We're back. So by now you guys have all done the certifications. So congratulations on your nifty suitable for printing Google ads certification. So you know a little bit about the theory behind building your campaign. So we thought it would be good to break up the the theory and vocab stuff with some actual information about how we have paid search campaigns in the past, in real life. I think there's such a density of material and getting through that class but until you put it to work and see it on screen with some actual campaigns, it's going to be really hard to retain. Yeah. So I think step one is before you even go in and create your first campaign, which we'll do in a second, it is really critical to get prepared. A lot of people overlook this. They're nervous, anxious, whatever, their boss is saying, ''We just got to get an AdWords campaign going.'' Taking these steps that we're going to cover in these next slides is actually a pretty big part of overall success with these things. That's right, and because you're going to start spending a lot of money, you want to make sure that you're actually getting something from that. It's pretty easy to waste a lot of money. Absolutely. So the first thing to get prepared is a little bit obvious, and hopefully everyone already is at the point where they know their business but I think it's worth being explicit about it to say, who are your customers and who are the people that you're trying to reach with this search campaign? Yeah, that's everything from where do they live to what types of language would they use to find your product or service. So it doesn't have to be a psychological deep dive, but spend a little time thinking about who they are, and again that idea of concentric circles. You want to start with the tightest target possible, spend the least, learn from that, and then grow out. So what is the need they have at the time when they come to you and are ready to become a customer? Absolutely. So once you've done that and then thinking about okay, now what are we trying to accomplish with this actual marketing campaign? Yep. I think for most people it's going to be, well, I'd like to drive more sales and get more revenue, but not always. Yeah. I mean, you're a local business. Do you want to make the phone ring? Do you want to get people in the store? Is website traffic valuable to you? Just you're a new business and you want to raise some awareness. So how you would operate those campaigns should be tailored to those goals. Absolutely. Because your experience, your face to the customer through this search campaign is all going to be online you got to have your website in order, prior to spending a dime on search. It's pay now or pay later. Sometimes that means you're going to have to invest a little bit to get that website where you want it, but failure to do that means Google's going to actually increase your costs and you're going to pay more click by click, so how can you really get set up for success before you start? This is now something you prepare for but also ongoing, you're going to need to continue to optimize your website, which you're getting some feedback. So some of the tools that folks can use to do these. I think Google Analytics is pretty ubiquitous and essential now. Yep. Easy to deploy, free, really robust. I mean there's other stuff out there for analyzing the behavior of your web traffic but they're really the predominant. Then AB testing the actual landing pages, tools like Optimizely can all help with that. Great. So now you're ready to actually get into search, knowing what's coming down the pike around what keywords you're going to need to enter and writing ad copy, knowing that that's all going to happen and you're going to need some work on structure. It's good to sit down first and prepare some of how you're going to approach those things. Yep. So again geography is a big deal, and maybe you are a business that is predominantly web-based and you would take customers from anywhere. But if you know just from being in this business a couple of years, the majority of your customers come from Seattle, Sacramento, and Chicago, maybe you want to start with some specific target. Yeah I don't know who those customers are but okay maybe. Yeah. Then you're going to have to deal with this concept of ad group structure and campaign structure as well, knowing, and we'll walk through some of your own experience on how you've dealt with that in the past. Yeah. There's not an absolute right answer, but hopefully, we can convey a few logical paths that you could apply and you could try structure a or structure b. So my favorite part of the math part is also critical. So knowing that you're going to spend a lot of money on this search campaign to drive customers to your site, you need to be really well versed and have some detailed understanding of how much you can spend and how do you make money because you're not doing this for charity for Google. So two of the metrics that you can optimize for in your search campaign are return on ad spend, and I'll take in a little bit about cost per acquisition as well. Return on ad spend is probably the gold standard for an e-commerce. So how much revenue do I generate across all the different shopping carts, so people checking out versus how much I spent on search? Yep. This is important that it's easy to gloss over, but this is what's going to impress your boss or your client if you're an agency or whatever, it's where the money they gave you to go run these ads in whatever capacity, you are doing that. What did you make? If that's profitable, then you can always build on that, and like okay if we spent $1,000 and made $4,000, what happens when we spend $2,000? So this is a metric, you want where higher is better. How much money do I make for every dollar I spend on advertising? In general, calculating your ad spend is easy. You can just look at your credit card balance and see how much Google charged you getting the revenue. If you've got your websites set up well to directly measure from your shopping cart where that value is, that's also pretty easy to directly measure. Otherwise, you're going to need to know your average check out value to be able to calculate that. Even so not without issue, attribution is always one that's a little bit hairy. I personally feel like the Google's in a position given that bottom of the funnel place that it lives, where it can often get too much credit because someone heard about you through word of mouth or some event that you did, went back home, searched for your brand name and clicked on a Google ad. You may be giving too much credit to that Google ad and not enough to the event. Google undoubtedly had an important role in that. That's what a lot of people when they talk about attribution, that means you've got multiple ways you're interacting with your customers, which one do you attribute the final action to. So you see that there's now multitouch attribution models and time came where they try to make sure they understand what are all the ways we interact with that customer and spread that value across all of them. That gets pretty sophisticated. In Google Analytics, you're usually going to see the last click attribution. Within online, there's good models to attribute but even if you're doing any offline advertising like events, then that gets even more complicated. So I would say generally build in a little bit of buffer there to account for some of the error in attribution. Then also make sure you're factoring in your profitability. So return on ad spend is about revenue. You need to factor in also your costs of goods sold. Yeah. If you're $100 product really only nets you 20 bucks, don't confuse that when you're coming up with your return on ad spend. Right. It's seems obvious but some people skip that step. Then you get into, speaking of hard to measure, there's a bunch of other factors that CMO's also love to talk about, "Hey, let's spend, let's lose a little bit of money up front because we know users are going to keep coming back and so we'll get repeat views. They'll tell their friends about how great our product." I think all those things are true but you need to be careful. Yeah. I mean there was a big case study of how Staples, the office supply company, they were paying two dollars per click to drive customers to the site for paperclips or something that only cost a dollar. But they had done the computation that even if they'd just buy that one thing, if we can give them a good experience and they shop for their office supplies over the next 12 months, the lifetime value makes that cost to acquire still worth it and you've got a long-term ROAS. For non-shopping cart related campaigns or businesses, a cost per action is generally the metric that we optimize for. So this is the total amount that you spent on your marketing with Google, divided by the number of leads or whatever it is the action typically it's a call or submission of a lead form that is what you're driving. That could be a couple of things to write, insurance has always used lead forms. You're not going to complete your policy online necessarily but maybe I want to talk to an agent or whatever, but this can also be a measurement of success in e-commerce. I've worked with some companies that have really high-value things: furniture, automobiles, satellite phones, these different verticals where you're going to spend a couple thousand dollars on the product. Maybe you want to get them higher in the funnel and so the lead is give us your email, we will send you our brochure on the top 10 sofas that we've produced over the last 10 years, you can learn about all the features or whatever, share it with your spouse and make your purchase decision. Not everybody is ready to purchase especially if it's a higher dollar item. So you can use cost per action in a few different ways. Pretty straightforward to measure if you're setup well. So again there's a few things you need to be aware of even if you are optimizing for CPA, that is the action of a leads submission. You don't make any money off the leads, you make money off the sales. So make sure you're also factoring in that conversion. That can be hard. You're tracking close rate. You're saying for every 100 leads, do we turn five of these into customers, sometimes that happens really quickly in some businesses. Those people don't become customers for six months and it's a real tracking headache. Yep. Very often you're moving between systems and so you break that connection, you don't know which keywords ended up bringing in the high value customers versus or the high value leads versus rights didn't close, so that can be challenging.
