PPC Streamcap – PPC Campaign Settings For Success

This week, with @adCenter watching in, Matt Umbro (@Matt_Umbro) hosted a lively PPCChat with the theme of “PPC Campaign Settings For Success.” The following is the transcribed Streamcap from the live chat:

Q1:What is your preferred method of ad rotation: “optimize for clicks”, “optimize for conversions” or “rotate?” Why?

  • Rotate for testing. – Paul Kragthorpe (@PaulKragthorpe) +
    • So I can see the data & make my own choice. – Katie Saxon (@ksaxoninternet)
  • Absolutely rotate. It gives you the ability to evenly test ads against each other whereas the other options don’t. – PurePPCCom (@pureppccom)
  • Rotate – so that I can test ads accurately. – John Lee (@John_A_Lee)
  • Mostly try to work on ‘rotate’ for ad optimisation – ensuring a fair test all round but must be kept on top of! – Anna George (@AnnaGeorge)
  • I like to manually decide based on the results. If testing I start even rotation. – Mark Kennedy (@markkennedysem)
  • “Rotate” all the way. Only way I get a true sense (and can analyze) what is working. Don’t want to give that control to Google. – Matt Umbro (@Matt_Umbro)
  • Rotate Evenly as default. I like to gauge ads and make the call to pull them myself. – James Svoboda (@Realicity)
  • Rotate, Rotate, Rotate. Can’t evaluating testing w/skewed data. – Crystal Anderson (@CrystalA) +
  • Rotate with human control. Can’t a/B test with the others, so should always be on rotate imo. – Eloi Casali (@Eloi_Casali) +
  • Rotate! If using ad level URL’s, test each url beyond CTR/CPC, and see which ads are driving more conversions/higher ROI. – Joseph McConellogue (@DigitalMrktr)
  • Even rotation, always. The software can’t make value decision on the quality of a click, but w/post-click data analysis, I can. – Theresa Zook (@I_Marketer)
  • Optimize for conversions when enough data is available after allowing ads to rotate. Our goal is to optimize to conversions. – Nicole Mintiens (@Tregesy)
    • Agreed, but do you trust that Google and/or MSN feels that way also? – Matt Umbro
  • Just like everyone else, rotate! Too many factors to let the googlerobotmonster make choices for me. – Aaron Levy (@bigalittlea)
  • For testing purposes it’s best to rotate ads evenly in order to determine the best performing ads. – Brian Gaspar (@BGaspar)
  • Also, revenue isn’t fixed for most of my clients, so good to optimize based on ROI of indiv ads & not just use # of convs. – Claire Remmetter (@cremmetter)
  • Initially rotate evenly ads to gather data. – Bryant Garvin (@BryantGarvin)
  • Letting Google optimize for me is like letting the gov’t choose which laws to pass: No thank you. – Tally Keller (@tallykeller)
  • Rotate: first you need data, then you may optimize. Not optimizing while testing. – Bart Schuijt (@BartSchuijt)
  • Q1a: Aside from the obvious reason ($$), why do you think Google is so intent on having you optimize for clicks and/or conversions?
    • It is the assumption that an ad with higher CTR is more relevant and therefore will give better user experience. But this isn’t always true. – John Lee
      • B/c unfortunately there are a lot of non-testers out there. – Crystal Anderson
    • Interestingly my ‘Google Professional’ asked why I ‘rotate’ and not optimise – didn’t understand why I want to split test. – Anna George
    • Automated tools can be good, this is one that I feel is often best left unused. Main problem with this tool is incomplete environment data from ad to all analytics conversions. – James Svoboda

Q2: How has the advent of advanced Location Targeting on Adwords affected your campaign targeting?

  • I hate that they took away a custom shape. – Mark Kennedy ++++
  • Frankly, I don’t trust any of it. Get too granular and you lose audiences. All based on IP; to a point IP’s guess on locations. – Aaron Levy +
  • Find that the feature has limited my ability to pick and choose which areas to target and/or avoid. – Brian Gaspar
  • I like the loss of the custom shapes. Cld never convince clients that wasn’t the way to really find & serve customers. – Theresa Zook
  • Geo-targeting is good if are hyper-focusing on a local product or service. But it can become problematic for larger campaigns. – Bryant Garvin
  • I always worry about what I am missing out on, and not what I am targeting! I find kw targeting more effective. – James Luty (@jamesluty)
  • Targeting using physical location has trimmed substantial amounts of unqualified traffic from campaigns. – Joseph McConellogue
  • Custom shapes never really worked. If your “shape” touched a city, you targeted the whole city. – Aaron Levy
    • True, but It was good for the small clients outside the metros, though. Franchisees and such. – Mark Kennedy
      • Target the entire metro, exclude specific cities you don’t want. – Aaron Levy
  • It’s been great for segmenting GEO-Keywords from GEO-Targeted Campaigns. Often have different CTR & Conversion metrics. – James Svoboda
  • All custom shapes have to be converted to the new settings by 2012 or Google will pick for you. – Mark Kennedy
  • Though, the ability to show for your geo-terms based on geo inclusion in query I’m a fan of (in some cases of course). – Crystal Anderson
  • I still find it very difficult to 100% trust the settings, I’m always testing to make sure they are serving as they should. – Andrew Baker (@SEOEdinburgh)
    • Yes, I recently discovered 95% of impressions were outside targeted area. But spoke to Google and they admitted tech error & are ‘working on it…’ (slowly, I might add). – Claire Remmetter
    • Yes, I spot-check accounts occasionally, just to check for geo problems. – Theresa Zook
  • Like all targeting options this is a “control” lever. It has + & – and you need to know when to use it & when to avoid. – Bryant Garvin +
  • Targeting using physical location resulted in lower CPC’s, higher CTR’s, on site engagement, and ROI. – Joseph McConellogue
  • Haven’t used the new geo-targeting options myself yet. – Dennis Petretti (@Denetti)
  • I think it’s like anything. Give it a test and evaluate. All verticals & markets are different, and so are techniques. – James Luty
  • I like the new feature cause they give you the option. B/f Google use to just assume you wanted to come up for the geo terms. – Bonnie Schwartz (@bonnieschwartz)

Q3: Do you find added value within adCenter campaigns when targeting by gender and age?

  • No. But I do it anyways in the hopes that someday I will. – John Lee
  • With prior campaigns I’ve found Bing to target higher AOV’s when using the gender/age target feature. – Brian Gaspar
  • Not very much due to current traffic levels. Should be great someday! – James Svoboda
    • For some clients that have a very specific audience yes. – Crystal Anderson
      • Audience types that you’ve seen good results? – James Svoboda
        • Yes..saw a positive impact on overall metrics once bid up our target audience and it was for targeting a Female audience. – Crystal Anderson
  • Nope. No added value whatsoever. IMO they might as well turn it right off. – Joseph McConellogue
  • I haven’t had much success with gender or age target yet. – PurePPCCom
  • Volume is so low as it is, that I haven’t dared restrict by gender/age while ROI is so high. – Claire Remmetter
  • @adCenter What value do you hope to add with this feature? – Matt Umbro
    • adCenter offers geo at various levels including radius targeting around biz location. add’l functionality in the wrks. To advertiser w/ age/gender skew to target customer, we offer this ability as a way to increase placement when we have the data. Boosting is the best way to use our age/gender. this allows for prominent placement for target, while not giving up unknowns. – Paul Corkery (@c0rk)
  • Gender/age not as applicable to our verticals but noticed big CTR boosts by using biz info in local geo-targeting in. – Nicole Mintiens
  • I think it’s just a ruse…making me feel like I have more control over my ad placement than I really do. – Tally Keller

Q4: If your client DOESN’T have a mobile site, how does device targeting change (if at all)?

  • I still break out campaigns by device for ad purposes and tracking purposes and budget. – Mark Kennedy
  • If no mobile site, primarily target PCs and tablets. – John Lee
  • We build mobile sites for our clients — no point in #mobile ppc without optimized site IMO. – Claire Remmetter
  • Working with larger brands, I tend to direct more searchers to store locator tools. – Joseph McConellogue +
  • Check initial conversion rates for mobile ads (analytics), if much lower than desktop & tablets, deactivate. Often the case. – James Svoboda
  • We have seen great success with Call Only for mobile. – Bonnie Schwartz
    • Depends on what your KPI’s are. From a lead gen perspective I’ve seen great results. – Brian Gaspar
  • Even if client doesn’t have mobile site I target to all devices for the click to call action. – Matt Umbro +
  • Depends if site displays okay on a phone as is. If not, consider testing call-only. Either way, break out into sep campaigns. – Aaron Levy
  • If a call is a conversion, no need for mobile site, use call extensions and drive to calls! Have seen HUGE success w/this! – Crystal Anderson +
    • I have been severely disappointed with Google’s call metrics/click to call features (and so have clients). – Matt Umbro
      • Be sure to get true cost-per-call, as they can be less than CP online conversion, even if calls drop off. – Joseph McConellogue
    • I’ve found hispanic markets respond particularly well to click-to-call. – Claire Remmetter
      • Google has data that shows hispanics convert better on mobile in general. – John Lee
  • Always have separate campaign (even for discovery KWs & volumes) and use click-to-call if no usable landing page. I use data to allow business to make decision whether to invest in Mobile Landing Page. – Andrew Baker +
  • Have still tested mobile w/o mobile site, but usually test will fail. – John Lee
  • Would still create separate mobile campaigns as found some with lower cpc and better cpa for different device with ad extension. – Anna George
    • Probably good to segment out with separate search and content mobile campaigns as well. – Brian Gaspar
      • Yes sorry meant device was another element for segmentation of campaigns. – Anna George
  • A bad mobile experience can be REALLY BAD. I would stay away from mobile targeting if you don’t have a mobile site. – Tally Keller
  • Separate campaigns for each device, enough volume is required IMO. – Bart Schuijt
  • I sometimes select tablet as some landing pages work fine, this will grow very quickly so worth it’s own campaign. – Andrew Baker
  • Evaluate your LP.. if it renders nicely on mobile and convs. actions can be taken, worth a test! – Crystal Anderson
  • Yes, smaller campaigns for certain but overall cpa has been better. – Anna George
  • Using device/network segmentation on new acct. Desktop/tablet/mobile Search/Content breakdown. Revealing results & optimization. – Lisa Sanner (@LisaSanner)
  • Don’t forget #PPCchat Google can get confused if tablets/mobile share an OS. Ipod touch = mobile. My droid tablet = mobile. – Aaron Levy
  • Q4a: Do you use same conversion tracking code on mobile site or designate it as “mobile tracking code”?
    • Never used separate codes, I guess I should try that. – Mark Kennedy
    • Me neither, there is mobile analytics code, would make sense to have mobile conversion codes. – Matt Umbro
    • Hate that duplicate calls & abandoned calls get tracked as conversions w/Google tracking, tho. – Theresa Zook

Q5: When dayparting & adjusting bids by times, do you find more value in spending more efficiently or improving conversion rates?

  • Spending more efficiently, if budget weren’t an issue I would run ads 24/7 (unless it absolutely didn’t make sense for client). – Matt Umbro
  • Yes, there are certain hours that create bogus leads for us…waste of money! – John Austin (@J__Austin)
    • Good point in lead gen campaigns, different in ecomm. – Matt Umbro
  • Normally to spend more efficiently, clients tight on budget! – Anna George
  • Conserve $. Like to serve off-peak hour “research” traffic but not willing to pay as much as for a click w/sales potential. – Theresa Zook
  • I think it’s a little bit of both. I only want to spend money during times when $ will convert for my clients. – PurePPCCom
  • I think it’s both, usually trying to spend more efficiently, does mean cutting out low converting times. – Crystal Anderson
  • Not just conserve money, but boost bids during high-converting times. Get more clicks when users are more likely to buy. – Aaron Levy
  • Instead of eliminting times all together, I often increase/decrease bids based on cpa/conversion rate by time. – Bonnie Schwartz
  • Really a case by case scenario which hinges on budget and identifiable high conversion times. – James Svoboda
  • Depends on cost per conversion. I focus on capturing the most traffic during the hours/days that drive the lowest CPConversion. – Joseph McConellogue
  • It really does depends goals/client. Eg… some conversions require staffed office phones (daytime). – Andrew Baker
  • I thought spending more wisely WAS improving conversions. – Eloi Casali
  • Feel like you can’t really do one without the other. More like spending effectively in improving conversion rates. – Brian Gaspar
  • If you are available 24/7 then it is about efficiency, if not then it is about making sure spend only when can convert. – Bryant Garvin
  • A client who generated most business via phone calls, Mon-Fri, will often get day parting set from 7 am to 7 pm. – James Svoboda

Q6: Describe a situation where you would choose to deliver your ads at an accelerated pace compared to a standard delivery?

  • When you’re having a hard time hitting budgets, don’t have a budget, etc. PurePPCCom
  • Anytime I’m driving convs under goal and am not hitting budget. – Crystal Anderson
  • Clients that have no budget restraints and just want as much traffic as possible (yes, it happens). – Pamela Lund (@Pamela_Lund)
  • Large enough budget and low traffic volume. Get the clicks & conversions while you can! – James Svoboda
  • When we had unlimited budget. – Aaron Levy
  • Also when you are bidding on terms with less search volume. – Matt Umbro
  • Accelerating ad display is a great test when your IS is limited by budget. Or your budget is unlimited. – Cassie Allinger (@_CassieLee_)
  • Topical events…maximise on a given situation, you just nail that traffic while it’s hot! Smash & grab. – Andrew Baker
  • When I’m killing cost per conversion goals and not meeting daily spend targets. – Joseph McConellogue +
  • Also, if its more of a branding/awareness effort vs. strictly CPA based. – Crystal Anderson
  • Cyber Monday. Where you really don’t know what budget you’ll need & can add budget during the day if you need to. – Lisa Sanner
  • When there’s budget left and the clients’ goal is to spend x euros / day. Also, when ROI is good. – Bart Schuijt
  • I find ACE really clunky because we use destination URLs, though…not nimble enough. – Peter Hughes (@hughespjh)

Q7: Have you run experiments in AdWords (ACEs) and if so what value(s) do you find with this feature?

  • I’ve only messed around with experiments a few times and they weren’t successful. I do want to look into it more though. – PurePPCCom
  • Love love love ACE. Great for bid tests (i”why is our avg. position so low?”) and LP tests. Helps take time bias out of tests. – Aaron Levy
    • What metrics do you use to judge tests? – Claire Remmetter
      • 100% depends on the test.. conversions, conversion rate, CTR, CPA, etc. – Crystal Anderson
  • Do we as PPC specialists under utilize ACE? – Matt Umbro
  • I think “we” do not like to “experiment” and prefer to test and analyze. – James Svoboda
    • Setting up own/more personalized experiments? – John Austin
  • I have used it for ad copy tests as well. Love to be able to see stat sign automatically. – Bonnie Schwartz
  • I think so. I’ve talked to a few that use it regularly but mostly it’s pretty new. I know I need to do more with it. – PurePPCCom
  • I use ACE for bid experiments all the time. – Pamela Lund
  • I find it hard to track what you’re testing exactly with ACE. I’d like to make annotations. – Bart Schuijt
  • I definitely under use ACE, I have a clever way to remind myself to use it. –

Q8: What additional settings would you like to see added to AdWords & adCenter? Why?

  • Though technically an extension, I want to see day parting for the call extension (and call metrics). – Matt Umbro
  • Block specific sites from network in adWords. Specifically search partner sites being able to block them! I would use network much more! – Bryant Garvin
  • I’d like to see reporting functionality integrated directly into AdWords Editor! – Lerna Terpanjian (@Lerna)
    • adCenter Desktop includes a Dashboard which can visualize performance data. May grant you some reporting capacity in-tool. – Paul Corkery
  • John Lee & Joe Kerschbaum hinted towards an new AdWords match Type that I’ve been wanting for a while! – James Svoboda
  • I would love if you could do ACE tests on campaign settings – ex. before day parting/after day parting-That would make me happy. – Bonnie Schwartz
  • Ad-group level extensions, easier sitelink reporting adCenter Extensions. Consistent functionality. Scheduled ad copy, AdWords & adCenter, I wanna launch my “one day sale” ads without staying up till 3am. – Aaron Levy
  • Start and end times (not just dates) for campaigns or ad copy. Going to basting turkey & launching campaigns at the same time. – Lisa Sanner
  • Savable custom date ranges. Would make looking back at results much easier. – Dennis Petretti
  • They need to let you separate search partners from Search. One place adcenter has Google beat. – Bonnie Schwartz
  • Wow where to start… much more / easier integration & reporting of product feeds. – Andrew Baker
  • Day Parting for Text Ads … with diff call-to-actions & destination URLs Please! – James Svoboda


Google AdWords Editor (version as of today’s PPCChat 9.5.1)

Bing Ads Editor


GoMo: Mobilize Your Site

Seer Interactive AdWords Campaign Experiment

More PPCChats

Don’t forget to stay tuned for the next #PPCchat on Tuesday at 12 noon Eastern, 9 am Pacific and 5pm in the UK. Same Chat time, same Chat channel.

Also, check out the next #PPCSpeak on Tuesday Nights at 9pm Eastern, 6pm Pacific.


Check out the PPCChat Twitter list to see and connect with all current and prior participants.

About the Author

This is a guest post by Paul Kragthorpe, Search Manager at WebRanking in Minneapolis, MN, #PPCChat Streamcap gatherer, rare SEO Blogger, SEO Padawan, Tweeterer @PaulKragthorpe, and Google+’er PaulKragthorpe.

Be Sociable, Share!


Leave a Reply

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

Recent Tweets

I am speaking at SMX East
PPCChat.co was rated one of the Best PPC Blogs by Boost CTR