{"id":17500,"date":"2021-09-29T17:02:45","date_gmt":"2021-09-29T11:32:45","guid":{"rendered":"https:\/\/www.karooya.com\/blog\/?p=17500"},"modified":"2021-09-29T17:02:46","modified_gmt":"2021-09-29T11:32:46","slug":"ppcchat-google-ads-and-their-latest-announcements","status":"publish","type":"post","link":"https:\/\/www.karooya.com\/blog\/ppcchat-google-ads-and-their-latest-announcements\/","title":{"rendered":"PPCChat | Google Ads And Their Latest Announcements | 28th September 2021"},"content":{"rendered":"\n

Google has made two announcements lately. The first one is about Keyword Matching<\/a> & the second one is regarding all ad accounts being migrated to data-driven attribution (DDA)<\/a>. This week’s PPCChat discussion was based on these two announcements. PPCers expressed their reactions to these two announcements, do these announcements change their thinking about attribution for Google Ads campaigns & more. Hosted by Julie F Bacchini<\/a>, here is a screencap of this week’s PPCChat session.<\/p>\n\n\n\n\n\n\n\n

<\/div>\n\n\n\n

Q1: What are your thoughts\/reactions to this announcement?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Well, I wrote an entire blog post about this topic… neptunemoon.net\/2021\/09\/google\u2026<\/a> It has gotten a lot of views over on LinkedIn, including 20 from people at Google. @NeptuneMoon<\/a><\/p>\n\n\n\n

This pretty much summed up my reaction. @beyondthepaid<\/a><\/p>\n\n\n\n

At this point not surprised. While the machine has without doubt improved in semantic understanding, they way they plan to execute does not make sense. Let us weed out negatives – so the bot can learn from us!!! @JuliaVyse<\/a><\/p>\n\n\n\n

The example of the very, very specific car part query matching to the broad query of “auto parts” was particularly head scratching. @NeptuneMoon<\/a><\/p>\n\n\n\n

I have to perhaps be the fly in the ointment. I’m not concerned with the KW announcement, & actually think it’s a good change. It makes sense to me to prefer keywords that match search terms (before ad rank), & will probably keep our accounts a little cleaner. @PPCKirk<\/a><\/p>\n\n\n\n

Need to do more frequent SQR analysis moving forward. @Reddy20449<\/a><\/p>\n\n\n\n

I mean at this point, they are all the same announcements. We no longer have control. @JonKagan<\/a><\/p>\n\n\n\n

I come at this from a B2B standpoint. We get enough bad matches without this change; it’s going to be worse now. Example: G matches “mobile monitoring” (an IT service) to “cell phone monitoring” (something parents do w\/kids phones). Not good. @beyondthepaid<\/a><\/p>\n\n\n\n

The match change isn\u2019t at all surprising. However the lede is buried. They really have started to show how far they will stem a term and that\u2019s the concern. Mobile and phone aren\u2019t always synonymous as someone noted yesterday. @armondhammer<\/a><\/p>\n\n\n\n

The match problem Adjusting matching allows Google to manage inventory and optimize yield of its advertising machine If only if only, Google would have taken all the input from marketers who voted with their negative keywords But yield was higher via opacity. @soanders<\/a><\/p>\n\n\n\n

If we agree the future is keyword less and all about audiences targeting… a la Facebook. Then this is just one step of 12 towards that future in 3 – 5 years. @duanebrown<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q2: Is this going to change the way you set up and\/or manage your Google Ads accounts? If so, how? If not, why not?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Not sure yet. I’m already using a lot of broad\/phrase + audiences to hone traffic. Will have to test and learn. @JuliaVyse<\/a><\/p>\n\n\n\n

We’re going to need to prioritize SQRs more than we even do now. @beyondthepaid<\/a><\/p>\n\n\n\n

Ever since exactish match has been the norm, my structure and management has moved to less keywords added and more keyword monitoring. This just reinforces that. @armondhammer<\/a><\/p>\n\n\n\n

I think this is a good change for our account organization because it will more accurately funnel search terms to the matched keywords we have set up (rather than randomly based on ad rank). I support that (if I’m understanding this correctly). @PPCKirk<\/a><\/p>\n\n\n\n

I hate to say it is going to get me to change anything, but I may be on the verge of no longer segmenting adgroups my matchtype. @JonKagan<\/a><\/p>\n\n\n\n

Testing broad now, if considering it, while at least the 0 click data is available to get an idea of the matching G Ads thinks is relevant. More SQR monitoring. @NeptuneMoon<\/a><\/p>\n\n\n\n

How matching affects campaign organization The key thing we need to find is liquidity in our campaign setups. We need the right data to flow so the machine can optimize on it. And we need to put up semantic walls to defend for example the brand. @soanders<\/a><\/p>\n\n\n\n

Not yet but in the future I’m sure it will. Our team is still taking it all in. Plus will be interesting to see more real work impacts in client accounts. @duanebrown<\/a><\/p>\n\n\n\n

As with all things tech. What is happening in the real world and ad accounts vs what Google wants to happen. This is tech and not someone doing the matching by hand… it’s never going to go 100% like they planned. @duanebrown<\/a><\/p>\n\n\n\n

As a Google Ads account manager, we’re all used to having to change the ways we set up and manage accounts. This is just another one of those instances. I will become using more negative keywords to minimise cannibalization between ad groups or campaigns. @C_J_Ridley<\/a><\/p>\n\n\n\n

I will also likely look at using different match types at different points of the funnel. Broad = TOF & Retargeting Phrase = MOF Exact = BOF. @C_J_Ridley<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q3: Do you have questions about how this will work that you would like to have Google Ads provide more information or explanation about?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

The aforementioned inventory question. Specifics on crossover keywords with multiple possible intents (and how to guide the machine). How we can provide feedback on bad matches beyond our negative lists. We want the machien to be better too. @armondhammer<\/a><\/p>\n\n\n\n

Will foreign language queries match to search terms with this? Because I found them matching to my exact match terms in campaigns only targeting English. @NeptuneMoon<\/a><\/p>\n\n\n\n

What are the effects of brand-only search terms being triggered by broad KWs? All of these KWs are other brands – is it just a case of excluding as I go or will Google do anything to prevent branded searches? I’m also not running Competitor campaigns. @dylanppc<\/a><\/p>\n\n\n\n

Yes, but there is a laundry list of questions for other things they haven’t answered me for years @JonKagan<\/a><\/p>\n\n\n\n

Following on from your post about having other languages match for your English keyword, I would like to know if this is due to the latest change and if so, if this is an intended result or an unexpected byproduct. @C_J_Ridley<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q4: What are your thoughts about all accounts being automatically migrated into data driven attribution (DDA) – with opt out as an option?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Personally, I think this is a good move. Curious how well it really works in accounts\/campaigns with low conversion counts though. @dan_patterson<\/a><\/p>\n\n\n\n

Makes sense to increase adoption faster. Will impact SMBs and those who aren’t paying attention most… assuming they don’t want their account to have it due to small data set. I think it’s a good move to get us all off last click for those remaining. r\/ppc is happy @duanebrown<\/a><\/p>\n\n\n\n

Here is what I told @MarketingOClock<\/a> last night: @PPCKirk<\/a><\/p>\n\n\n\n

There’s some great stuff there, and some not so great stuff. Last click overvalues adwords generally so happy to move away from that. But DDA uses some interpretation, and that’s potentially less reliable, especially for smaller advertisers @armondhammer<\/a><\/p>\n\n\n\n

I am happy to see an announcement that has an option to opt out. Automatically migrating existing campaigns seems like an interesting choice? Why not just push it like crazy in recommendations? Concerned about DDA for lower volume and long conv. time accts though. @NeptuneMoon<\/a><\/p>\n\n\n\n

We already have credibility issues with attribution from what once was decently accurate, and now is more an estimate. DDA could make that worse. If you have someone that gets 10 leads a month and DDA estimates 15 from adwords – not a good look. @armondhammer<\/a><\/p>\n\n\n\n

In general I’m a fan of DDA but many of our clients were opted in as of yesterday (if not sooner) with no warning. We need to be able to give clients a heads up that things are changing. @beyondthepaid<\/a><\/p>\n\n\n\n

I like that there is an opt-out option, even if I may not use it. @TheMarketingAnu<\/a><\/p>\n\n\n\n

We’ve also seen what happens with this from FB. View thru conversions are 80% BS, and 20% real. They’ve effectively disappeared and it’s thrown the whole system into a tizzy. I fear the same with DDA @armondhammer<\/a><\/p>\n\n\n\n

I’m dubious about the performance of DDA within smaller accounts, as Google has provided little data (irony) on why they deem it suitable to remove the data volume threshold now, or what the expected impact this may have on smaller accounts. @C_J_Ridley<\/a><\/p>\n\n\n\n

I have migrated most (if not all) of my conversions away from Last-click though. I’m currently choosing Data-driven (when enough data) or Position-based as the successors. @C_J_Ridley<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q5: Google Ads is also removing the prior data thresholds for using DDA (current documentation lists 30,000 interactions and 30 conversions in 30 days as the minimum). What are your thoughts on this?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Was having a convo with some of my coworkers this morning. It seems as though data integrity could be a problem with no thresholds. @beyondthepaid<\/a><\/p>\n\n\n\n

Given the wide variety I’ve seen for conversion rates on lead based landing pages, I worry that using someone else’s data to estimate mine will lead to real issues. A lead gen that offers a free T-shirt is very different than one that asks 5 qualifying questions. @armondhammer<\/a><\/p>\n\n\n\n

Change is in our with lower 30 conversion\/month down to 15 to get campaigns over to smart bidding. Net positive unless you hate smart bidding. @duanebrown<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q6: Does this change your thinking about attribution for Google Ads campaigns? If so why and if not why not? Will it change what you\u2019re doing now or in the near future for attribution?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

I want to understand in better detail how this will work for lower volume accounts. I would love to use data driven attribution, but want to understand where the data is coming from or how the tech has evolved to not need the previous thresholds. Just explain it. @NeptuneMoon<\/a><\/p>\n\n\n\n

Hoping to be able to use the “Model Comparison” in Google Ads to see how the data would change before actually switching to DDA @_RileyDuncan<\/a><\/p>\n\n\n\n

People drop $3M on a superb owl ad, and have nowhere near the data we do for our $5K a month advertisers. It’s worse, but it’s not always the end of the world. @armondhammer<\/a><\/p>\n\n\n\n

Not going to change much beyond a couple small ecom clients and having to get them over to DDA. @duanebrown<\/a><\/p>\n\n\n\n

It won’t change anything except make me monitor the DDA in small accounts and check how it performs. I’ll definitely be looking at the attribution model comparisons more often. @C_J_Ridley<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

Q7: What questions do you have for Google Ads about this change?<\/strong><\/h4>\n\n\n\n
<\/div>\n\n\n\n

Sent my questions yesterday via Ginny’s Twitter announcement but nothing new has come up yet. @duanebrown<\/a><\/p>\n\n\n\n

I feel like my question is always the same. How can we provide feedback to the system? I know it’s web scale, but it needs to be more than a monolith. @armondhammer<\/a><\/p>\n\n\n\n

What data do you have to support the reasoning behind removing the data volume threshold, and what do you recommend to account managers that are now auto-switched to DDA and have seen performance significantly dip? (As I suspect this may happen to some people) @C_J_Ridley<\/a><\/p>\n\n\n\n

<\/div>\n\n\n\n

PPCChat Participants<\/strong><\/p>\n\n\n\n