Posted by & filed under PPCChat.

This week’s PPCChat session was inspired by the Google Ads Expert Series “Welcome to 2022” presentation. Host Julie F Bacchini sought PPCers reactions about this presentation, what are their thoughts about having only 1 RSA per ad group, opinion about enhanced conversion and more.

Q1: Did you watch any of the “Welcome to 2022” presentation from Google Ads? If so, what are your general thoughts and reactions to what was shared?

I watched it, of course! I am here for you all to watch and digest these things. It was not all that surprising, really. Automation and “simplification” will be the mantras of 2022, I think. A few solid answers on match types and manual bidding staying. @NeptuneMoon

For the first time in several years I did not watch. I caught some recaps, but this timing is for the birds. Who said presentations all January? cause they suck! @JuliaVyse

Noooooo. @stevegibsonppc

Yes, I watched it and for me watching it is more of making sure I am ready for the direction Google is moving in. This allows me to make sure brands I work with are getting the most out of what Google has. Do our personal thoughts matter at this point? lol @lchasse

For once, without sarcasm, I missed it as the invite I think went to spam. @JonKagan

I didn’t watch but I bet it was about automation. @jord_stark

Did not know it was happening. Did I miss anything? @duanebrown

No, I read your Twitter thread and your article on it instead. Much more enjoyable than a 80min stream. @C_J_Ridley

Q2: Google Ads clearly wants advertisers to “simplify” their account structures by using broad match + smart bidding + RSA and including DSA (dynamic search ads) in ad groups. What are your thoughts and/or experiences with this suggested method of organizing accounts?

It’s not a bad approach for an smb with a squarespace site, but DSAs are REALLY dependent on the content and structure of your site. I do not think this is widely applicable. @JuliaVyse

1 RSA. ONE? 1. sirs, dearly beloveds, sincerely yours, I am NOT building new ad groups for every. single. promo we run on brand and broad-theme keywords all year. are. you. kidding. me. @JuliaVyse

I see mixed results with these structures already – in some accounts this works really well. In others not so much. The biggest takeaway is to test and not assume outcomes. @navahf

I expect we will be hearing about “The Golden Triangle” A LOT this year (image is slide from presentation). Expect a big push for smart bidding + broad match + RSA. Not thrilled for my low conversion volume accounts…@NeptuneMoon

They have been kind of wanting this for a year plus now. The good news is it is actually starting to work well. I am still not a huge fan of broad match, but in some cases it can work well with “automated” bidding. The days of complex ad structures are numbered. @lchasse

Inherently, I hate the concept and I don’t trust Google. That being said we tested it lightly last summer and worked concerningly well @JonKagan

Scary – automation on speed. Was any audience overlay suggested?? @TheMarketingAnu

Simple makes sense to not have a mess to maintain. Since we use DSA as a sweeper campaign, better on its own. Broad does seem to work on more “commodity” type products. @duanebrown

On ratios of entities: RSAs: Agree with 1 RSA per ad group, just not one ad per keyword. Keywords: I still think there needs to be exact match but having broad can actually outperform due to auction price. DSA: This is a trust fall too far for me. Extensions: yes. @navahf

The recommendations are fine for larger accounts with more volume, but for smaller accounts who don’t have deep pockets, we can’t afford to run broad match for months on end while Google figures it out and we add negatives like crazy. @marksubel

Well… Google can want all sorts of things, but I’ve never taken their garbage advice at any point in the 16 years I’ve managed campaigns, and I’m not going to start now. @stevegibsonppc

I’ve only implemented pure broad match, DSAs and smart bidding a few times in lead gen accounts and all I’ve got in returns is high CPAs, low CVR and a hefty list of search terms to review. @C_J_Ridley

I expect us to see this shoved down our throats by Google Reps. I’m not a fan of the structure as I think more than 1 RSA is needed, and until we have more data readily available and more control over campaigns, I don’t see much success coming from such structures. @C_J_Ridley

Broad match with automated bidding works well when you have a lot of data. This strategy has helped bigger accounts but is a recipe for disaster for smaller accounts. I use RSA but I do pin my qualifiers. I use DSA as a keyword mining tool. @sonofgorkhali

Q3: How often do you check your SQRs (search query reports)? Google Ads suggested checking them EVERY SIX MONTHS for their new simplified account structure…

I used to do it weekly but I can’t see anything anymore. @timmhalloran

We check SQR’s weekly and sometimes more regularly for accounts that seem to be looser on Google’s side. @marksubel

Usually 3x a week, sometimes less for bigger, more established programs. Every 6 months you say? tell that to auto co’s who litchrally create new search terms with each product launch. who came up with this? @JuliaVyse

I check them every 6 months since that is what Google recommends… Okay, no way I do that. When launching I generally check every week until I am seeing things settle, then it is about once a month. Wonky things can pop up due to news cycles trends, etc…@lchasse

Multiple times a day…checking every six months sounds like an April Fool’s joke (but it’s January) SQR is clutch for #PPC management + collaboration with other teams (#SEO #social #analytics) – I’m not ignoring it XD. @navahf

EVERY.SINGLE.DAY I don’t always do anything with the report i look at – depending on volume and relevancy but at least I check. @TheMarketingAnu

For mature accounts under, say, $10k/mo, 1-2 hours every quarter is probably fine for actionable data. @ferkungamaboobo

At account launch I check very frequently. As things align more with what I am actually targeting I move to a bi-monthly or monthly basis. @jord_stark

I’m sure they’d WANT me to check them every 6 months. More money for them. How often I check them depends how long I’ve been managing the account. The longer, the less frequently I need to do it. @stevegibsonppc

I check SQRs weekly. Interestingly, one of the case studies shown had SQR checking happening much sooner in the process. @NeptuneMoon

I usually take Google’s recommendation then double it. So I only check my SQRs once a year. But in all seriousness we check them frequently depending on data volumes. If a new launch then we check daily. For consistent campaigns with history we do it less often. @AndrewPPC

We check SQRs weekly. We have a script that runs (from @ThatSearchGuyNL) that shows us trending search terms. We start there and then dig in. @beyondthepaid

What drives us nuts is that @GoogleAds can’t figure out yet a search query searching for a location OUTSIDE our target area. How is that not possible? We have to add geo-negatives like crazy sometimes. What a waste. @marksubel

We do minimum weekly just to make sure irrelevant searches aren’t getting in. With Google making changes on SERP and how shopping ads work. It’s a safety measures. @duanebrown

As Reggie would say “Well, it depends.” For a small volume account once a fortnight maybe even once a month if I feel good about keyword discipline. For bigger accounts once a week maybe even twice if it’s early on or going through a critical time. @C_J_Ridley

I check SQR daily and weekly. Daily if clicks >10 and conversion is < 0.10 Weekly if click = 0 and impression is >100 This helps to cut down a lot of noise, especially with a larger media budget. @sonofgorkhali

Q4: Another recommendation was to only have 1 RSA (responsive search ad) per ad group (and no ETA or expanded text ads). Thoughts? Experience with doing this if you have?

This was a suggestion from a Google rep I spoke with – basically a lot of “what’s the point of doing it differently? what are you gaining? what are you actually doing with ad copy data?” I had 0 defense and loved the clarity. @ferkungamaboobo

I’ve been hearing this from Google for awhile. If this is the case I would love to hear what their best practice is for testing ad copy. Another RSA ad? Guess not. Add headlines to the existing RSA? What if you don’t get enough impressions for performance data? @jord_stark

We are doing this now for newer accounts and really don’t have a good feel for how well it’s working to be honest. We just know everything is moving to RSA’s so for new accounts we just do it at this point. I think experimenting with “pinned” headlines will be key. @marksubel

This got a “hell no” from me. I’m making sure all clients have ETAs in their accounts now so the RSA can learn. That said, I agree with one RSA. There are standard ads that outperform ETAs still – not hurting winning creative. @navahf

We try to do 2 ETAs and 1 RSA / ad group when ad spend has it make sense. Otherwise it might just be 1 or 2 ETAs @duanebrown

I do agree that multiple RSAs per ad group aren’t strictly necessary – but i put a post about this and some ppl did give me good testing scenarios as to why to have multiple. We will need to drop ETAs all together soon – so maybe start testing that? @TheMarketingAnu

The “one RSA” idea makes logical sense if you believe Google’s AI works. (More sense than having more than one.) @stevegibsonppc

We are testing 1 RSA and 2 ETAs currently. If ETAs do well (which in many cases they are beating RSAs), we’ll leave them live as long as possible. In future, we will test 2 RSAs, 1 with pinning and one without. @beyondthepaid

Let’s be real for a moment, shall we? ETAs have been getting less & less impression share for a while now. If you have an RSA in the ad group, it gets the lion’s share of impressions. This is the end of testing as we have known it. More data on RSA was promised. @NeptuneMoon

This only works for me if G essentially replaces ads with extensions. ie: someone searches for Burger. am I just changing the headlines for my RSAs depending the calendar? or a new ad which gets a learning period delay for a 2 week promo? mess. @JuliaVyse

I have at least 1 RSA in each ad group. I also have the legacy ads that still outperform ETAs! I run whatever works the best and continue to test. I am however noticing Google is forcing the issue with showing RSAs more than other ad types. @lchasse

Aside from seasonal/promotional text for easy ad on/off automated rules, I don’t fully understand why you would want more than one. You can’t get a great A/B test betwen 2 bc too many variables, so might as well just put your diff ideas into the one RSA, right? @PPCKirk

My work is in the finance industry, where CPC is insane. There is no way I am running one RSA and not ETA. Many times I run multiple RSA with pinned headlines. I need to pin qualifiers – I don’t want to spend $40 a click to drive traffic that doesn’t convert. @sonofgorkhali

Q5: Enhanced conversions will be a big push in 2022. Have you used them yet? What are your thoughts on them? Here is the documentation on Enhanced Conversions: support.google.com/google-ads/ans…

Not at the moment. my main conversions are foot traffic and web traffic to sites not permitted to use tracking pixels. it’s a whole thing. @JuliaVyse

I’m in love – they have helped a lot of tough regulation industries regain access to data and shore up conversion gaps. The match rate has been pretty fantastic IMO. @navahf

We’re finishing getting them set up on all client accounts. Don’t have them included in conversion column yet but once we know it’s working and setup correctly …. We will start using it as our source of truth. @duanebrown

Yes, This one goes into the win column for sure for clients. I don’t have it running with every client, but with the ones we have setup now, it has been helpful already. @lchasse

Q6: What has your experience been with broad match lately?

Still reluctant to use broad match in anything other than RLSA campaigns. We get far too many consumer-focused queries for our B2B clients with broad match. @beyondthepaid

Honestly it’s exact that I’m worried about. @JuliaVyse

I have been hard on Google for broad match (rightfully so), but lately I have seen it work in some accounts (only with automated bidding though). You do need to pay attention to negatives closely and set a testing budget to see if it will work. @lchasse

Even though I do not have any broad match keyword running, it sure feels like I do… To be fair, I am also not using their recommended trifecta of broad match + smart bidding + RSA. I am using RSA all the time and smart bidding sometimes. @NeptuneMoon

Really good tbh. Broad match the match type isn’t the bad guy we remember from years ago. Broad concepts without proper targeting and negatives are. @navahf

Bidding on “dinner” vs “restaurants with gluten free options” represent two very different levels of search intent scope. Building in transactional intent into the syntax of the keyword allows broad match to do its thing and bring in cheaper clicks. @navahf

I do think is is VERY interesting that audiences were not mentioned at all in this presentation. Maybe you can shed some light on this for us @adsliaison – seems like a big oversight? @NeptuneMoon

A year ago I would have never considered it. Now it performs quite well. Stay on search terms for sure though. @jord_stark

Honestly, I don’t understand how broad match can even work. This is especially true for ecommerce. Until Google gets halfway decent at sending people to the correct landing pages, broad match is pretty useless for e-commerce. @CJSlattery

Like if I broad match “baby shoes” that’s going to trigger on “size 6 baby shoes” or “blue baby shoes” and if I’m sending them directly to the all shoes CP that converts at a meaningfully lower rate than the correct LP for that query. @CJSlattery

Kind of a mixed bag w/ Broad match. Larger budget & clients, can sometimes work ok. For smaller budget and clients we feel it leads to too many clicks the client can’t afford for the budget. We have to do SQR negatives like almost daily with it. Work in progress IMO.@marksubel

Broad match works well if you have a plethora of data and a bigger budget. We do run broad match, but with all guard rails, you can think about. They not only drive volume but help RAM. @sonofgorkhali

Q7: What are your biggest questions about Google Ads in 2022? What would you like more clear guidance on?

Best practice for testing new ads given their current recommendations. @jord_stark

Any timeline on when more data on RSA performance will start being available? Are audiences being demoted in importance for targeting? @NeptuneMoon

Honestly, a crystal ball of what changes will be coming which is not necessarily possible. Other than that, just having more data on our campaigns, like Performance Max. On my end taking a class on GA4 as well to make sure I can support merchants. @lchasse

RE: max conversions or CPA Max bidding, can @GoogleAds actually get better performance w/ smaller volume? Most times we see it doesn’t. So not clear on how much data it actually needs to start working. They recommend it sometimes with like 1 conversion a week. @marksubel

PPCChat Participants

Related Links

Leave a Reply

Your email address will not be published.