Posted by & filed under PPCChat.

During this week’s PPCChat session, host Julie F Bacchini gathered expert perspectives on brand vs. competitor keywords. Do they generally advise clients/brands to advertise on their brand terms? And do they separate brand term advertising in your reporting?

Q1: Do you generally advise clients/brands to advertise on their brand terms? Why or why not and can it vary? And is your answer different by platform?

I tend to be in favor of advertising on brand terms, but it depends on the situation. @revaminkoff

I generally do advise to advertise on brand terms, but in a separate campaign with them as negatives in all other campaigns. @NeptuneMoon

Favourite answer: it depends, but I don’t immediately cut off brand terms because “organic is covering that” many of my clients will straight up lose thinking that way because Gov website templates are not fast to upgrade and not very organic-friendly. Plus the competitor issue. .@JuliaVyse

If a client needs to control their brand story, if there isn’t a lot of brand loyalty, if there is a lot of competition… those are all cases where it can be important to be on brand terms. @revaminkoff

Minus a couple clients who don’t want to advertise on their brand terms. We always advertise on brand terms. @duanebrown

@NeptuneMoon oh yes, definitely brand should be in its own campaign. @revaminkoff

Also, if a lot of your traffic is on mobile, you want that top of screen positioning. @NeptuneMoon

imo you should show up as much as possible by paying as little as possible. think a lot of folks over the years killed their brand terms and are paying the price as the Max’s overindex on bidding on competitor terms (aka your brand terms) that said, it should be treated differently, reported differently and maintenance should be charged differently @Aaronlevy

Yes, 100% (usually). Now, if they have a super strong organic presence or there is literally no competition, then I can forgo the brand ads. We have to remember in the day of “expanded, AARs, Pmax, etc…” folks end up bidding on brand terms unless they purposely turn off those terms using negatives, which not many do unless business owners are friends (yes, I have had a couple who made sure not to bid on their friends businesses). @Ichasse

I have 4 scenarios: @teabeeshell

  1. New brand, stimulating traffic before organic takes hold
  2. Defense, only if meaningful competition
  3. To promote deals that can’t/don’t surface in organic listings
  4. To model conversions for parallel non-brand efforts

Our SEO coverage on our brand is VERY good — I think on limited budgets, in this case, it’s okay not to advertise on brand. And I am vehemently against advertising on other people’s brands. Seems that we wind up doing it anyway because of the way Google matches search terms to keywords. My bottom line tends to be: If you can spend the budget without advertising on branded terms, do that. You will miss some new customers sometimes, and your conversion rates will be lower, but you are more likely to be getting new customers who you wouldn’t have gotten anyway. @EronCohen

…having said that, we do have some campaigns where we do advertise on our brand, via broad match (non-branded keywords).  The powers that be think it’s a win. I disagree, but I can say it does lead to more new customers than I thought it would. Question is whether we’d have gotten them anyway. @EronCohen

The budget needed for brand typically is VERY small compared to the non-brand campaigns. If it was more expensive, then I would try not to run brand a lot more often. I had a couple big brands who wanted to test turning off brand an in each case, we lost about 30% of top-line revenue compared to running them. You won’t lose it all, but you do lose some, and for the cost, the 30% was too big to ignore. @Ichasse

Usually, check whether someone else is already advertising on those terms and share it with the client, and recommend it if competition is already bidding on those terms. More than search, I find it becomes a necessity when it comes to Amazon. But definitely needs to be reported on separately as it tends to skew the real picture. Sometimes the demand generated from Meta is captured from brand search Google. @alimehdimukadam

I can say that branded search term CPCs tend to be cheaper than non-branded, but not always. I think it’s because people search for our brand and then a service. The “service” part has alot of competition, while the “brand” part of the search term does not. I think the most important question is: Would you have gotten the conversion anyway via organic? In healthcare, I think the answer tends to lead towards YES because you were either referred to our brand or are already a patient. @EronCohen

Brand search is like insurance; it makes sure people looking for you get to the right place, and you can get them directly to a well-optimized page (which may not be the homepage). And in our current world of broad match and PMax, you bet Google is matching your brand name to competitor names and even general search around your specific industry/offering. @robert_brady

I also feel the need to add another point–there are different types of branded searches and they aren’t equal in terms of commercial value. To delineate three of them:

  1. General branded searches where they just name your company and not much else to help you discern what they are looking for.
  2. Branded searches that also contain a product or service
  3. Branded searches that contain likely non-commercial terms (eg – <company name> login)

Of the three above, I think the second one would be most valuable to bid on. @EronCohen

Generally, no, unless they’re having trouble ranking for their own name organically. @williamhboggs

Q1A: How do your views differ on brand vs. competitor terms?

Well, these days you can accidentally end up advertising on competitors’ brands without knowing it as brand terms match interchangeably in Google Ads…@NeptuneMoon

I’m against bidding on own brand terms UNLESS (a) competitors are bidding on them, or (b) the brand does not show up in the top position organically for their own name (this is rare).
Special scenario (c) the client has more budget than they can spend on cold searches (0% SLIS(b)) and just wants to capture as much as possible. @williamhboggs

That being said, it should be a strategic decision to advertise on competitors’ brand names. If you’re not deliberately doing it, then do you best to keep up with competitor negatives. @NeptuneMoon

Running directly on brand terms can be an expensive and fruitless undertaking unless you have a clear landing page or proposition as to why you’re an alternative. @revaminkoff

Though to @NeptuneMoon’s point, Google keeps matching everyone on everything these days…@revaminkoff

defence vs offence! I’ll say how mine are similar – you’ll get the best results from both by being clever. Just bidding on “brand name” or “competitor name” won’t get you much of anything incremental in brand terms, and you’re q score will be so low it’s underground for competitors. bid on review terms, “vs,” specific features, “cancellation” etc. – haven’t seen many people doing it, but could be wise to explicitly exclude [competitor name], then let the max’s pick up more interesting terms. @Aaronlevy

I find competitor terms mostly a miss in my neck of the woods. Typically, if someone is ready to go to a store, they’re not going to be persuaded to swerve to your shop by one search ad. and they’re REALLY pricey! Now for Govs, we bid on all kinds of ‘thematic competitors’ because we’re trying to change behaviour and interrupt misinformation. @JuliaVyse

“Alternatives to” can work really well @revaminkoff

I think we also tend to think that if a person is searching for a specific brand, they have already totally decided on that brand. I don’t think that is always the case, as evidenced by competitor terms sometimes converting. But expectations should be very low for a competitor campaign. Anything you get there, I position as a bonus. @NeptuneMoon

I always ask two questions: @teabeeshell

  1. How likely is it that a person searching explicitly for X brand will see your ad and instead convert with you?
  2. Have we structured copy to appropriately command this situation?

I’d say that for lead gen and auto, I can see it. for QSR, like, try selling someone a whopper when they want a big mac. waste of bucks. @JuliaVyse

I have had a few brands who wanted to go after other brands (mostly because they were targeting our brand, you could tell on purpose). If you have a plan for it and make specific ads for it, I say go for it. Just understand when you do, it will be more expensive CPC wise, so have enough budget. Also, ensure your ads are something that will make them click yours vs. theirs. Why pay more? We have xyz services that others do not, etc…@Ichasse

It is pretty common now to see, but I had an HVAC & plumbing client years ago that used “we answer our phones” (and they really did) in their competitor copy and it was very effective. @NeptuneMoon

I’d usually be pretty bearish on brand campaigns, but after what I’ve been seeing from how Google matches brand keywords to competitors, I’m thinking the same is true for everyone else. So even if other brands aren’t running competitor campaigns, they’re likely bidding on your keywords whether they know if or not. Defensive spending almost feels like a non-negotiable (from what I’ve seen recently). @AnthonyMcDaniel

@AnthonyMcDaniel good point. I’d take cues from Auction Insights. If the competition takes a </= 10% IS, and not consistently MoM, the risk of minimizing branded spend is low @teabeeshell

Q2: What are the most effective ways for a brand to test the impact of advertising on their brand terms? Does this vary by platform or campaign type?

One way is definitely to turn it off for a period of time and see what happens. @NeptuneMoon

@NeptuneMoon we call that holdout testing so you can charge more for it. @JuliaVyse

Also, take into account impression share. 80% on brand is good, but not all brands should go for 90% or 100% impression share because of how expensive the CPCs can be. @duanebrown

We do holdout testing, and we’ve been known to develop scores that include brand term weights as part of the calibration. @JuliaVyse

This can be a little complicated. You have to measure not just paid traffic but also organic to get the difference. It is not as easy as turn it off and just measure the paid difference, since the hypothesis is that organic will pick up the difference. Kind of like using MER to measure marketing spend efficiency with the ratio. You have to measure the impact of brand search over all other forms of conversions to see if any other channels picked up the difference (direct, organic, etc…). @Ichasse

I think we also hear about more lately “these conversions would have happened anyway” and that is often not based on any actual data, but more the agenda of the person making the statement, IMO. @NeptuneMoon

Holdout test and/or geo holdout. We did a handful of these in various formats at tinuiti (sometimes scientific, sometimes with engine help, sometimes uh, by accident). @Ichasse makes a great point at measuring total impact, not just search impact https://www.searchenginejournal.com/should-i-bid-on-branded-terms-in-sem/373258/ @Aaronlevy

@JuliaVyse You work with any catalogers, lol? I ask because that is a VERY common term to see how impactful catalogs are with groups of folks. If they got emails, catalogs, etc… We use holdout testing a lot! @Ichasse

Holdouts are the gold standard IMO, but few test in this way to reach statistical significance. Lower cost alternatives include: @teabeeshell

  • Review Auction Insights – Determine if competitors actually exist, are meaningfully stealing traffic, and represent a threat
  • Review Demand vs. Organic Volume – Determine what percentage of branded traffic you already capture organically

@Ichasse, I did actually! My former agency was YP owned, and every so often on a smaller client we’d get the directory team saying they increased search volume with the ad in the book. things got spicy here and there. @JuliaVyse

When working in search, it is also important to continually repeat that search ads capture demand, they do not generate it. @NeptuneMoon

And I look closely at new markets. We can see where people are searching your brand, and there isn’t a service available, so worth looking into a new location based on demand. not a perfect test, but interesting. @JuliaVyse

So if a competitor gets buzz you might see a drop off in your brand volume @NeptuneMoon

Those catalog folks will defend their catalogs until their dying breath, lol @Ichasse

I’m in a weird spot with one particular client just now where they are KNOWN, like brand recognition and even brand search volume is not an issue. We need to work on getting customers back after some rough moments in the fall. @JuliaVyse

@Ichasse they most CERTAINLY will! god it was so weird and like, we couldn’t argue in front of the client obvs, one team. So in the prep and debrief we were very um…forthright @JuliaVyse

This is why having a metric like MER is very helpful. You can see overall impact of specific programs if you do holdouts on them. You unfortunately also have to take into account latency of spend has a trailing decline (yes we have tested this as well). You can literally stop spend for a month or two on the whole account and only see small declines, but after a couple months it will start to increase more. @Ichasse

Holdouts by geo. It’s something that should be done not just for branded search but also for channels. It would have happened anyway – partially true. By running it – you are ensuring it does happen without leaving it to chance. For DTC & Ecom, the demand you generated from organic social can end up going to marketplaces like Amazon if you are also selling on Amazon. @alimehdimukadam

@Ichasse, and in this short-term, quarter to quarter world, that’s really hard to deal with. You can literally start the test with one exec and have a new CMO by the time the data is valid! .@JuliaVyse

Look into TOS and new customer percentage. @revaminkoff

Q3: Do you separate out brand term advertising in your reporting? If not, why not?

Yes, absolutely. @NeptuneMoon

I show a rollup and separated views. There’s no reason to only do one. @JuliaVyse

The next question is on what you do when you take over an account that has brand mixed in everywhere. @NeptuneMoon

I’m with @JuliaVyse share both rolled up and separated. @Aaronlevy

100% because brand searches are considered customer retention vs. non-brand is much closer to customer acquisition. The return on spend needs to be tracked differently, because they are 2 fundamentally different audiences. If 60-80% of your non-brand are new to file customers vs. only 10% of the brand is new to file customers you have to track the results differently. One needs to be profitable right away where the other can be profitable on the 2nd/3rd purchase depending on the goals. @Ichasse

Brand pulls down the average CPC and CPA so keeping it mashed together in non brand makes the averages look good, but it is deceiving to view it all together for strategic purposes. I think brands should have a clear idea on what it costs them to acquire customers through brand vs non branded terms. So, you can show everything together, but should also have a breakout, IMO. @NeptuneMoon

We don’t break it out as a seperate tab in the paid media report. Our Google Ads Summary report shows in broken out. @duanebrown

I always show two things:

  1. Percent of spend allocated to branded efforts
  2. Percent of conversions (by volume) that come from branded queries

This applies pressure to keep branded spend controlled and highlights how efficient branded efforts are (should be). @teabeeshell

I like to show competitor terms separately too because they are also their own animal. @NeptuneMoon

Yes. Even reporting on the search terms reveal timely insights about what people are actively looking for Could be brand * reviews brand * specific sku. And if there is decent volume – then specific brand * X query can turn to a dedicated landing page or a campaign on it’s own @alimehdimukadam

Yes! This is a must.@williamhboggs

always, though there is always some roll up of total monthly spend, or total search efforts, but the brand/non-brand splits is one of the first splits to discuss. . @Chriskostecki

Q4: How do you handle a situation where a client/brand has their brand terms intermingled with non brand terms in campaigns? Do you pull the brand terms out or leave them where they are? Does this vary by platform or campaign type?

Saw this in an audit yesterday. Always recommend breaking out brand in their own campaign. @duanebrown

I will always split them out. They tend to eat up most of the budget, inflate performance metrics, and again – I’m on the “don’t bid on own brand terms unless needed to be competitive” train. @williamhboggs

I tend to present an ideal state – brand/generic separated – but if things are working, I try to not bulldoze in with changes. Start with sqrs and pulling high performing terms into their own area and work it through. Unless you have a single product client and can move a little faster. @JuliaVyse

This one can be tricky… I don’t like having brand in non-brand campaigns, at all. But it is very common for brand terms to either be in as targets or not excluded as negatives in a lot of setups. And, then the client has the performance expectation of a blended campaign. Which sucks, because it is not what I want to use to drive strategy. It makes non-branded seem more efficient than it actually is. And it can be really hard to split things out and have the client still be happy when performance “drops.” We know it did not drop, but from some clients’ perspectives, they can’t get past the change in the numbers. @NeptuneMoon

The quickest path to highlighting the bread of the concern is the SQR: @teabeeshell

  1. Extract the spend on branded terms, and show this as a percent of total.
  2. Extract the percentage of branded conversions.
  3. Perform any necessary calculations to show efficiency/inefficiency (CPA, CR, etc.).
  4. Determine the role of the platform. Is Google Ads merely a backstop for other platforms, capturing low-hanging demand that they create? Could it be working harder to capture non-brand demand and/or generate its own net new demand?

My goal is always to cap branded spend at/under 15% of total unless there is a specific, declared intention to spend more. (It’s rare that I hear a compelling reason.)  @teabeeshell

@teabeeshell how did you get to 15% as a specific blend? I have seen a variety. @JuliaVyse

I will start to work on separating them out if possible. Beyond @NeptuneMoon’s point about client expectations it can be tough because Google has a lot of data on a blended campaign and starting a new campaign starts the data at zero. Sometimes you can have them measure “new customers” in their own systems to understand what the impact is of that particular campaign, but in the end it is a little ugly and may take some time to work it out and get it setup as it should be. @Ichasse

Let me add this too… Google will absolutely encourage you to NOT take brand out of PMax. And will tell you that most advertisers who take it out put it back. And, putting it back is not a guarantee to get back to previous performance. Ask me how I know this…@NeptuneMoon

Most probably do, because brand is a cash cow for any brand with a pretty well known name in their segment. lol @Ichasse

I think we should also acknowledge here the gap between what we as data driven marketers think or want for performance and what many businesses are ok with. We see situations where we think we can do more or get better or more efficient results. But for a lot of businesses, how it is running right now is plenty fine. @NeptuneMoon

Never underestimate your contact not wanting to have to explain to the powers that be how the numbers they have been using are not accurate too… @NeptuneMoon

@JuliaVyse My 15% branded (max) somewhat a bespoke percentage. There are fringe cases where extreme competition and/or Amazon activity dictates higher. In the absence of either, this historically has shaken out for me that I can demonstrate strong performance without over-indexing in spend that sandbags performance. Many clients appreciate that this is a non-issue, and they can respond accordingly to entities that might reach out directly to them, trying to poach their (my) Google Ads business.@teabeeshell

I also ensure that branded spend is NCA-only. Of they choose to opt into Amazon, they pay that tax knowingly. I want branded spend to capture truly new, not folks being lazy, repeat purchasing on the back of a paid ad. @teabeeshell

For example if the client is hyper fixated on their CPA and their campaigns are all blended, it can be very, very hard to change that. They might not be able to see past the “old” CPA figure and feel may feel like you broke it. When in reality you just separated it for a clearer view of their true costs. @NeptuneMoon

Separate it immediately once you have identified it – regardless of the platform @alimehdimukadam

I have been brought in as a consultant or strategic set of eyes on accounts and recommended breaking out brand and keeping it clean, only to be told “we can’t do that, the client will freak out when they see the CPA that way.” And I always say, ok, but it doesn’t change the underlying reality…@NeptuneMoon

Separate separate separate. @revaminkoff

I combine brand and non brand in pmax until we have some conversions, then I will split brand out. That way the damned thing knows what a customer looks like. @DanThies

separate with negatives, pause brands and set to non brand, build a brand campaign, negate brand terms from non brand campaign. @Chriskostecki

Q5: What is the best or worst advice you have seen or been given relative to advertising on brand terms?

Running some automated rules or Max conversion bidding strategies on branded search – you may end up finding sometimes it’s not an auction at all. The machine will willingly take in your inflated bids. @alimehdimukadam

Simply, “You have to do it.” No, you don’t! It’s relatively simple to portray how it’s often a waste of spend, at least at historical levels. Walk through the customer journey, review SERPs, understand the organic contribution, and operate accordingly. @teabeeshell

Anything that is in absolutes should give you pause. This topic was inspired by content I have seen recently saying you should not ever spend on brand terms. As though that was universally good advice… @NeptuneMoon

Self-bidding. I’ve had clients here are there who are “very into Google” and have their own strategies in mind. Please put brand terms across 7 differently themed campaigns with the same location targeting. uuuuugh. @JuliaVyse

Also, “PMax over-indexes into branded, and there’s no way around it.” Also nope. I’ve run PMax campaigns that are 99.9% non-branded since Day 1. It’s very possible to insulate these efforts in a non-branded fashion. @teabeeshell

“Run brand at target impression share” (and at any price). If you’re in a really competitive category this can get crazy competitive, and there are better ways to bid. @revaminkoff

PMax really wants brand in it and that pressure can feel intense, especially from reps. But you can absolutely run PMax with brand excluded. @NeptuneMoon

Competitive also generally equals expensive – this is true. @revaminkoff

I also think a lot of discourse around brand terms is from an e-commerce perspective. So if you are in lead gen, take with a grain of salt or at least look closer before you apply or abstain. @NeptuneMoon

The worst advise is always to run search as display expansion, regardless of what type of keyword it is. @Chriskostecki

PPCChat Participants

Related Links

Stop wasted ad spend with Karooya

Stop the wasted ad spend. Get more conversions from the same ad budget.

Our customers save over $16 Million per year on Google and Amazon Ads.

Leave a Reply

Your email address will not be published. Required fields are marked *