On this week’s episode of Polishing Your Digital Marketing, we sit down with Shawn Walker, VP of Strategy and Product Development at Symphonic Digital. Shawn is a digital marketing expert who works to develop strategies for digital media investment to maximize ROI.
When asked about the importance of Google ad bid modifiers, Shawn says, “I think the number one thing is people don’t know they’re there. Number two, if they know they’re there, they focus on other things that maybe they think have a greater opportunity to be more efficient, like working on ad copy, or adding keywords. I would argue this is just as important. If you’re using the correct keywords but you’re not targeting the right people, that could be just as bad as having irrelevant keywords.”
We chat further with Shawn about…
- What bid modifiers are
- What levels they can be applied to
- How they can help to optimize a digital marketing campaign
- Why human supervision is still necessary when automating your ad campaigns
- And more
Mentioned in this episode:
Steffen Horst: Welcome to the Performance Delivered Insider Secrets for Digital Marketing Success Podcast. Today, we’re bringing you a new episode from our series, Polishing Your Digital Marketing. Here with me is Shawn Walker, who is the VP of strategy at Symphonic Digital. He is the mind behind all the great digital marketing campaigns we develop for our clients. Shawn, thanks for joining me again.
Shawn Walker: Yeah, good to be here.
Steffen: After our inaugural episode last time when we talked about attribution, today’s topic is about Google Ads bid modifiers and why you should not ignore them. Before we dive really into the meaty things on a topic, what is or what are bid modifiers, Shawn?
What Are Bid Modifiers?
Shawn: Bid modifiers are really targeting specs that you can have in Google, that you can either just target, so say you just wanted to target 18 to 24-year-olds, versus just observing the behavior of 18 to 24-year-olds. So you can target or you can just leave it on observation and see how it performs, even though you’re targeting multiple ages. And then you could bid on that. We’ll get into that more but essentially, it’s just a targeting parameter that you can use to explicitly target or just observe.
Steffen: And I assume that does not have an impact on the overarching campaign for which you have made certain setup. It’s really as you said, for specific parts, whether that is as you I think mentioned from an audience or a device perspective, or location. I don’t know if there are any other areas that modifiers can be used for.
Shawn: Oh, yeah, absolutely. There’s, I want to say six to seven. Could be wrong. There’s age, there’s gender, there’s household income, which is a really great one that we can expand on. There’s devices, right? And then Google also has certain lifestyle traits or in market traits. So let’s say you want to go after somebody who’s looking at Honda cars for, you know, looking at the brand, let’s say. You can actually target those people. So it gets pretty granular. And like I said, you don’t have to just target that.
You can just leave it on observation, which is a setting in Google, just to see how it performs. And what I like to do is include a lot of targeting and just see what works because sometimes we just don’t know who our audience is. We can take a best guess. But it’s good to throw a bunch of targeting parameters in there and then just see what performs and then you could start bidding to that or explicitly target it.
Steffen: Yeah, it sounds like a great tool to before you actually put something in place, you kind of check your strategic thoughts on something, right? So that what you assume that would perform, you’re basically test driving before it has an impact on the campaign before you’re spending money on that particular part.
Shawn: Yeah, exactly. And you’re not paying any more for it. Just leave it on observation. Check it out. And you might find that your age group might be different or maybe you’re targeting people who have an active lifestyle, when maybe it’s, that’s not your audience. Maybe it’s someone else. Maybe it’s pet lovers or something like that. So it’s good if you have a base, but this is great to kind of sense check what you’ve assumed in the beginning or if you have light data, this is a good way to have more robust data to back it up with.
Steffen: Yeah, it sounds like a great time saver at the end of the day because otherwise, I mean, as I said, you would have to run a test, which means you will have to spend money and you probably would have to either duplicate, you know, an existing campaign existing ad group in order to gear that towards a specific device or audience, for example.
Shawn: Yeah, yeah. You know, I think in some cases it can save you a bit of time. And I think the bigger thing is it gives you more flexibility with optimizations. So, you know, there might be some people listening and saying, Oh, I use these things. But how often are people checking them? Or are you using everything? Because you can have a campaign that’s running smooth but these bid modifiers can add that polish, you know, as we like to say. So let’s give an example. So say you’re fundraising, for example, right?
And we know, or we have data saying that people with disposable income donate more in larger amounts. So what you can do is go into household income and target people who are in the top 50%, right? And avoid the lower 50% because we know their propensity to donate is going to be a bit lower, right? Or leave it on observation and check it out, see what happens. But essentially, you’ll have data in, you know, let’s say a week or two weeks, sometimes it’s quicker than that, to make those intelligent decisions.
Then you could say, you know what, we’re not going to target the lower 50%. We’re just going to go heavy with the top 50. Or maybe it’s so specific, you just target the 10% of people who have disposable income, right? So that’s the power of this. I think a lot of the settings people might know but they’re neglected often. Most of the time when I go into accounts, I don’t see everything used and seldom do I see even half of the things used. Or if they are used, they’re not current, right?
Maybe somebody put a setting in there where they bid down something, you know, because they were pacing hot on their budget one day, let’s say. But now that they have higher budgets, they can be spending more. So instead of bidding down, they should bid up on the groups that are performing well that have a good cost per conversion or a big return on ad spend. So it’s maintenance, right? You have to maintain this, I mean, some accounts I look at every single day, but other accounts, at least every other day. And if you’re not checking bid multipliers on everything at least weekly, you’re really missing out on a lot of efficiency.
Steffen: Yeah. Why do you think some account managers, some paid search managers don’t use bid modifiers as frequently or even don’t check up on them as frequently. Is there a particular reason from your perspective?
Why Some Account Managers Don’t Utilize Bid Modifiers as Much as They Should
Shawn: Yeah, you know what I think? I think the number one thing is people don’t know they’re there, right? Number two, if they know it’s there, they focus on other things that maybe they think, have a greater opportunity to be more efficient, like working on ad copy, adding keywords, things like that.
But I would argue this is just as important because if you’re targeting, let’s say you’re using the correct keywords but you’re not targeting the right people, that can be just as bad as having irrelevant keywords. Where if you’re actually targeting the right audience with broader keywords that might work for you. You know, sometimes we don’t know people’s intent based off the keywords. You know, it’s very broad. So this is a way to actually target people. Let’s say, for example, let’s say you’re targeting someone with like fishing tackle, right?
You might want to target outdoor people, you might want to target people in certain locations, right? Who are by lakes, because you can do location targeting as well. You might want to target devices. So maybe you just want to target people on mobile because they’re on the go, right? So you can offer like coupons to stores, things like that. So it just depends. I’m just, you know, rattling off some ideas here, but essentially, you know where I’m going with this. So, there’s a lot of flexibility.
Steffen: I find the location part really interesting just because obviously, US is being such a large country and if you run a national campaign, even if you run a regional campaign, there are so many individual cities, DMAs, whatever you want to break it down to that can and will perform differently. You might have nuggets where a product that you’re offering finds a really strong audience for whatever the reason is, and then being able to say in what in, I don’t know, in San Diego County, we want to bid up there because we’ve seen a strong response there.
And then compare to that, in I don’t know, in Los Angeles County, we bid down because we’re not getting any traction there, or the cost per is not strong enough. I mean, that’s really getting detailed into optimization and really making sure that every dollar that business spend or an agency and foreign clients spend is working hard to generate a strong return on investment.
Shawn: Oh, absolutely. I mean, think about it this way. If you have a luxury product, like you’re selling, you know, boxes of champagne or, you know, Sterling watches or something like that, you’re not going to target areas that aren’t affluent, right? Or you can but it’s just not going to do well. You want to target cities that have dispensable cash, right?
People who have money to spend on luxury items. Or, you know, if you just want to go broad and you’re targeting national, at minimum, put all the states in there because to me, California versus, let’s say, South Carolina or something like that, they’re completely two different countries in my mind. So you can’t treat them the same.
Steffen: Yeah. And I think that’s when I said earlier probably saves time, that’s where the time-saving effect. And also, the overview in the account comes into play, right? If you were have to kind of duplicate your campaigns 50 times because, or 52 times in order to have one for each state, right? I mean, that’s not only a lot of campaigns in your account, but it also will make it so much more complex to manage all of these campaigns and keep an overview at the end of the day compared to being able to put a modifier on, bid modifier on depending on one campaign, depending on the individual state.
Shawn: Yeah, exactly. So, you know, back in the day, you had to split them out by state if you wanted to do that. But now if you have a similar message, let’s say, then you don’t really need to break it out unless you’re specifically saying something about California versus Texas, you know, versus Florida and you need to have different copy, that’s another story. But if you’re doing national targeting, it does not hurt you just to break out your states by uploading them to locations but not having to build 50 different campaigns, like you said, per state, right?
Steffen: Yeah. Yeah. Shawn, on what levels can bid modifier be applied to? So when I’m talking about levels, I mean, you know, campaign level, adword level, keyword level.
Shawn: Yeah. So I think historically, it was just campaign but now it’s campaign and ad group level. So it gets pretty granular, which is great, right? So we know every theme is a little bit different depending on how you build things out. So getting that flexibility to actually have a bid modifier at the ad group level is fantastic.
Steffen: Yeah. So in situations where an account is either overspending or underspending, how can bid modifier help? Maybe let’s start with underspending.
How Bid Modifiers Can Help With Overspending and Underspending
Shawn: Sure, yeah, I mean, so this is a daily occurrence for me. So, if an account is underspending, typically what you want to do is go find your top performers, right? So you look in your devices, you look in your locations, your income, age, gender, and you start boosting the ones that have a good cost per because essentially, what you’re gonna do is you’re gonna boost your efficiency and spend more, right? So it takes a little bit of manual work, but it has pretty significant outcomes.
Steffen: Okay, on the flip side, how about overspending?
Shawn: Yeah, so typically with overspending you, you know, you do the opposite. So instead of finding your most efficient groups, you find your most inefficient groups, things that aren’t delivering conversions, things that have a poor cost per conversion or low return on adspend and you start adding bid modifiers to it to decrease, right? So you might have minus 25% or minus 50%.
Or if you just see something that isn’t converting at all, you could either exclude it or do a bid modifier of minus 90%, let’s say so you’re not really spending on it at all. So that’s, you know, and really, at the end of the day, you should be doing both. But if you’re having a spending problem, then you end up doing the first thing I recommended, which is boosting and if you are spending too much, then you’re going to have to cut the fat essentially.
Steffen: Yeah. What are the limitations to this feature? Are there any?
Shawn: Yeah, so one of the problems is Google automates a bunch of stuff. If each, you can choose target return on adspend, you could target CPA if you want. And I’ve seen some of these modifications kind of behave a little bit odd. Because if you let’s say you have a target return on adspend, it will actually remove all your device bids because the system is trying to do the lifting for you. It’s trying to find what your best device is, who your best people are. So you’re kind of hands-off if you try to do a fully automated bidding strategy with Google.
And in some cases it works great and other cases it doesn’t, you know? And those cases really are do you have ample conversion volume for machine learning to do well, or do you not? Or sometimes you’re interested in consideration and awareness and maybe conversions are secondary, depending on what business you’re in, where that strategy might not make sense and if you use it, you’re just going to end up shooting yourself in the foot.
So it really, if you know what you’re doing, sometimes it’s better to stay away from the machine learning and do all these tricks. Because machine learning takes a while as well. It’s, if you go into system it will say, eligible and learning but if you know after a week or two of messing around with this that people aren’t converting on mobile phones and they’re in the top 10% income bracket and they’re all female.
I mean, start making those changes, right? Do the thing that you have visibility on. And another limitation I was just thinking about right now is I don’t think it’s clear for reporting, right? If you just have a blank slate in the system and this system is kind of bidding to all kinds of stuff and you haven’t really added any audiences, then you’re not really going to see who your audience is. So I think at minimum, you still want to add everything, even if you’re not using the bid modifiers.
Steffen: I mean, for the machine to learn and to make autonomous decisions, right, it needs data to function well and then to make the right decision. Is there kind of a rule of thumb of how many data points or how many data sets an account should have before you turn that on?
Shawn: Yeah, you know, I’m trying to remember, don’t quote me on this, but I think last I looked there was about 30, it was either 30 a week or 30 a month. I’ll have to get back. But there’s a threshold on how many conversions you need to have before you turn on machine learning or it might be weird. Or also, if you aren’t running every day, let’s say you don’t run on the weekend. or, you know, you’re slow on like Tuesdays and Wednesdays, or, you know, something like that. I don’t think this is the best thing to do with machine learning because it doesn’t factor in every single thing, right?
It’s just going to go off your numbers and start making changes to that. But if you have an account that’s pretty consistent, now I would say, like a Home Depot, for example, we’re always selling tools and nails, screws and all that stuff and it’s pretty consistent, then this might be great because it’s high volume as well. So high volume and consistency is great for machine learning. But otherwise, you know, if you’re a small business, and you’re not a giant, and you see some inconsistency, then it’s probably best to just stick with doing some manual optimizations.
Steffen: So earlier in the show, you said that when we do audits for prospects of a client that want to bring over accounts to us, you see quite often that bit modifiers are either used wrongly or not tended to properly. Could that also be because those people think other optimizations are more important than changing bids and adjusting them based on the various points that you can do that? So geolocation, device audiences, etc?
Building Confidence In Bid Modifiers
Shawn: I think so. Yeah. I think it’s also like familiarity, right? Like, it’s, if you don’t know what, if you’re not using it every single day, you don’t trust it, right? I almost treat it like fishing. Like if I haven’t caught a fish on a lure, I don’t want to use that lure. I don’t feel confident in it. So I feel confident in these bid modifications because I use them all the time and I know they can make a huge difference, right? Just selecting age is a huge thing, right? So it’s, I think it’s one of those cases you have to go through it and see Oh, wow, this actually is pretty big.
And sometimes it has bigger impact than the keywords you add, right? So I think the piece of advice I would give people is think holistically, don’t sit there and say I’m just gonna do negative keywords today or, you know, I’m going to add from the search query report. I think this needs to be part of that everyday optimization, where if you need to spend a little bit of time scrubbing through negative keywords, maybe half that and save some time to do these bid modifications. Or, you know, if you’re doing keyword bids, maybe save some of that time and apply it to these modifications as well.
Steffen: So what I hear from you is basically that sounds like a hygiene task that the paid search manager should literally have on his or her to-do list for daily basically.
Shawn: Oh, absolutely. Yeah, I mean, anytime I’m optimizing an account I go to this. Sometimes even more so right? Because you can’t, well, it’s seldom you can add high volume keywords every single day. At some point you kind of tap out. But this is shifting, always. You know, sometimes you’re underspending, sometimes you’re overspending. It’s rare that you’re perfectly on every single day. So you have to constantly go in there and make changes. And this is, to me, kind of an easier way to go about it.
Steffen: I mean, again, what I said earlier, I think I like this feature a lot because it allows you to test things. At some point like when you manage an account, I don’t want to say there’s nothing else to optimize because, you know, there should always be something to test whether it’s ad copy, landing pages, and so on. But, you know, but the function that you can, for example, test audiences or testing certain geos and see how in device performs in certain situations.
I mean, that gives you such a multitude of opportunities to, again, test drive a strategy or an approach before spending money and having a consistent test and learn approach in place. So if, you know, for the people that listen to this and then so far kind of shied away from bid modifiers or haven’t used them and embraced them fully, how would you set them on the path to get started with them or feel more confident using them? What is the easiest way to make a change on their account today to get started with bid modifiers and as you just said, feel confident with it?
Shawn: Right. So the easy ones to me to recognize are, go look at device first and compare the same campaign and same ad group and see what device does well because they’re always vastly different. So go in there and see If computers are doing well, then you could bid them up. If mobile isn’t doing great, then bid them down. And eventually, you’ll get them to a good equilibrium between mobile, desktop and tablet. I would also do age because age is an easy one. You know, 18-year-olds and 65-year-olds are completely different. So you can go in there and Google is great because you can see the graph, right?
You can see this nice line chart, or sorry, bar chart that says conversion rate, cost per conversion. Just have a look and see if, you know, what audience is doing well. And like I said, the ones that are doing well, bid them up. The ones that aren’t bid them down. And I guarantee just doing those two things device and age to start with, I guarantee that you’ll get at least like a 10% boost. If you do that daily for a week, you’ll get a 10% boost in efficiency. I see more often, but I don’t want to promise anything but at least 10% I could stand by.
Steffen: Yeah, I mean, I honestly have to say and I started with search engine marketing in 2004 as my first job coming out of college and back then obviously CPCs were low and you did not have the opportunity to be so fine-tuned in your optimization approach. Maybe didn’t make so much sense back then because prices were lower and you were able to achieve great results for clients without going so detailed. However, these days where marketplace search engines have become so competitive, prices have gone up because of them.
Even if you look at how brand terms have increased in prices, you know, 5, 10 years ago, you still got your brand CPCs for 5, 10 cents or less even. Nowadays, they can go up, you know, into 50 cents and sometimes without any competition on your brand terms. I feel like all these tools, all these additional features that are available in this case in the Google platform just allow you to be more nimble, to be more focused and to be able to be more detailed in order to make the budget of a client or that you manage account for to work harder and achieve the results that you were able to achieve more easily 10, 15 years ago.
Human Supervision is Still Very Necessary
Shawn: Oh, absolutely. And I think on top of that, there’s two things. There’s other platforms now. Seems like everyone has gone the advertising route, you know? You got Facebook, Twitter, Instagram, Spotify, Outbrain, you know, the list goes on and on and on. So some of that share comes away from Google. So you have to be a little bit smarter. And I think also, it’s important that we have these opportunities to make changes because otherwise if it was just a set it and forget it type of thing, then our job would be obsolete, right?
Why would they need someone who’s a professional and experienced that this when they just click a button that and everything works? But we all know that it doesn’t work that way. I mean, you know, we have self-driving cars that still crash into the walls. So we’re not, you can’t 100% rely on machines. You still have to operate them with the right mind and whoever builds a machine, you know, you’re really trusting that person. So, you know, I think Google is I like to say frienemy because they do help and it’s a great platform and they gave us jobs here, right?
But at the same time, they’re running a business and it’s gonna be in their best interest to make money and but also do well for you. But there’s so many automated things and, you know, they have smart campaigns and Google Express accounts where you literally just press a button and it does things for you. But I’ve seen just, you don’t know what you don’t know. Those accounts always look terrible if we ever acquire them. It just like, it’s sad. But that’s the point is sometimes people just want something quick but quality is really important and that’s where the detail comes in, you get quality this way doing things like this.
Steffen: You know, I think human supervision input is so important. I mean, you know, everyone wants automation and we want that too because it takes a certain part of, that certain tasks that are tedious for our managers away and they can focus more on the meaty things like strategy or finding nuggets in data points and then making decisions based on that. But at the end of the day, even software solutions need human oversight in order to perform better than when you just leave them alone at the end of the day.
I think that’s a good point to end today’s podcast. You know, as this is our second episode for the series Polishing Your Digital Marketing, I just want to say again, you know, we’re looking for great guests to join us here. Guests that might have digital marketing related challenges and would love to come on here, introduce the challenge they have and then we would love to talk through the challenge and identify solutions for the challenges they have. If that sounds like you, send us an email to firstname.lastname@example.org.
You can also go on to the website and reach out to us through that. Just connect with us on LinkedIn. It’s another possibility and tell us about the challenge you have. And we’d love to help you solve the issue. Otherwise, thanks everyone for listening. If you like the Performance Delivered Podcast, please subscribe to us and leave us a review on iTunes or your favorite podcast application. If you want to find out more about Symphonic Digital, you can visit us at symphonicdigital.com or follow us on Twitter at Symphonic HQ. Thanks again and see you next time.