PPC evolves fast, and each new evolution is the opportunity to tap into new avenues of growth.
In this video, RJMetrics and Elite SEM share some of the biggest opportunities emerging in PPC and discuss why you should consider giving them a try and how to start experimenting with them.
Tristan: Hello everyone. We are excited to have you today. Thanks for taking the time out of your busy schedule to show up at our webinar we have about 6 new PPC experiments that you should try today.
Before we get into our content, take a second to tell you a little bit about who's talking to you today so you can associate some faces with the voices you are going to be hearing. My name is Tristan Handy. I run marketing for RJMetrics, and I am a bit of a dabbler in PPC, we've done some of it here but I don't claim to be a super expert at it. I am joined by Mark Weisinger, who's the director of marketing at Elite SEM. Marc, you want to say hi?
Marc: Hey, everyone. Thanks for having here. Excited to be here to talk about some PPC experiments.
Tristan: Cool. So a little bit of housekeeping before we jump into it. Afterwards, if there's anything that sparks your interest during the presentation, and I certainly hope that there is, we're going to be doing a Q&A session where you can either tweet your questions and the hashtag you can see on every slide is #PPCevolves. Feel free to chat at that hashtag and we will be doing Q&A there at the end. You can also use the ready talk interface to chat us questions there at the end.
We're also recording the entire webinar and we're going to be posting that probably within the next 24 hours. We'll make sure to send you an email follow-up with the recording and the slides so that you can catch anything that you missed or forward it around once you're so impressed with all the amazing things we've told you today. To add a little bit of enticement so you stick around for the whole thing, not that you wouldn't, but we're going to be giving away a free iPad at the end to somebody who is nice enough to share the webinar on Twitter beforehand. Stick around for that, we'll be announcing the winner.
Marc, do you want to take it off?
Marc: Sure. Yeah, to really get started, a lot of what we're going to talk about is how PPC has evolved. When I first started at Elite around 10 years ago, I was the third employee at Elite SEM. SEM was really just search, there wasn't great display, there wasn't great re-marketing. We weren't doing a lot of mobile marketing, it just wasn't as big, there weren't iPhones.
PPC has evolved a lot, just as you see Homer evolving. Not only in the last 5 to 10 years but even the last 1 to 2 years, we have seen a ton of big changes where existing platforms, new campaigns, so we're going to walk you through a lot of these changes and really get you up to speed on what tests are most relevant to today's PPC world and walk you through a lot of those changes.
Some of the things that we're going to talk about today, we always hear mobile is so important, we're going to cut through the fat of that and really start to understand the big picture when it come to mobile. We're going to talk about exhausting your search ROI. For most of us, I'm sure search is a really part of what we're doing online. How do you get the most out of it? How do you maximize your results?
We're going to talk about supercharging your ads and your feed, so we're going to talk about Google shopping, and we'll just talk about some cool new opportunities for some existing platforms to really expand your reach and get into new stuff. So what are the newer tasks? What are some of the more cutting edge stuff outside of maybe your realm of what you're currently doing now? Ultimately, we're going to talk about how smart decisions based on data and performance, and smart testing is going to get you better performance and better ROI.
To give a quick introduction, a 15-second commercial about Elite SEM, so Elite SEM is a digital marketing agency. We specialize in SEM or paid search, SEO display advertising, re-marketing, conversion rate optimization and feed management. As you could tell, with PPC evolving, we have evolved to do a lot of things that we consider ourselves experts in. Additionally, one thing that we're really proud of and happy to tell is we won the best agency at the U.S. Search Awards, not only in 2013 but just recently last month so we're back to back winners. It's voted on by our peers, so other agencies and people in search marketing, so we're really proud about that.
Tristan, you want to tell us a little bit more about RJMetrics?
Tristan: Yeah, absolutely. Let me see if I can get on the right slide. There we go. RJMetrics, for those of you who may not be familiar, we're a complete analytics platform for online businesses. We help businesses make smarter decisions with their data and we're super proud to count many of the most innovative businesses online as our customers. First, you can see Hootsuite, Fab, NoMoreRack and a bunch of others.
We suck in data from a ton of different data sources. One of them being AdWords, and then we help, which is analyze that and grow faster as a result. We're going to be returning to that just at the very end on how you can use a platform like RJMetrics to manage your search better.
Last thing before we get into the heart of it, we wanted to learn just a little bit about who's on the call today so that we can target the content that we're going to be delivering based on the audience that we have. For those of you on the call right now, how do you manage your PPC program? Feel free to respond and then we're going to be sharing the results with everyone as soon as we've got the responses in.
Marc: Cool, and this is great. What we're going to be presenting is going to be applicable to everyone. Whether you manage a team that manages it or you're a part of an in-house team, all of these trends and experiments are going to touch upon what's changed, what's the latest and greatest in SEM, and how do we better manage it? If this is new to a lot of folks or this is more learning what you should ask your team on or you're going to manage these stuff yourself, it will all be very applicable.
That's great. It looks like about half of us are actually on the client side, managing it in-house. We have about 20% there using an agency so maybe we'll get some better questions for agency, and then 10% for someone else on the team manages it, and then 10% for I actually work in at an agency. That's pretty cool. Thanks. That's super helpful. We have one more question coming up later and without further ado, Tristan, if you're ready, I'll get right into the content.
Tristan: Yeah. Let's do it.
Marc: All right. Awesome. Again, we hear about mobile all the time. If you guys are going to conferences, if you on other webinars, you're hearing this is the year of mobile, mobile this, mobile that, it's enough to say that mobile is important but how do we really understand what's important, what we need to takeaway, what's something useful I can actually go back to my desk or tell my team that I want to do for mobile? So we're going to talk about how you best modify for mobile and understanding how we want to treat mobile, not only in terms of goals and experiments but how it works in today's PPC world.
The first thing that we need to talk about, and there's two things to really understand when we talk about mobile in terms of the landscape. The first one is what I like to call mobile micro conversions. This is a slide that comes from Google and I think it does a really good job of showing, for desktop, normally we're tracking one thing. So we're going to track our spend and then we're really looking at conversions, how many leads did I get, how many products did I sell, and that should make your metric that you're using. What was my cost for conversion? What was my conversion rate? Everything is really based around one goal or one conversion metric.
For mobile, we have that as well, so let's call that mobile commerce. How many people search for something on my brand, got to my site, filled that Elite form or purchased something? Right, we're still going to track that on mobile but there's so much more when it comes to the mobile landscape. We have people downloading your app, we have people looking for a store locator because maybe they are looking to walk into a brick and mortar store. We have people who call, so either clicking on a call extension, they click on a phone number on the site.
There's all these other events that are happening on mobile that are things that step number one is we need to measure correctly. We can't make the right decision if we don't understand what people are doing. If we're only tracking mobile commerce, and we're not using analytics or AdWords or an internal system to track how many people actually download my app, how many people call us, how many people are looking for something in store. Then really there is a lot more to do to set up mobile to measure it the right way.
We know that there are problems with mobile attribution and I'm hoping that we get some questions in the Q&A to really discuss that. The first thing we need to understand is with mobile, we have to start to measure things outside of just mobile commerce because people have different intentions when searching or browsing the web on their mobile device.
The second thing that we need to understand, and because a lot of people are actually managing this in-house, I'm hoping this isn't that new to that many people because this is a mouthful and this is hard to really get across in one slide but we're going to do our best. About a year and a half ago, Google launched what was called "enhanced campaigns". Previously to enhance campaigns, you could have one campaign for desktop, you could have one campaign for tablet, you could have one campaign for mobile.
That would give you the ability to have separate budgets for desktop, tablet and mobile. If you want to spend most of your budget on desktop and then only have $100 a day going to mobile, you had the ability to do that. You could bid and budget exactly what you wanted per each device type. So everything was really great, all the marketers were really happy.
About a year and a half ago, Google switched to what is now called enhanced campaigns. This is worth knowing outside of Google as well because Bing is now switching to this as well. What enhanced campaigns did was they took your desktop bid and budget and they combined that with tablet. Let's say you're bidding $5 on a keyword on desktop, you're now bidding $5 for that keyword on tablet. There is no way to switch the bids between desktop and tablet and it's one campaign budget.
If you're budgeting, let's say $1,000 a day, you're going to spend whatever traffic is out there for desktop and tablet, you're not able to budget separately between them. So it took away some of the ability to have a separate budget and maximize your bid, bidding lower or more for desktop or tablet based on what was more valuable to you, what was converting better.
Now, you're probably wondering, all right, what does this mean for mobile? Unfortunately, mobile is now a part of this big campaign. Let's say we want to also advertise our mobile for search. What you do is you would opt your desktop and tablet campaign that is now together, also into mobile. In your same campaign where you have one budget for desktop and tablet, now you also have to share that budget with mobile and this introduces us to a big theme or something that I really want to get everyone to understand, which is modifiers.
If you're bidding, let's say, $5 for desktop, the way that you actually put in your mobile bid is you use what's called a modifier. You're able to bid up or down based on what your desktop, tablet bid is for mobile. So you could bid negative 50% or plus 200%. If you're bidding $10 in desktop and tablet, you could take your bid on average and lower by 50% for mobile bids or higher for mobile bids. It sounds a little bit confusing, it is a little bit confusing, but essentially, you can't if you want to spend let's say $10,000 on desktop and $2,000 on mobile, you can't set separate budgets for these things. It all lives within one campaign and then you have to do some reporting and some digging to actually monitor how is my performance changing, how am I doing on mobile versus desktop.
That's understanding the landscape of mobile so we understand that we need to start tracking more things. We understand that enhanced campaigns and there have been some changes to the way that Google allows me to actually bid and budget for different device type traffic.
Before we understand our first experiment, we need to talk about the search engine results page and some of our goals for desktop traffic versus mobile traffic. In our example on the left, you will see a desktop search for Florida Gators t-shirts. I went to the University of Florida, say what you want about our football team, we're working through some stuff but who doesn't want a Florida Gators t-shirt, right?
If you do a search for this, our number one goal on desktop is traditionally to have the best CPA. You want to be as high on the page as possible but ultimately, you don't want to pay more for traffic than you can afford, so being profitable and having a good conversion rate and making sure your cost per order on search is really your number one priority. If you can get a higher position, great, but there's a lot of opportunities on the search page that you don't have to be in position one or position two. You want to bid what's effective.
When we talk about mobile, we have to optimize very differently. If you look at the mobile searches on page for Florida Gators t-shirt, you will see that there's only two paid with listings, the organic listing's only half of what's on my screen right there. So really for mobile, our number one goal in terms of priorities is we want to be in the top two positions. If we're not there, we're not getting any traffic and no one is going to click on our ads. So we're not going to spend any money, we're not going to be able to measure any conversions or micro conversions because we're not going to get any traffic.
Now that we understand those goals, and at least have a baseline idea of what our first goal is for desktop versus mobile, let's talk about the experiment. For the experiment, we want to download, analyze, adjust and repeat. Let me explain what that means.
In terms of our first experiment, what we want to do is we want to look at our campaigns to see where we're actually spending money and see how that varies by device type. What we do is in either our campaign or ad group tab, in the AdWords interface, we're going to download a report, we're going to include a segment or go to the segment tab and we're going to add device there. We're then going to download this so for whatever date range you want. For me, I chose this month.
You'll see we have two campaigns. One on the top, one on the bottom for ad group, one in two within one campaign or two campaigns, exact and broad match. You can see exactly what your bids are for desktop, what they are for mobile.
Again, it's worth noting, you can't just change your bid from $2 to $2.50. You have to change your modifier. But things that we're going to be looking to do is download our report, analyze it, look at the data. Am I in the top position for my mobile data? If not, we want to consider raising our modifier. Am I converting at a profitable CPA for mobile? If not, you're going to want to make a change so maybe you lower your bid. In instances where we're number one on desktop, we maybe could bid lower and spend less on mobile or vice versa.
What we want to do is we want to actually look at our data as quickly as possible for the campaigns that matter to us, understand our goals, and then start tweaking there. We're going to change our bid modifier and we're going to actually see, okay, I made a change, I'm going to be more aggressive on mobile because I'm, let's say, in the second position and I want to be number one to get more traffic. You're also going to want to monitor how much you are spending on mobile versus computers because let's say you're spending a majority of your budget on mobile, then maybe you're not getting as much desktop traffic as you could. So you want to lower that and then take this on a week to week basis. Analyze, download, make changes and then repeat. Continue to use that as your experiment.
When we talk about any of the experiments that we're doing . . .
Tristan: . . . you've suggested doing that download analysis about once a week?
Marc: I would say at least once every two weeks. It depends on how much traffic you're going to be getting. If you look at a week, and you see you're getting no mobile traffic and you want to be getting that but on desktop you tend to be in a really high average position, the odds are you're going to have to look at that more frequently to see how high you have to jack up your mobile bids and your modifier to get there. You also don't want to want to be spending too much, so once a week would be great. Once every two weeks is good if you're on top of it, and then ultimately this is something that never goes away. This is something we always want to be testing because our competitors are always making changes, doing things differently. Our click-through rates are going to vary. Our performance is going to vary, so it is something you want to stay on top of very much so. Great question.
In terms of the mistakes to avoid, from all our experiments we're going to go over certain mistakes. So for mobile and modifier or mobile modifier, let's think about the baby and the dishes, right? You're going to have to wash your baby, you're going to have to wash the dishes. Maybe you could wash your baby in the same sink that you wash the dishes, everything gets washed, everyone is happy, right? You could do that.
Potentially, it's a mistake, I would advise and I'm sure most of the parents, which I am not yet, on the call would probably say you know that's maybe not a good idea. You maybe do a better job washing the baby separately and washing the dishes because you want to get different things out of those.
I think the same thing for mobile and desktop. We don't want to blanket bid for mobile and desktop, have a 0% modifier, not be analyzing their performance, the average position for mobile and not making changes. So you really need to look at these stuff differently. Sure, you're still just bidding and you can't set a specific bid for mobile but you have to see directionally how things are working, what the performance is, and make sure that you're addressing each of those individually because again, it's important to. They work really differently, the search results page are different, and you're also tracking a lot of different things.
The second mistake to avoid is really being careful about mobile overspend. When you hear so much about how there's a lot of mobile traffic, that is true. If you all of a sudden decide, "We saw this webinar, we're going to decide to start testing mobile traffic. Let's optimize and put that into our desktop and tablet campaign." If you normally spend $100 a day on desktop and that's super profitable for you, and then you add mobile at a 0% modifier but you're also taking mobile, if on a daily basis, you started spending $50 on mobile and $50 on desktop so you're not spending $100 on only desktop, you might see your conversions go down. You might see that you're actually spending and inhibiting what you're doing on desktop and all of a sudden, you're not getting as many conversions as you used to or your ROI is really bad.
Or maybe you're in the top position for desktop and for mobile, you don't have to bid as much, and there are so much more mobile traffic that, again, you're spending a lot on mobile, you have to be really careful of that. We want to understand the conversion events, make sure that we're tracking it correctly and then when we do optimize mobile, we want to watch it because, again, the budget lives in a campaign that's being shared between devices, so if desktop performs really well for us, we don't want to rock the boat. We don't want the cows to light our barn on fire. We want to keep our barn intact and do some testing.
In terms of some clients of Elite SEM that do mobile really well, I want to shout out Lilly Pulitzer. So Lilly Pulitzer does mobile well on two fronts. In terms of cross device, you will see that they have a responsive web design. On desktop, tablet and mobile, the actual site experience will change but still provide you with the same usability across all different devices.
The difference with mobile is you can essentially browse the whole site with your thumb and things like searching, if you want to click and then search is available but for something like desktop, that's where it's more spread out, that's where they are expecting you to be clicking on drop downs and different navigations. For mobile, they really simplified it and made it a great mobile user experience.
Additionally, for search, Lilly is able to see what they need to be bidding on desktop to be number one and then also pulled back our bids on things like mobile where we're still looking to get the traffic but ultimately if we could bid less for it and save some money there and still be in the top position, we can do that. They've been really successful in doing that. Another thing you will notice is on our mobile ad, we have things that we're going to talk about extensions, click to call. For people that are coming on a mobile experience, let's bid accordingly, let's maintain the top position but not pay too much for it and let's give people the type of experience that we think they are looking for.
That's our mobile experiment, and that's our introduction to modifiers. When we talk about search, search is going to be the bread and butter to most of the companies on this webinar. Search is what's been around for a while. People are actively looking, raising their hands saying, "I want this product or service that you provide," so search is really easy. There's a lot to search, we have to match up what keywords with the right ad and the right landing pages, but ultimately, we're all doing search.
What I'm trying to get out of this portion for search is not how do we just do search, but how do we really exhaust search as much as possible. How do we get the most out of search? To do that in terms of understanding the landscape for exhausting our search, I want to introduce you guys to two concepts. The first one is extensions. We want to showcase the best looking ads and get the best ad performance for our users through the different types of extensions that are available with Google, and we're going to go through those, the most applicable versions.
Additionally, we want to understand all the bid modifiers out there because there's more than just device type modifiers that we could use to actually look at performance, bid more aggressively in areas where out conversion rates are better or making more sales, and save some of our budget, actually bid lower in areas that aren't working as well where we maybe shouldn't be spending as much.
When we talk about extensions, the first one I want you guys to think about is site links. Site links, I'm sure 90% of us on the call probably know what these are. Essentially, you do a search on Google in addition to just your normal text ad coming, you're actually able to have additional clickable links that take people deeper into your site to increase your conversion rates. If someone, let's say, is looking for Elite SEM, and we're not sure if they're interested in SEO or SEM or product feed management or conversion rate optimization, if those are our site links, they're able to actually on Google.com click deeper into our site to get deeper into the funnel, and hopefully, convert quicker, right? Not necessarily have to navigate our site.
We talked a lot about mobile earlier. When we talk about location and call extensions, these are super useful for businesses as well. If you have brick and mortar locations, especially when people are looking on their mobile devices, what you can do is these are actually all linked through Google My Business. If you look at your Google My Business page, have all your locations in there have the right address, have the right phone number. I know when I'm looking for restaurants or local stores, I'm on my Google maps doing searches, so you are able to opt these into not only the search results page but also map listings.
When it comes to mobile and good extensions for getting people to the right place, let's say you don't have a great user experience, what you are able to actually do is you can actually in the bottom left have what's called the click to call or call extension. If you don't have a great mobile user experience, you have a great sales team or customer service or they can actually call a local store, you can actually have them use what their phone was meant to do when they are searching on it and make a phone call. That's super intuitive, works great.
In terms of the other things for extensions for dressing up your ads, giving a good user experience, we have seller ratings and call out extensions. These are both pretty easy to implement. Seller ratings, you have a third party site, verify and get reviews for how people like using your site. These are really good for click-through rate, shows that you have a great quality product, people like using your site, like your brand.
Call out extensions allow you to A, take up more real estate on the right and also tout a couple of other business best practices that you have. If you have free shipping, you have 24/7 customer service, if you won an award, these are great examples and great use cases for actually putting more into your ad, having a better experience.
Then finally for extensions, we have app extensions, app installs. Let's say someone is looking for your brand and you actually want to drive mobile commerce or mobile users in general to your app experience. More and more of our brands are focusing on apps or including apps in their e-commerce experience. You can actually have people directly from Google.com in the search results page, download your app from the Google play store or the iTunes library, so that's really cool.
Or let's say someone already has you app and they are looking for a product that you sell within it, you can actually use what's called deep linking and on the example to the right, you'll actually see that let's say for instance, someone's looking for a product that you sell with your app, you can take them directly to that page, to that sale, to that experience within your app. I think you're going to see more and more of app installs and deep linking as more and more people get applications. In terms of dressing our ads up, providing good user experience, those are all the main extensions that we need to know about.
Now that we've covered extensions, let's get more into the testing and let's talk more about the modifiers. We've already talked about device type modifiers. This is where I have my desktop and tablet bid. I am actually going to bid up or down based on how I think I do on mobile. If I do better on mobile, maybe I bid more. If I'm not in a high position, maybe I bid more. If I'm spending too much money and I'm landing the barn on fire, maybe I'll bid less.
The other modifiers or options for modifiers we have are day of week location, time of day and household income. Let's think about this for a second. If we sell products and we're an e-commerce store, do we think that across all of the 50 states in the United States or between the U.S. and Canada, that we're just going to have the same conversion rate, the same lifetime value, the same ROI and average order value of every single person that buys something from our site, regardless of where they are from? Absolutely not. We tend to do the majority of our sales in certain states. During certain days of the week, our site will convert better or worse and the same thing with certain time of day.
Just like for mobile, when we talk about doing our first experiment, we want to be able to start identifying and analyzing the differences in traffic and performance and then tweak accordingly.
Then our final one at the end, Montgomery Burns, one of the modifiers available to us is you can actually bid higher or lower based on household income. If you have a really luxury product and you want to bid more for people that make over a certain amount of money for their household income, Google actually will allow you to do things like that as well. The way that works is it's based on zip code and the census, so it's not exact science. It's not tracking Montgomery Burns specifically, but his zip code within Springfield is where you could essentially bid more to show up in front of certain people.
When we talked about this first experiment that we wanted to do, we want to talk about similarly the mobile, let's analyze, let's take a look at location, day of week, devices and let's essentially see how they do and start to tweak our modifiers. It's the same kind of cadence that we're looking at with mobile. We also want to be experimenting with the types of extensions that are out there.
It's worth mentioning that nowadays, Google doesn't actually let you pick. If you have more than one extension live for your campaign, it doesn't actually let you pick which one serves. It will choose the extension it thinks is best for user experience and your extensions are a part of quality score. Quality score is a huge part of advertising with Google, paying less, having your ad show up in a higher position, getting more clicks and traffic. So you really want to maximize different extensions, test them, see which ones give you the best experience.
When we talk about the modifiers, we want to do a similar thing that we were doing with mobile. In terms of where you download these for time or geography, we are actually going to go into the dimension tab of our AdWords account. Instead of viewing by let's say conversions, we're going view by time. You can view by day of week, weekday, hour of day for time of day. You can also look at this by geography as well.
What we're going to do this, I download this sample report right here. You'll see that on Sundays and Mondays, we tend to convert a little bit worse than we do for Tuesdays, Wednesdays and Thursdays. Our cost per conversion is around $30. Our conversion rate is around 2%, where in the middle of the week, it's around 6%. If I could tell you, "Hey, let's take half of your budget for Sundays and Mondays, and then put that into Tuesday, Wednesday and Thursday when it is going to convert better for you, we can do that," wouldn't you want to?
Wouldn't you want to bid a little bit less in the time of day or the day of week or the geography that you don't do as well and put that money, bid more aggressively. Instead of being in the 2.4 position on Tuesdays, maybe try to get in the first position and use that budget there? I think so. Our test here, the experiment here is let's look at our data for time of day, day of week, geography, let's even test things like household income. Start to modify, make those changes. See how it affects your overall performance. If it's working well for you, continue to download, analyze, adjust and then report.
Tristan: When you're doing this type of experiment, how big in scale do you have to be? If you spend a couple of thousand dollars a month, is that enough to really be tweaking this kind of thing or do you need to be spending $50,000 a month?
Marc: Well, that's a great question. Ultimately, the more time we have, the more data we have, the more these trends are going to be very apparent in terms of Sundays and Mondays are just not our best days and Tuesdays, Wednesdays and Thursdays are.
Another thing that you want to be really careful - and we're going to talk about mistakes in a second - is if you are, let's say, looking at your whole account, you want to be careful that you're opting out, you're not looking at let's say a big sale that you had on Tuesday because that's going to skew all your data. You want to look at like campaigns so you're going to want to look at non-branded separately from branded. You're going to want to look at exact match separately from broad match and all those different things.
Then ultimately, I mean, you can start looking at these stuff as soon as you have data, the bigger the sample size, the better. Even if you're spending, let's say, $100 a day or a week, after a couple of months, you should have a sense of geography, right? All right, where are my conversions coming from? Maybe get a ton of traffic in California but you don't get that many orders, right? So maybe we're going to bid lower in California even though we're getting a lot of traffic to see how some other states might do.
Like what I was saying before, this is an ongoing thing. This isn't something that we do once and never look at it again but the more data the better. You could do this with minimal amounts of data. You just don't want to make too quick or rash decisions where you're cutting something off completely if you only have let's say less than 30 clicks.
Tristan: Got it. Cool. Thank you.
Marc: Yeah, no problem. In addition to only looking at like campaigns, in terms of mistakes to avoid, we talked about how you don't want to just bid the same for mobile and desktop, it's really the same for geographies, time of day and day of week. If we do better on Tuesdays and we don't do as well on Sundays, we shouldn't be having the same bid and potentially even the same budget for those types of campaigns or those different things that we can modify. Let's not be lazy, let's not wash our baby in the same sink that we wash our dishes. Let's actually look at that data and let's modify and bid for things based on what they should be bid based on how good the performance is or how bad. We're going to end up saving some money and being able to be more aggressive with those savings in times that work really well for us.
This next one is one of my favorites because we're continuing Simpson references, and this is one of my favorite Simpson character,s and because this is a really big mistake. On one hand, Marc is telling me, okay, multipliers are great, we need to figure out what we are doing on mobile and get into first position but we don't want to bid as much in California, and Sundays we don't do that well but on Tuesdays, we do really great. We have to be careful because bid modifiers are actually multiplicative. So let's use the following example.
Let's say we do really well in New York, so we're going to multiply our bid in New York by 50% and on mobile we're not in the top position so we're going to multiply that by 50%. On Tuesdays we do really well so on Tuesdays we already multiply our bid by 50% as well. If you have a $10 bid, you are getting 50% times 10 and then that times 50%, times 50%, so you can get really out of hand really quickly with the modifiers. Unfortunately, Google has not made this really easy to test multiple things at once so our recommendation is actually you should try and not test many modifiers at the same time.
Test the day of week. If day of week works really well for you and you feel like your test and your experiment was conclusive, it's working, potentially, you might build Tuesdays through Thursdays in its own campaign with a new bid and then you could modify for things like mobile or things like time of day within that or things like geography. Be very careful with the bid modifiers because they are multiplicative. You don't want to, again, have those super high bids on mobile if you're trying to be aggressive and take advantage of some of what we were talking about in terms of being as aggressive in areas where you convert the best.
Just a reminder, I see a lot of questions in chat. That's awesome. We want to continue to get a lot of questions, we're going to have a whole Q&A at the end. Don't forget to send all your questions to #PPCevolves, with an "s" at the end, because we're going to address all those questions at the end and hopefully, we'll have some great conversation about all these stuff.
Let's avoid overspending, we don't want to have to have a degree in mathematics to actually understand what we're bidding, so avoid the multiplicative bid modifiers. We're going to test one thing in a time and we're going to be careful and we're not going to do too many tests at once. Also, that's harder to really know what worked and what didn't.
A final example of modifiers of a company who does it best. One of Elite SEM's clients is a company called Magoosh. Magoosh is a leader and a renowned online test prep. You come to Magoosh, they help you get higher scores on the GMAT, the GRE. They actually have a score guarantee where you don't pay unless you get certain percentage of points higher. They're a really cool company.
As you can see with their ad, they have exhausted all of their site links. They've really tested this over months where these are the best links. These are the things that they really want to tout. They are really nice looking ads. They have worked on their ad copy, they also have a click to call extension.
Additionally, they use geography, time of day and day of week to actually bid more and be in higher positions in the areas and times where it works the best. They have had so many success with day of week targeting that they actually even run specific promos on certain days of the week now because of bid modifiers and all the success that they have been having.
Now that we've gotten to do the first two big experiments, we're going to have one more big experiment and then we're going to have three quick hitters in terms of new opportunities, more cutting edge stuff. We want to take a time, have everyone take a breath, talk about before we get into shopping and product feeds, how optimized is your product feed? This is for questions for people who, we are talking about shopping.
The answers are, "What is a product feed?" so product feed, I've never heard of that. "No", it's not that optimized, right, like I had one but I have no idea, probably isn't that great. "Yes, but it's not great," and then, "Yes, it's very well optimized." Do we not know what it is, do we have one but we have no idea if it's that optimized? Do we have one that no, it needs work or do we have one and we're the product feed masters? Let's take a minute.
Tristan: Looks like the results are rolling in. Do you want to take a minute to take one question from the audience while we're waiting for everyone to respond?
Marc: Sure. What have you got for me?
Tristan: We have a pretty straightforward one, I think. Somebody wants to know, regarding detailed site links, does quality score affect whether AdWords display the detailed versus the simple site links?
Marc: Yes. That's a great question. You'll tend to only see site links on the top left of search results. They tend to be ads that do have a really good quality score, they won't always serve, and the same holds true for detailed site links. My recommendation would be in any area that you're using site links or that you have the opportunity, so any campaigns for search, you want to have site links in there because your goal is to essentially get to that top left, have a good quality score, have good ads that match your keywords and have a good ad to keyword and landing page quality.
The goal would be to get detailed site links on every campaign, to have site links because if someone's looking for it, could we take them somewhat deeper in this page and then also to have details based on those. Like the audience member is guessing or asking, quality score is a factor so if you're not in the top left position, you might not always see them even if you have them, but you want to show for them if Google would allow you to.
The search result page is always changing, your position is always changing. In that chance that you actually do get into the top left position even if you're not there on a normal basis, potentially to have site links and then detailed site links with the good quality score is something that you'll be looking and aspiring to. Yes, it does factor and you won't always see them, but best practice would be to have those in there.
Tristan: Cool. We've got some data back. It looks like over half our audience either isn't aware of what a product feed is, and I count myself as one of those people before we started prepping for this webinar, or they know what it is and they don't have them. Almost 30% says, "Yeah, we do have a product feed but it could be better." Almost 15% says, "We have an amazing product feed."
Marc: Yeah, shout out to the 14% that have an amazing product feed. Good job guys. For the rest of us in the majority of the audience, so because we're not 100% sure what a product feed is or that's not something that we feel super confident about, I'll discuss a little bit about what it is and then we'll get into the next session. Essentially, a product feed is nowadays mostly for e-commerce clients and we're going to talk about extensions and product feeds outside of e-commerce, but what your product feed is is it's an XML or Excel document that actually shows all of the different products and inventory that you're selling on your site, the title of that, a description of it, what size it is, how much it costs, shipping, all the specific information, is it new, is it used, all of that information text of your products.
That's how Google actually, when we go to Google and we search for let's say "Nike sneakers" and when you'll see those product ads, so I'll go back a couple slides real quick, and I'll actually show an example of what this looks like or I'll fast forward to one of the companies that does a good job.
When you look at on the left-hand side, so someone's looking for a grill. The way the product feeds work are Google, we have a product feed, we put it in our merchant center which is something that essentially feeds our most up-to-date product feed to Google and then we don't pick keywords to decide what shows up on search. It actually will look at our feed, look at out titles and then decide, "All right, company X has this product, company Y has this product. We're going to show this specific products to people and then let them click on this picture ad and go directly to that product page."
That's essentially what a product feed is. It is a representation of all the different products that we have in our inventory and we're sharing that with Google or Amazon or eBay or any other shopping comparison engine company. Again, the biggest, most important thing to understand here is we don't choose the keywords that serve product ads, serve shopping ads, so it's very different from search.
Product feeds are so important that a year and a half ago, we at Elite SEM actually created a division to manage product feeds because it's such an important part of showing up in Google shopping. So really, we're going to talk about in this section how to maximize your feed, get the most out of shopping. For those of us that answered no or we think we could do better, we're going to touch upon that and hopefully, that's useful for the group.
In terms of understanding the landscape, I want us to think about Arnold and I want us to think about the Terminator and I want us to think about hopefully not the best series or movie in a series, the third movie, the rise of the feeds, how important feeds are playing a role in terms of shopping and what's going on in e-commerce. In the latest Elite SEM white paper that I finished about a week ago that's going to be on the site in the next 20 or 30 days, we're designing it right now, we saw that e-commerce clients spent 93% more on shopping campaigns in the first half of 2014 versus the first half of 2013. That means, if I was spending $100,000 on shopping campaigns in 2013, I spent almost $200,000 on shopping campaigns.
Now, why is this happening? Google is focusing the search results on shopping. Users love shopping. They are clicking on the ads. Google is supporting this, they are getting behind it. They are making the experience better on mobile across all devices. Google is really driving a lot of the additional search that's going to shopping, and users really love it. It's great customer experience, people are really liking it. Feeds are becoming a more and more important part and a bigger part of a digital marketing mix because a lot of non-branded traffic now, instead of going to search ads, is going to feeds.
The second part of understanding the landscape of shopping is just recently, so for those who don't know what a feed is, this is going to be news to them and for others, hopefully it's not as much news. Google Shopping used to be called product listing ads until about a month or two ago. Google changed product listing ads to now, Google Shopping and that's a name change, right? What else is actually different there?
It used to be that even though you could bid and target things based on your feed, so in this example, let's say we are bidding on men's shirts because we sell men's shirts and that's a target on our feed that we're going to bid for. You would see that you spent $100 on men's shirts and that you got 10 conversions and your ROI was positive. It wouldn't actually show you product data. You wouldn't know that you were actually selling mostly blue men's shirts or a specific type of men's shirt or even one specific style was driving all of your conversions but a very small amount of your cost. Maybe red men's shirts or red shirts were driving half of your cost and zero conversions. We weren't getting product data, which now we're getting with Google Shopping, so they have given us much better reporting.
We can actually optimize a lot better for specific products and segments, which again, the theme today is how do we make smarter decisions based on data and performance. That's a huge change from product listing ads to Google Shopping. Now we know the product data.
Additionally, Google rolled out two additional things. One is more custom labels. It used to be that in your product feed, you would get two tabs or two columns that you could essentially use to have custom values in there, whatever you wanted for bidding and budgeting purposes for targeting. Now, Google's increased that to five. What you can do with this is you can put in additional things that aren't in your product feed, like seasonality or best sellers or sale items or things that you maybe have a higher profitability on. Or let's say products that are under $50 or over $300 because potentially, you might want to bid less or not bid at all on products that are less than $50 because they are too competitive, your margins are too thin but you might want to budget more and bid higher on things that let's say were bestsellers, had higher profitability and sold for over $300.
In addition to your custom labels, Google now has what's called campaign priorities. When you talk about men's shirts, we had green shirts and blue shirts and tuxedo shirts and T-shirts, all these different types of men's shirts. Potentially, you get some intermingling on which product will actually be served based on the keyword. What you are able to do with campaign priority is put a low, medium or high priority on if the product's served.
Let's say you are targeting budgets or all products because it's a best practice to actually have one campaign that just targets everything in your feed with a really low bid. What you want to do in that case is put that campaign at a low priority. We really only want that to show up if nothing else is going to show and it was the last chance to actually serve a product. But if something else of a higher priority that has a higher average order value or is a better bestseller or has a higher profit margin, that's what we want to serve. That has a high priority.
In terms of the test here, really it all comes down to understanding your business, so speak with your merchandizing team. If you're an agency, speak with your client. Begin to understand what the custom labels are that you should be implementing and putting on your feed. Start to bid and budget and segment that stuff out, and ultimately, segmentation is really the key here. The more we segment out products that work versus ones that don't, budget separately, bid separately, we're just going to do better.
Really, your first experiment with this is having that set down. Take a week, take a couple of hours. If you could have additional fields on your feed to budget or bid for, what would those be? Let's add them. Then start to mess with things like campaign priority and segmenting for things that are more profitable for us or less profitable.
When it comes to mistakes . . . Oh, question?
Tristan: No, just because of time, I want to make sure that we can get through all these. Do you want to fast forward to and do quick hits on the last three experiments?
Marc: Yeah, let's do it. Really quickly, we want to not bid the same on all the products. We want to make sure our feed is reliable and has the right things in it. Formatting is a big thing. This book for "Cooking Your Dog", it's "Cooking FOR Your Dog", but small changes like this actually have a big impact on what shows up, so manage your feed correctly.
Then yeah, we can skip over, we can skip to the new breeds of opportunities to make sure that we cover the lost opportunities outside of shopping. When we talk about new opportunities, we're going to talk about ads formats that are on established platforms. Our users are there, we know we can get a lot out of these because the people are there, but there are some new ad types and because it's Halloween, we're going to do some spooky new breeds of opportunities, some hybrid, new stuff, cutting edge.
The first one we have is YouTube TrueView ads. Anyone that is familiar with YouTube knows that if it was considered a search engine, it would be considered the second largest search engine. There's ton of traffic on YouTube. Our users are there. What we are going to talk about is TrueView ads. Everyone knows what TrueView ads are. We just might not know what they are called.
TrueView ads are skippable pre-roll videos. Before you're actually going to be watching video content, you'll see a video, a pre-roll video for an advertiser with the ability to skip within five seconds. If you skip these, it actually doesn't cost any money. Within 30 seconds, you are able to have these 300 by 60 banner units. We see about 10% of people who actually watch the videos, click to the site. The goal with these is to pair these with re-marketing, pair these with companion banners, and actually try to get people to click. Clicks are very cheap, between 10 and 15 cents so definitely, something to pair with remarketing, something that we've seen a lot of great results from.
The next one is going to be Gmail-sponsored promotions. In terms of follow-ups post-webinar, we have an unbelievable article written by our own Shannon Greene. It really goes in depth, in terms of how you actually set up, manage and work and optimize your Gmail sponsored promotions. These show up in Gmail. They are these really rich media interstitials that the best part about this is really the targeting.
You could target mailing lists, you could actually target age, gender but you could also target domains. So if there is another business or another mailing list that you think is super complementary to the type of product that you are offering, you could target everyone who gets a domain or an email from let's say RJMetrics and say, "Hey, maybe you'd be interested in this product that I have," or any other competitor, things out there.
The next is going to be dynamic re-marketing audiences. We talked about dynamic or we didn't talk as much about. Dynamic ads are where you're actually, based on your product feed, see ads for people or for the types of products that you've been looking at previously on someone's site. If I've looked at three products and then I see ads for those, those are all connected to my product feed and they lead to better click-through rate and better conversion rates. The way that audiences work is they are generated by what's in our URL, so you can re-market to people based on what's in our URL.
Here you will see that even though I'm looking for a flight from Toronto to Athens, so I'm giving Jet Airways a lot of information about me, that I want to go from Toronto then I want to go to Greece. If I don't want the booking, there's nothing in their URL that they would know that I actually was looking from Toronto. But there's a way to tag custom parameters within your website and actually use this. It's in the help center. It's really easy, so you can create re-marketing list and re-marketing ads based on all the information that's actually on your pages even if it's not in destination URL.
We found this leads just to a bigger, better scalable list and it's something definitely worth testing whether it's for travel or for e-commerce. For instance, for Harley if you've a 2012 bike, that's a lot of important information that's maybe not in their URL that you could use for marketing purposes.
Great, so Tristan, we covered shopping a little bit to get through but we're ready for measuring and managing PPC performance.
Tristan: Sure, great. Sorry, I'm having some problems with my network connection. I'm trying to call in on the other line so at any point if I fade out, hopefully I can fill over the phone. I just wanted to take a couple of minutes to talk about how the PPC market has changed over the past five years. I have been involved in PPC since 2009 when I worked at Squarespace.
Back in 2009, Squarespace was really just printing money, doing super basic PPC ads. Really, we had two main campaigns. One was build a website and the other one was create a blog. Every single week, we just checked in with our agency and they told us what the performance was like and it was always amazing. We're like, "Okay, great, we're going to keep doing that." We grew, expanded to $50, even $100,000 a month. Things were going really well, but it doesn't really work like that anymore.
Today, if you see on the left, you've got a couple of cows there and it's easy for those cows to eat all the grass if they wanted. It's just of there and they can walk around and munch it, but on the right, that's a little bit more of what it looks like if you're in the PPC market today. You got to kind of be agile, and run in and grab the food amidst all these other cats what you can.
The reason that things have changed so much is that Google released a lot features and we talked about a lot of them today. But every time Google releases a feature, it is an opportunity for all of the different people that are competing in search to outdo each other. If you can take advantage of a new feature more quickly and more effectively than somebody else, you're going to beat them out and vice versa. If you're not making and taking advantage of those features, somebody is going to do that and your ROI is going to suffer.
So how do you do this? How do you keep ahead in this kind of world? Really, the secret to this is data. Speed equals agility equals ROI. The faster you move, the more things that you learn, the higher your ROI is going to be. This is a typical iterative process. You test, you measure and you optimize. One of the things that you notice from all of the experiments that we've talked about today is that Marc has been talking about the step where you first go into the AdWords interface and you download a spreadsheet.
Well, we really think that too many people are wasting too much time downloading spreadsheets and crunching your numbers. Really, you should and this stuff should happen automatically. There are opportunities today, and I would say really, look at in terms of your work for today, if you're downloading spreadsheets by hand and crunching those numbers once every two weeks, you should be looking to do something that's more real-time.
There are options today out there, including Google Docs has some cool options where you could grab stuff from APIs. There's an open source project where you can grab all of your data and stick it into a database. Of course, the easiest way, I think, is really to use a tool like RJMetrics. We connect directly to AdWords and we suck in all your data, and we let you look at graphs like this.
Really, you can see, we're using the throw spaghetti at a wall approach. We think that this is kind of the best approach to optimizing your SEM. We're trying out so many new things all the time. Our active campaign list has hundreds of campaigns and at any given time, we're seeing what we should be investing more in. By the way, this is our actual internal advertising that you're seeing, so what we should invest more in, what we should kill, and where we should tweak things and test new options.
All of that stuff happens automatically. It gets pulled in and we're able to, using our own tool, stay on top of all of these stuff that's happening. I read a blog post on how we did this, and this blog post is less about our tool and more about my time running PPC campaigns as a marketer and how different it is if you have all these data at your fingertips all the time, how you can diagnose problems when they come up and respond quickly. But I read a blog post on this. You can check it out on the RJMetrics blog and it has really step-by-step how to go through doing this process.