes if you’re sending people directly to a landing page, so you can’t buy from there, sometimes they get a little antsy about even allowing you to use their re-targeting capabilities on landing pages.

[10:43] So the next step is to run a series of tests on each platform for a specified amount of time. I’m partial to basically setting aside $100 and running the test for a week. And that’s a very small test. I understand that it’s probably not going to be representative, but at least it, kind of, dips your toes into the water. It gets you, at least a little bit comfortable, with how that platform operates – what things you need to make those things work. You know, there are certain platforms, like Twitter and Facebook, you can put images into those advertisements. But when you’re doing that, the requirements for those images are different sizes. And they will appear differently to the users, based on whether or not it’s showing up on the desktop or on a mobile device. You have to play around with those things a little bit. And when you’re first setting these things up, it can take way more time than you ever thought imaginable to like just tweak your images, and stuff like that, to get them to show up inside the advertising platform. So it can be, kind of, nightmarish process – in terms of the amount of time that it takes – but you do still have to go through that process to figure out, what is it you need to do? And then be able to take those things, and write them down, and say, “These are the requirements for this particular network.” And then hand them off to a graphic designer. Once you get through this, if you decide to double-down on a platform, you can then take those requirements, hand them off to somebody, and say, “Hey, I need you to build me images that go with this, this and this. And these are the different sizes.” It allows you to repeat the process in a way that is a lot more effective in terms of your time, as opposed to the first time where you’re just trying to figure things out. You’re trying to get familiar with it, get comfortable enough to be a little bit dangerous. And then from there you need to start optimizing. But from there you don’t want to go through the process of trying to optimize before you even get started.

[12:27] Rob: I second your surprise at how long it takes to set these things up. Every time I do paid acquisition I think, “I’m just going to go in, by a few ads, set up a budget, and be out.” And it’s always hours later, because I wind up finding images, and crafting headlines, and running all the split tests, and doing all of that type of stuff. So give more time to this than you think you’re going to need. There’s not a shortcut, like any other marketing approach.

[12:53] Mike: Yeah, like three or four times as much time as you think you need.

[12:56] Rob: Yup. Exactly. The other thing I would do – probably before I ran a test, if you haven’t done much paid acquisition – is to pick a single channel, pick one of these. And I would educate myself. I would dive deep into it. I would probably buy a course, either on Mixergy Premium, or Udemy, or from AppSumo. I have taken courses on all of those. There are some really nice experts who give you some good insight on how to work the system. Or you can even find an expert on it. So like Amy Porterfield, is an example, is a Facebook ads expert. So I bought a course from her on how to improve my Facebook ads. Or if you find a good blog, or a good podcast, on the topic, typically they are monetizing that by selling a paid course on whatever topic you want – Twitter ads, of AdWords. That’s what I would do, and that’s what I have done. So when I started really doubling down on Facebook ads with Hittail, I bought like four or five Kindle books. I listened to several podcasts, and I bought some paid video courses. And I went through all of the material quickly.

[13:55] I didn’t spend weeks doing it. I mean, I did this all in, like, one day. I skimmed through the books. I tried to find out the things that people were saying in common, and the differences. And you can find yourself almost becoming an expert by proxy, if you just immerse yourself quickly, and give yourself like an instant PhD in that technology, just by hearing a bunch of stuff. Then you go and use all of the best practices that you’ve developed, and then you hammer on a test. Now, you said do a small test, $100 in a week. I tend to want to move faster than that, and I will throw more money at it. I might budget $100 a day, and sit there and watch it. Because I want to know quickly if this is going to work or not. And if I get four or five days into it, and I’ve spent a decent amount of money, I can start to tell where the trend line is going. When I’ve made paid acquisition work in the past, it typically is an investment of a couple thousand dollars before I’m starting to become confident in my ability to execute on it. And sometimes it’s more than that. But that’s, in my head, kind of a mental marker that I use of like, “Yeah, I might get a few grand into this before I really know if it’s going to work.”

[14:58] Mike: Got it. Yeah, see, I run them, kind of, in parallel at the same time, on multiple channels. Maybe at some point I would get a little bit more confident and, say, dump $100 today into like three, four, or five different channels at the same time. But I also found that trying to flip back and forth between the different ones, and keep track of which ones were performing, and which ones weren’t. You could easily rack up a pretty hefty advertising bill, in just inside of a week, if you’re paying $100 a day on five different channels.

[15:25] Rob: That’s the thing. I only test one channel at a time, because I need to be an expert on it. Like, I can’t become an expert on five channels, and then try to run them parallel, or else I’m running around like a chicken with no head. When I do a channel, I immerse, I focus on that channel for maybe a solid week. And I don’t really do anything else, except for – I mean, I do my other projects and stuff – but I don’t try to scatter myself. Then if that channel doesn’t work, I can move on to the next one. If it does work, then you can start trying to figure out that repeatable process of, “How am I going to run this long term?” Because I’m not going to spend 20 hours a week on this every week, even though this week, getting it set up and monitoring every – I mean, I would literally monitor it every couple of hours. Like, log in, look at it, see what it’s going to do, stop ads, start new ones. You know, I was really into it. Again, I’m not saying this is the only way to do it. This is how I do it. I like to dive into things anyways, and be pretty intense about them, and find out very quickly if it’s going work or not, rather than taking a more hands-off approach that I think you’re talking about.

[16:20] Mike: Yeah. I think it’s interesting that you have a different approach to it than I did. Where what I did was, as I said, I would run several of them in parallel, and then after that week I’d basically look at that channel I thought was performing the best, and essentially double-down on that one, and optimize it. So I ran through an ad, and I compared Twitter against Facebook ads, and found out that Twitter was performing about four times better than Facebook was. And then I went in, and I optimized the Twitter campaign, and I essentially doubled conversion rate on the Twitter campaign. So at the end of the day I basically made eight times better than what I was getting on Facebook. Clearly, there’s different ways of doing it. I think it depends on how much time you have to play with, and how much your budget it, and how quickly you want to move forward with the paid advertising.

[17:04] Rob: Yeah. That’s a good point. There was a point where I ran through, I think, about twelve ad networks, in the span of maybe three or four weeks. Which means I obviously did do some of them in parallel. And what I did was, the bigger ones that had more information about them, and that I just had more confidence that they would work. Those were the ones that I dove into, and I only did those one at a time. Right? So that was like AdWords and Facebook. Whereas the ones – like the B2C ones I was talking about, 7-click and Chitika, and these kind of weird things, and Bing ads – I would set them up and let them run in parallel. Because it’s not like there’s a way to become a master at those. You just, kind of, had to throw it out and see what was going to happen. So I did do more parallel stuff with the things I didn’t think were going to work as well.

[17:48] Mike: Yeah. That totally makes sense. So the next step in this process is to document everything that you’re doing, so that you’re able to refer back to it later. I use a combination of things like screenshots, and spreadsheets, and Google Docs – pretty much anything that I need to be able to come back and recreate it, or ask questions in the future. I do that for a couple of reasons. Partly, is just so I can acquire information, and set it to the side so it acts as a true snapshot, and partly because I don’t necessarily trust the platforms to not change underneath me. So if I need to find something later on, I can do it. And some of them actually make it very difficult to go in and tweak some of your settings. I remember back when Facebook was still making a ton of changes to their advertising network. There were things that would just break every other day. So, I kind of learned quickly to just take screen shots, and make it so that you could go back and do it again. That’s not to say that you need to document everything heavily. Just make sure that you’re taking screen shots, so that if you take a screenshot of like the demographic information that you’re going to be targeting for one advertising network, it’s a lot easier to just go refer to the screenshot, than it is to go there, log in, and the try to compare it to another network that you’re trying to advertise on.

[18:59] Rob: Yeah, I haven’t done a good job at this step of documenting things.   I typically leave it in the system where it’s in. You know, if it’s in the Facebook ads system, I’ll refer back to it, and go and try to look for trends. And I’ll graph stuff, and I do review the data, and see who’s clicking, and optimize and stuff. But I don’t think that I have a single, external doc that documents anything, that is shared with anyone. Because I’ve pretty much done most of the ads stuff myself. Now, with that said, if I was going to be outsourcing this, or even having someone in my team do it, we would need some kind of a way to communicate and document the tests we were running, so that we could share our findings, right? There’s this concept of a “growth spring”, and it’s similar to the “coding sprints” with SCRUM. It’s like a one to two week, or it can be a 30 to 90 day, thing of trying to grow your company by using a single tactic. And so if you do a growth spring with Facebook Ads – however long that takes – you really want to document the experiments you’re running, and what you’re learning from them, so that you can share them with others. To date, like I said, I have not been very good at this, and I’ve kept it all in my head. So if I was going to pass it on to someone else, I would have a bit of documentation to do beforehand.

[20:09] Mike: Well, I think that also, kind of, goes back to the difference between how you have done it, and how I’ve done it. Whereas you do that really hard, deep dive into one advertising network, where I’ll step back and do things a lot slower, and compare the networks against each other. In that case, I kind of need those screenshots in order to make it easier for me to be able to compare between the two. Versus, if you’re kind of going deep in one particular network, it’s not as important, because you’re tweaking within that network, and you typically have all the information right there at your fingertips, and you can just look at it. It depends on how much you think you need to document some of that stuff, too.

[20:46] The other thing I like to do is compare the results from one platform against each other. And I think that the best way to do this is to make sure that your advertising is as similar between the different platforms as possible. But do keep in mind that there are going to be differences between advertising on LinkedIn versus on Facebook and on Twitter – partially because of intent, but partially because of the audience as well. So you might say one thing to the people in LinkedIn, and something different to the people on Twitter. And another thing I like to do – as I said before, comparing against like Twitter and Facebook – double-down on what’s working, or try to optimize it. I found that on Twitter I was able to do four times as well as Facebook right out of the gate. And I found that using the Twitter cards, for example, I could essentially double my conversion rate, by essentially eliminating steps by forcing people to click on the advertisement, and then go over to a landing page, and then enter their information. With the Twitter cards it’s literally one click. They can just click a button that says, “Learn more, and send my information over to Mike Taber for what he’s working on.” And I’ve found that by getting rid of those extra steps, it doubles my conversion rate on that.

[21:55] Rob: Yeah. The thing to keep in mind when you’re doing paid acquisition is that it’s a lot harder than you think it is. You don’t just set up and ad, and instantly have positive ROI. It takes work. It takes optimization. It takes comparison. I mean, we’re recording a whole episode here, and we’re barely touching the surface of the amount of, not only up-front optimization, but the ongoing work it takes to run an ongoing campaign. You know, because it adds burnout and that kind of stuff. So keep that in mind – as you’re embarking on this – that this is… The nice part about paid advertising is it’s very quick, right? It’s like instant ramp-up, instant ramp-down like a faucet you can turn on and off. And it scales really well, if you can make it work, because any of these channels that we’re talking about are pretty highly scalable. But the downside is that it often will not work for you. A lot of channels just won’t convert, and you will spend some money that you won’t get back, and you won’t have a positive ROI on that.

[22:49] Mike: So let’s talk about some general thoughts on paid advertising. One thing that you just mentioned was “ad burnout”. Why don’t you expand a little bit on that, on exactly what ad burnout is first.

[22:58] Rob: Sure. The idea is that ads have a certain lifetime, and over that lifetime you’ll see this gradual decay of click-through rates, and of results, because people basically get blindness to your ads. So if you use the same image for months at a time, over time that click-through rate will just naturally fall. So you can’t just – as a rule – post an ad, and expect it to work forever. There are some exceptions to this. AdWords is one where you don’t tend to need to renew your ads, because it’s intent-based, right? It’s search-based. So the people who are seeing it tend to be new people all the time, and that’s a good thing. If you’re on Facebook, and you’re targeting a demographic, then someone who’s interested in “email marketing” will see that same ad over and over, and over time – even if they’ve clicked it once – they’ll become blind. Your audience is so big, and you need to change it up. So that’s the idea. “Ad blindness”, “ad rot”, “ad burnout”, it’s all the same idea. It’s that the performance of your ads will decay over time.

[23:56] Mike: Going back to what I talked about, in terms of documenting things, one thing that I’ve seen is that the ad networks that I’ve worked with don’t tend to track what your conversion rate is over time. Because of ad burnout, you kind of have to do it yourself. Because all they do is they give you this raw number, and it’s within this snapshot of time, or it’s like the total amount of time that you’ve been running the ad, and it will just be a number. So, if you’re not paying attention to what that number is, yesterday, and the day before, and the week before that, then it can slowly drop over time, and you don’t notice. So that’s something else that you, kind of, have to keep in mind. As I said before, in terms of being able to drive people to a landing page, it’s very helpful. And if there’s any way to bypass that process, or shorten the entire time that it takes in order to acquire somebody’s email address, then that’s definitely an avenue that you should explore and look at. I mean, like I said, on Twitter I was able to double my conversion rate just by using the lead cards, as opposed to sending somebody to a landing page, and then they have to enter their information.

[24:59] Rob: Right, and Google AdWords has that same capability, where you can capture an email right there in the ad. I’ve never actually tried that, so I don’t know its effectiveness. But I imagine that if someone is interested, it’s a decent way to go.

[25:10] Mike: Another option you have is that if you have an email list already. So if you have an existing product – where you have customers, and you’ve got information about them, like their email address – then you can take those email addresses and upload them to some of these ad networks. What they use is something called “related audiences”. Essentially, what you do is – you’re going to be able to target people who are like that list of email addresses that you uploaded. I know that Facebook does this, Twitter does this. I don’t know if Google Ads does this or not, but –

[25:40] Rob: I don’t think it does.

[25:41] Mike:– what it does is it allows you to target other people, for advertising, who are like your current audience. And that’s extremely helpful.

[25:48] Rob: Yeah. So it’s similar to re-targeting, right? Where re-targeting, you are essentially targeting based on the action that somebody has taken – which is visiting your web site. In this case, you’re targeting a specific group because they are demographically similar to your audience – to people who have actually signed up for it. And “demographically similar” I’ll put in quotes, because that’s Facebook’s judgment call, right? They have these algorithms – so they have this social graph, where they can link people together and see similar folks. So it’s, kind of, a black box for you, but you get the recommendations based on that. I’ve used the “related audience” stuff on Facebook. I have not had luck with it, but I have heard that it is doing very well for some folks.

[26:32] Mike: It hasn’t worked nearly as well for me, but I do know some people who are making it work extremely well. Something else you can pay attention to is, any sort of related information you can get from the different advertising networks. So if you can get age, or location, or demographic, or platform information you can zero in on those particular things, and some of those are going to convert better for you. I haven’t figured out exactly why that is. So, for example, I’m running a Twitter advertising campaign right now, and the people who are using Android devices are converting more than twice as well as people who are using IOS devices, which is kind of bizarre.

[27:10] Rob: I’ve used this quite a bit, especially on Facebook. I found out that there was a certain age bracket that converted better. Absolutely. There are like 14 states in the U.S. that convert way, way better – like two or three times better than all of the other states. I didn’t notice anything with platforms, in particular, like you said. But you can get these out of the Facebook Ad platform, once someone clicks on stuff, once you run some ads, and then you can go back and see who clicked on it most, and narrow your demographics. Now this means you can’t scale this ad up, as well, because you are starting to exclude people. So you won’t get as much traffic, even once you ratchet your budget all the way up. But those will be your highest ROI clicks, and so early on that’s what you want to shoot for. Then later on, if you are exceeding your ROI, you can always dial those back a little bit. You can widen that age range, or you can go to those states that maybe don’t convert as well, when you’re really trying to scale it up and drive the maximum amount of traffic.

[28:02] Mike: So, another I’ve found is, do not trust the metrics that they’re giving you. You have to take anything that they’re telling you with a grain of salt. Especially when it comes to Facebook and Twitter, because they use what’s referred to as an “engagement”, which is not necessarily a conversion for you. The other issue that I’ve found is that their idea a lead, and yours, can be wildly different, depending on whether they’re counting it just based on a click, or all the way through your acquisition funnel to get an email address. Also whether or not they’re charging you for that lead, or not. There are some ads that can be shares, for example, and depending on whether or not it was showed to somebody – and then they clicked through it, and you were charged for it. Or if they shared it, and one of their friends clicked through and signed up for it – you might be charged for the first one and not the second one – and that skews your numbers. So be really, really careful about how you’re looking at a lot of those numbers. And make sure you fully understand exactly what you’re interested in. Whether it’s email addresses, or impressions, or actual sales all the way through the sales funnel. It really depends on why it is that you’re running these ads, as to the specifics of what you’re looking at.

[29:15] Rob: Yeah, I agree with this one. I’ve seen some folks touting the Facebook news feed ads, over the right-hand side ads – because you get more clicks, or the clicks are cheaper, I guess is what it is. And the click-through rate is higher. But if you actually look at what a click means for a news feed ad, it means someone clicking any link on that ad – and there’s like five or six links. So if you’re trying to drive them off site to your web site, half of those links – or more – don’t go to your web site. They go to like your Facebook page, or some other random place – completely not useful for you, and not going to contribute to your conversions. But they count that as a click, and you get charged for it. Then, on the right-hand side ads, any link there is actually going to lead off-site to your web site. So every click there is a real click to your web site. So like you said Mike, these things can be misleading, and I hope they’re not being intentionally misleading, but it has felt like that to me, at times, where the numbers don’t really add up because of the way they’re defining certain thinks, like clicks.

[30:14] Mike: It’s complicated because your perspective, as somebody who’s buying these ads, is going to be different than their perspective, as somebody who is creating the mechanism for you to run the advertisements. If any of the advertising platforms reach out to you, and offer any sort of a free one-on-one, to help you improve your advertising, definitely take them up on the offer. It gives you the ability to ask them in-depth questions about exactly how things work, and you can start questioning the numbers that you’re coming back with – or that they’re coming back with – and ask them, “Hey, what does this mean?” or, “I saw this, and these numbers don’t add up. Why do they not add up?” And you can start getting an in-depth description, or explanation – directly from them – about why some of those things don’t add up. Sometimes it’s very helpful, and sometimes you look at it – and they’ll look at it – and they’ll say, “Yeah, we know. We’ve heard that a number of times. This is how you can, sort of, get around it” So sometimes they do have work-arounds for you, but it’s always worth it to take the time to follow up on those things, and be able to get in front of them the questions directly, so you can get answers directly from the vendor.

[31:20] Rob: Yeah. I would agree with that. They seem like they might be a waste of time, but I agree. If there’s quirks in the system, these are the folks who know it, and these are the folks who can alert you to it. I did let Google, at one point – they wanted to like optimize one of my campaigns. So they copied it, and asked if they could create an optimized version based on their “best practices”. And this was three or four years ago, probably. It was an AdWords campaign, and I ran theirs along with mine, and theirs was awful. It was horrendous. It was spending a ton of money. The budget was high. It wasn’t getting many clicks. It wasn’t getting conversions. I eventually stopped it. And I don’t know what the story was, because I genuinely expected them to have the knowledge to be able to do this better, but the campaigns I had running were far superior to it. So, I think in terms of that – which is super-specific advice – I think, take it with a grain of salt. And if they do run or optimize a campaign for you, you just need to watch and compare, like anything else. But just in terms of meeting with them – like you said – and getting advice on the platform and all that, I think that’s a worthwhile hour you can spend on the phone with them.

[32:18] Rob: If you have a question for us, call our voice-mail number at 888-801-9690. Or email us at questions@startupsfortherestofus.com . Our theme music is an excerpt from “We’re Out of Control” by Moot, used under creative commons. Subscribe to us in iTunes, by searching for “startups”, and visit www.startupsfortherestofus.com for a full transcript of each episode. Thanks for listening. We’ll see you next time.

Twitter Digg Delicious Stumbleupon Technorati Facebook Email

3 Responses to “Episode 225 | How to Run a Paid Advertising Campaign”

  1. Claus says:

    Hey, recently started http://www.cmsassist.io which offers Joomla support. Want to run some paid ads campaigns but don’t have the budget or time to try out all different providers. Anyone here that can recommend one to start for my audience (small biz running on Joomla). I’m thinking of Twitter or Facebook, but also open to google adwords, LinkedIn etc.

  2. Claus says:

    Cool. Thanks for pointing me in the right direction Rob.