How to Choose a Bid Strategy

‍

How do you choose which bid strategy to use for your account? Which bid strategy is best and is there even such a thing as the "best" bid strategy? How do you increase your value as an agency, beyond managing campaigns? These questions answered, and more.

Full Q+A Transcript Below:

‍
Isaac Rudansky: All right. Welcome, everybody to another exciting session of Q&A Live with myself, Isaac Rudansky, and Patrick ‘The Great’ Gilbert and Danielle Immerman. We're here to talk about all things digital advertising, answer your questions, try to see if we could add some value to your lives, to your campaigns, maybe to your relationships. But who knows? Who knows where the road might take us? 

I hope everyone's enjoying, I guess, the tail end of winter. Danielle, you were in Florida. How was that?‍

Danielle Immerman: It was amazing. So happy to be back in the cold, frigid New York weather.‍

Isaac Rudansky: The background might be deceiving, but Danielle is actually back here in New York. And if you have any questions, guys, if you're listening to this, and you have any questions about your campaigns, post them to the thread beneath this video, and we'll see them and we'll try to get to your questions. 

We have a bunch of questions from the prior weeks that we've saved, that we didn't get a chance to get to. So Danielle, do you want to start us off with a question that we left off with, from one of the members of the group?‍

Danielle Immerman: Sure. Why don't you give me a minute to find those and start discussing them? ‍

Isaac Rudansky: Okay. In the meantime, Patrick, why don't you tell me the target impression share? ‍

Patrick Gilbert: Sure. So-‍

Isaac Rudansky: One second. Wait. Not yet. ‍

Patrick Gilbert: Okay.‍

Isaac Rudansky: I forgot. Many of you have not yet purchased a copy of Patrick's book. It's called Join or Die: Digital Advertising in the Age of Automation. Patrick, is that the softcover?‍

Patrick Gilbert: No, it’s the hardcover. I have a soft cover over here. ‍

Isaac Rudansky: This is the hardcover version. It also is in paperback, and Kindle. It's available on Amazon. It's gotten unbelievable reviews. People have written to us, to myself and to Patrick, saying that this book has literally opened their eyes to techniques and strategies that they did not even know existed. Everything here is explained clearly, simply, easy to follow, but these concepts are actually pretty advanced. So not only will you understand more than you did going in, you're going to look like a hero at your company. 

The truth is in the pudding in the performance that you're going to get from your Google campaigns, your Facebook campaigns. If you want to understand the intricacies of Facebook and Google's AI, and how machine learning works, and how you can put it to use in your own campaigns and save yourself a lot of money, this is the book to read. It's very practical, straightforward. And the cost of the book will pale. It'll be a fraction, maybe a one in a 10,000th of the amount of money that you're going to save by not wasting it on ineffective strategies. So head over to Amazon, get yourself a copy of the book. And it makes a great gift as well. 

So Patrick, kick us off with target impression share.‍

Patrick Gilbert: So, target impression share as a concept is a type of bid strategy. One question we get often is about how to choose different bid strategies. 

We see a lot of activity in the Facebook group about somebody making a post saying, “I'm starting out. It's a brand new account. I have no historic data. What should be my … What bid strategy should I choose? Or which bid strategy is best?” There's a lot of these broad questions. Obviously, the unfortunate answer is that it constantly depends on a million different factors. The book, there's a full chapter outline to all the conversion-based bid strategies. So maximize conversions, maximize conversion value, target return on ad spend, target CPA. 

Once you understand how any of these work individually, you can replicate that out in terms of how the algorithm is going to make decisions for whether it's on Google Facebook anywhere else, whatever bid strategy it is. When we talk about maximize clicks, it's a bid strategy type that does exactly what maximize conversions does, just for the click metric. So instead of optimizing for conversion rate, it's really optimizing for click-through rate. 

Target impression share is unique because it's not using either of those variables. It's optimizing for an impression share. This is tricky because impression share is a relative metric. It's based on auctions that you enter and is not indicative of the market at large. 

I don't really address this in the book because we really don't use it very often, but the new motto of AdVenture Media in terms of how we go about campaign management is never always and never never. Meaning while we rarely use target impression share or maximize clicks or one of these more obscure bid strategies, it's not as if we never use them. So there's always a time and a place. 

One example that has come up recently, is using target impression share for broader keywords, more top-funnel keywords that have a much larger audience set. In doing so, you can essentially attempt to own a space. A few weeks ago, we did a podcast with Kirk Williams of ZATO, and Mike Ryan of Smec talking about smart shopping. The example that kept coming up was what if I want to own the coffee grinder space? We were talking about optimizing for search impression share. If you haven't had a chance, please check out that podcast. It's on the How to Hide a Dead Body podcast feed, but that's a whole separate … The smart shopping conversation is totally separate here. But we were talking about how to own a space, and target search impression share is one of the ways to go about it. 

What you can do with target search impression share is essentially lower your actual CPC. Because what you're doing there is you're bidding high on auctions where there might not be a high predicted conversion rate, which kind of gives you a competitive advantage. So generally, if you're using a conversion-based bidding strategy, you're increasing your bid when that auction is likely to result in a conversion. However, if you're looking at a given keyword as a top-of-funnel-type keyword, you're not expecting that person to convert in that setting. So you're almost okay with the fact that you're expecting a lower conversion rate. 

There's plenty of auctions that will take place that you can gain arbitrage, as we call it all the time in this Q&A session. But you can actually gain a competitive advantage by bidding in those auctions, bidding based on search impression share for more top-funnel keywords. So that's a great example of when you use that bid strategy. Isaac, do you have any thoughts or clarification? Because I feel like I just talked in a circle.‍

Isaac Rudansky: I just think it's interesting that we're even at this point, talking about target impression share. Because if you go back four years, we'd never have used it in any instance. 

It does speak to your motto. And I think it speaks to a similar, or maybe a sister concept that the more experience you have, and Patrick, you and I were speaking about this the other week. The more experience you have in any given domain, the more confident you are in that domain, to reevaluate your positions. This seems like a dichotomy, but it's not. Because when you're less knowledgeable or less secure about your approach, you become more ideologically oriented in that approach. 

Once you're ideologically oriented in an approach, because it maintains your own sense of identity or security, it becomes almost impossible to reevaluate or to say, “Okay, well, maybe we should look at this again, or look at this from a different angle. Or maybe there are instances where this is right, instances where it's wrong.” That's the opposite of being ideologically possessed. 

So as we've experimented, and we've become more knowledgeable as an agency, and we've maintained a commitment to our education, we've actually been much more willing and open to testing new ideas to the benefit of our clients. Which is a great attitude to have, because I don't think we would have ever thought about or really talked about using a target impression share bid strategy in the past. But it does make sense that there are instances where it can be valuable and certainly worth testing, especially if you don't have enough conversion data in an account. 

I remember recently, we all spoke about … So let's say you look at maximize conversions for a moment, right? So, maximize conversions is telling Google to ignore an imposed target CPA, because Google, it's not to ignore. It's mathematically inaccurate to say that it's ignoring CPA. Because if they want to maximize a volume of conversions within your budget, it has to also take into account the most amount of conversions I can get for the lowest cost per conversion. Maximize conversion value, it might be more accurate to say that Google's ignoring-‍

Patrick Gilbert: CPA.‍

Isaac Rudansky: The cost for the … Right. CPA. The cost for that acquisition, but maximize conversions is often misrepresented. It often gets a bad rap, because people think that it's like, “Oh, I'm just handing over … I'm just handing Google over the keys and they're going to spend as much as they want,” while it's like, no, that's actually not the case. It's mathematically inaccurate. It cannot be. CPA has to be a part of that equation.

However, okay, so now you're looking at an account that has no data, to go back to our practical example. Brand new account, no data. I've seen accounts get off to a good start using maximized conversions. I've seen accounts get off to a bad start using maximized conversions. How would you approach that question? 

There's a limited budget. Let's say, for example, it's an ecommerce store and the client is spending $50 a day $75 a day as a test. And there's really only enough money to test the first two weeks, three weeks with one of the two bidding strategies. How would you evaluate? Is it just 50/50? Are there indicators that you could look at to determine whether you should use maximized conversions versus a targeted search impression share strategy?‍

Patrick Gilbert: Yes, and I think this actually ties back into the target search impression share consideration. It comes down to what you're targeting, and how much volume. Take a step back. It comes down to your targeting. It comes down to, one, how much volume exists with the keywords that you're targeting. And two, how much confidence you have, that these keywords are your target audience. That's where we get to the whole funnel concept. 

So, the bottom funnel is an expression that we use to describe a keyword audience of people that when they search for that keyword, we know with 100% confidence that they are in the market, or at least searching for the products and services that we sell. As it gets more broad, it becomes middle funnel, top of the funnel. 

We’re a digital advertising agency, right, so we do a lot of PPC, so on and so forth. If someone was searching for a marketing agency, that's something that could mean what we offer. But we don't do SEO. We don't do web development. There's a lot of things that people think about when they refer to as a marketing agency like us. So technically, that's not a bottom-funnel keyword if we were running ad campaigns on our behalf. Okay, we've explained that a million times I won't go too much deeper into that. 

But if you have bottom-funnel keywords, if it's a PPC agency, which I think is what we could be described as, and you know that there's a lot of volume for it, and you know that based on the average cost per click, and with your given budget, you can afford several hundreds of clicks per week, then I think you should be confident in maximized conversions. That's where I would start. However, if either you couldn't afford that many clicks, or if it was a little bit more ambiguous, then you might want to consider a different bid strategy.‍

Isaac Rudansky: Okay, that makes sense. Danielle, did you pull up any questions?‍

Patrick Gilbert: She's not going to read it to us?‍

Danielle Immerman: What do you want? Story time? ‍

Isaac Rudansky: It's International Women's Day, apparently. So we're going to read the question ourselves. ‍

Danielle Immerman: That’s right.‍

Isaac Rudansky: How do you increase your value as an agency, beyond managing campaigns? Actually, this is a topic right up Patrick's alley.‍

Patrick Gilbert: Interesting. ‍

Isaac Rudansky: We could both talk about this. Maybe, Patrick, you can pull up your slide with Ronnie wiping out. Because this is your slide, right? When we have time to do other things. ‍

Patrick Gilbert: We play ping pong. We were on a ski trip. Ronnie fell down a mountain. We stopped going on ski trips.‍

Isaac Rudansky: Smart bidding is working. You're hitting client goals. What's next? I'll kick it off, and then I'll let Patrick give his take. 

So this goes back to how you identify as an agency first. I've been saying this for a long time, internally and externally. If you identify yourself as an AdWords management company, you're finished. It depends on the time horizon, I don't know. But at some point, you're going to be finished. Because Google is working on these things that are working, that obviates the need for agency management or as much human management. Especially if you define human management in the way that it used to be defined, where you're building keyword lists, and you're managing match types and you're calculating max CPC bids and you're adding negative keywords, all of those things are going to go away and they are already going away. 

If you define yourself as a freelancer or an agency that is trying to find creative digital channels to drive profit for your client or for your business, then your job is only going to become more important. 

So for us, I'll tell you what this means for us. We develop what I call consultative relationships with our clients, what does that really mean? It's not just a buzzword. We leverage the skills and experience of our team. There are people sitting out there that know data and Excel and regression analysis. I was going to say better than I do, but they actually just know it. I don't know it. It's not about better or worse, it's infinitely better because it's the difference between zero and one. There are people who know creatives, creative strategy better than I do. And we find ways to get involved with our clients and helping them with interesting things outside of smart bidding. We'll give a couple of examples. 

We were recently hired by one of the UK’s largest sports books to determine if it was worth buying one of the US’s largest sports media outlets. It's a complicated question. They want to expand their sports book into the United States. And the question was, “Should we go into the US market, branded as ourselves with very little recognition, with very little brand recognition? Or should we spend a lot of money buying a well-known brand, and getting access to all that existing traffic?” 

So, that's not anything to do with smart bidding. It has nothing to do with Google ads, but we were hired to figure that out. It was a really fun, exciting project. We were hired because the people we've interacted with over the years, knew enough about us, about the way we think, about the way we manage projects, that they were clear and they understood that that's something that we could deliver on. That was a really fun, exciting project. We had another exciting project.‍

Patrick Gilbert: Wait, can we stop there? I want to ask. How would you advise somebody to bridge the gap between being an AdWords management company, and ultimately being able to do that high level of consulting? What are the middle steps there to be able to get from point A to point B?‍

Isaac Rudansky: That's a great question, actually. I think there are a few things, and there are probably multiple ways to go about it. But one is you have to do the work. You have to take on some of these projects, and you cannot wait for those projects to come to you. 

I'm comfortable saying this, because we've always been this way, to a large extent, where a client asks us, or the client even indicates that they're experiencing this or that problem. And we'll say like, “That's fine. We'll get involved. Let us know. We'll help out.” That's what got us into understanding wireframes and web design and consumer psychology. It's related to what we do because the better your website performs, the better campaigns it will perform. It's not exactly what we do. But brands will always have a need to communicate their message clearly to a consumer, and we try to look for those opportunities. We try to insert ourselves outside of the border of what we were comfortable with. And everyone could do that, provided you have some work, right? 

There are two different questions. How do you get your first clients? Okay, we've had that question. We could have that discussion. Now you have a client. You're running their ads campaigns. That's a whole business. It's a complicated machine. You have access to cash flow. You see clients. You see email addresses. Maybe there's some data partner that you could figure out a way to extract really cool information out of a customer list, and that could be a cool service. 

We saw another example. I'll give you a perfect example from last month. We had a client, a law firm. She just randomly brings up on the phone call that they're spending all this money on TV advertising, and she doesn't know if it's working. I said, “Okay. Well, why don't we charge you, and we'll answer that question for you? We'll tell you if your TV advertising is working.” We have people who could figure that out. But I really feel that even if we didn't, I probably would have still made that offer to the client. I would have said, “Well, we're going to figure it out. We'll have to find the people or find the partners or find the vendors who could run these types of regression analysis by looking at the data stamped or the timestamped data from the TV agency and matching it up with Google Analytics, and finding a statistically valid portion of traffic.” 

She didn't ask us for it. She was just bemoaning the stress in her life, about her business. And we jumped on the opportunity to say, “Let us help you solve that,” because those are the types of projects that are going to help identify us. Now we've done that project, we could have a case study on our website. So, that's the second thing I was going to say.

The middle steps are not getting the work. Taking the work. Do it. Do it for free. If you have a client that you're running campaigns for, there's no excuse because every business is complicated. Every business has hundreds of different angles that you could approach it from. You have enough skills to lend your insight to one of those areas. Doesn't have to be about data, or TV lift analysis, or creative or wireframing. It could be a lot of different things. Just find what those things are, and offer them work. If you could charge your client for it, awesome. But if not, do it anyway, because now you have a case study. 

Then you have to promote that work. You put that case study on your website along with the other case studies. And you say, “Look, we did this interesting project.” Maybe it was about data. Maybe it was about customer persona development. Maybe it was about branding, or who knows what. And now people could find it. People could sort of seeing that you've done these interesting outside things that are not constrained inside the typical AdWords management, smart bidding, whatever. 

But it's a really important question because you have to start doing those things. Otherwise, you'll be left behind because Google is just going to release an update. And they're going to be like, “Okay, no more keywords left,” or, “No more negative keywords and no more search terms, and no more manual bidding.” It's like, well, what are you going to do? Then eventually, Google might be like, “Look, give us your credit card, tell us your goals, and we'll send you a monthly report.” What are you going to do then?

But now's the time to start thinking about finding more. That's what I'm trying to define in a roundabout way, what I call consultative relationships, where you're trying to solve and you're willing to solve problems, and you go out of your comfort zone, and you put yourself in very awkward, uncomfortable situations where you have to now deliver, and you're not really sure how you're going to do it. You feel like there's a lot riding on the line, but this is sort of a separate conversation. If you fail, there's really not that much riding on the line. But if you succeed, you've done yourself a lot of good. 

So that's my take on that idea. Patrick, what do you think? And then we could leave off with this good topic.‍

Patrick Gilbert: I really like what you just laid out there. I was writing a few things down because we've been struggling with this question for years. You were the one that, I don't know, six years ago, had said, “Hey, we need to assume that Google is going to actively get rid of agencies, or have a need for us to not exist anymore,” not as if that's what they want, but it's going to go in that direction. 

You had said, “We need to envision a world where reporting on our work inside AdWords campaigns is like a very small piece of a larger client meeting.” We've spent a lot of time trying to figure out okay, well now like, what is the agency look like when we become … When that sort of thing happens?” 

We have been lucky enough to develop into this area. We have these new consultative relationships, and they really run a large … They're pretty diverse. It's not as if we do one thing specifically. It's not as if we went from, “Okay, we figured out smart bidding. And now we do this specific thing.” It's more of we start asking questions with the client to figure out what else is happening in their business, aside from traffic acquisition? What other problems exist? As you noted before, every single business is complicated. It has a lot of problems. And if you just sit back and listen, you'll be able to figure out what those problems are. 

Then there are two ways you can go about it. One of which is you can hire people that can help you solve those problems for clients. That's what we've done to an extent where we've brought on account managers here that weren't … They didn't necessarily want to be a career account manager. But this was they came on, they figured this out. And now they get to do certain projects for clients that, as you described, you and I couldn't do. We couldn't do this sort of data analysis for clients or lifetime analysis, or TV lift, or any of these things. But we have the people on staff that have those skills. So you can hire for those things, and then present that to clients. Or figure out what you're good at, or what you are interested in learning, and then figure out a way to do that for clients. I'll give you a few practical examples. 

Years ago, one of the first consultative relationships I remember was with a client that had a call center. You have the experience, Isaac. You have experience in sales, obviously, and in managing a team of people and incentivizing salespeople. So, the owner of that business was having a problem with he couldn't figure out how to manage his sales team in a way that promoted or motivated them on a consistent basis. You sat down with him and said, “Look, if this were my team, I would do it this way.” You presented a few different models to him. That was completely outside of the scope of anything we had worked on before. But you had experience with this, and it was valuable to that client, and that was something that was great. 

But separate from that, think of all the clients that need help with their CRM integration or conversion tracking, or they can't figure out how to get good graphic design for their display ads. They have no idea how good storyboarding for video production shoots will work. They don't understand usability on a landing page. They are trying to bring a mobile app to market, and they have no idea where to even start. Their copy on their landing page stinks. 

There are a million different things that every single business is dealing with that fall under the marketing realm that anyone watching this right now has some sort of skill set in one of these different unique areas. If you just really explore that and start talking to your clients to figure out, “Hey, look, maybe I can get in the door to help them.” At least be a fly on the wall while they're developing their mobile app. Now, all of a sudden, a year later, you have a lot more experience. 

I think we lost Isaac, so that's good. ‍

Patrick Gilbert: It's kind of funny. Alright, so I think that's a really good question, though. Going back to it, the question was, things are working well. We figured out smart bidding. What do you do next, to continue to provide value to your clients? 

It really comes down to figure out what problems exist in that business, that you have the capacity to solve and play the long game. That's still something we preach all the time. This is not something you're going to turn around and start doing next week, next month, or even six months from now. Do the small things now gain the experience and a year from now 18 months from now, you'll be able to develop into that sort of freelancer agency, whatever. ‍

Isaac Rudansky: The long game is such a big idea. That first project that I was referencing with the sports book, so that's something that came through a relationship that was over two years. ‍

Patrick Gilbert: Yeah.‍

Isaac Rudansky: Over two years in the making. I have always struggled with the long game. I'm drawn to digital advertising because of its immediacy. It's like the anti-long game thing. But it's such a big idea. You wake up a year from now and you're like, “I feel the same way I did a year ago. I really wish I started losing weight a year ago. I really wish I started doing writing a year ago or doing anything a year ago,” because look, I'm not that old right now. It's just I feel the same as I did a year ago. It's like, “Why don't I start?” But it's true over the course of four years, or five years or six years.

I heard a quote recently where it was people tend to overestimate what they can accomplish in one year, and drastically underestimate what they could accomplish in 10 years. So the long game, it's also things get compounded. It's not linear. You bring on some clients. You can start doing some work. It takes a long time. You develop some relationships. You're posting your case studies. You're posting to LinkedIn. You're doing your thing. Then you have one opportunity that leads to the next, but it compounds because those opportunities get progressively larger, and they interact with each other, and they expand. 

So I think that's a really good point. Playing the long game. It's hard. We should really talk about that more in another session. But for now- ‍

Patrick Gilbert: Wait, we have one more question. Someone commented and it's a quick one. It's about in order to get conversions using display, do you use smart display campaigns and pay for conversions, which is an optional description? Or do you use the standard display with a smart bidding strategy targeting a more specific audience and pay for conversions? Or I think you mean here, pay for clicks or impressions. Thanks, and a big hug. 

Okay. Both is the answer. So the one thing I'll say, so Google has this pay for conversions option, which sounds great on paper until you waste a lot of your client’s money. And you sully some relationships, which is something that we absolutely did when this feature first came to be in 2018. Be very conscious of what's happening here. So with the pay for conversions option, what you do is you tell Google, “I'm willing to pay $100 per CPA for conversion.” Google will go out and they will spend their own money advertising. Serving your ads on the Google Display Network. When they get a conversion, you will just get charged $100. So if you get 10 conversions, you'll get charged $1,000. And that's just how it works. 

This works great in theory, but it's very risky. If you don't have proper conversion tracking in place, you're going to immediately spend a ton of money on bad traffic. If you have a lead generation campaign in place that is not validating the quality of the leads in the back end, you're going to waste a lot of money. So, that's the risks involved with that. 

The other side is if you're not getting conversions, this is really Google betting on themselves saying, “We'll go out, spend our money, and we're just going to charge you when we do well.” It's a performance-based bidding model. It's pretty cool. Actually, it's really cool that Google offers this. If they don't start to get confidence that they're going to be able to drive conversions within a certain amount of time, the campaign is going to slowly shut off. 

So, try it out. I think if you're an e-commerce store, definitely give this a shot. Do the paper conversion model. If you have conversion tracking set up in the right way where it only goes through when a conversion or a purchase takes place, I think that's probably the least risky model. So that's what I would just recommend. 

As far as the other standard option of standard display campaign where you're using smart bidding, that's probably what we use more often. And then depending on what the bid strategy is, maximize conversions for a remarketing campaign or target CPM or something like that. Depending on what your goal is, you can figure it out that way. But I would say definitely play around with those different bidding options. Figure out what works best depending on what your goals are. All right. Anything else, or was that it? ‍

Danielle Immerman: Yeah, we have one last question. ‍

Patrick Gilbert: How can I get one of those-‍

Danielle Immerman: Where did you get your T-shirt?‍

Patrick Gilbert: Me?‍

Danielle Immerman: Yeah. ‍

Patrick Gilbert: Wait, what? Okay, how do you get an AdVenture Media T-shirt? I don't know. Isaac, we should open a shop. ‍

Isaac Rudansky: We should do an online store. ‍

Patrick Gilbert: This is actually from a former AdVenture Media client that produced these for us. So I don't even know. ‍

Danielle Immerman: It’s so nice. ‍

Isaac Rudansky: Yeah, well, he charged-‍

Patrick Gilbert: It’s really comfortable. It's one of my favorite shirts. I probably wear it too often. I hope people think I have multiple. I only have one. ‍

Isaac Rudansky: Guys, thanks for joining us today for your great questions. Get yourself a copy of Join or Die: Digital Advertising in the Age of Automation. If not for the content of the book, at least for my foreword. Where is it here? It's on page … Is the foreword after the introduction, or before the intro?‍

Patrick Gilbert: Before the introduction. It's foreword. It's before the words.‍

Isaac Rudansky: Right. Not before the table of contents? ‍

Patrick Gilbert: Yeah, those don't count as words. Isaac Rudansky: So, it was forwarded by myself and David Sable. It's not only a book about the technical aspects of advertising. It's a book about agency growth. Patrick talks about these very questions. How do you not be just an AdWords management company? How do you get beyond smart bidding? And how do you ultimately provide value to your clients so you can retain them and grow them and grow the business and grow your agency? Join or Die, available on Amazon. We'll see you guys next week, and looking forward to it. Thanks for joining us today.