• Justin Sous

    Great article, Brad. Thanks also for clearly defining how you were measuring account size. I think there are a lot of different assumptions and initial reactions when someone mentions account size, so you clearly defining this set yourself up for a well-done post!

    Just one comment on small to medium sized accounts for Search. You mentioned the following: “If you are doing a lot of geographic targeting, you might end up with a few additional campaigns because of your geographic targeting, but beyond location targeting you usually don’t have many campaigns.”

    This may have been completely true a few years ago before Google has developed more ad extensions, but especially with site links, if your account has a variety of products or services, to fully leverage ad extensions you should have as many campaigns as you see fit. For example, a business with 10 essential products may want to be able to leverage site links for each one. This can boost CTR significantly. Rolling up your sleeves and doing the necessary work allows you to win big in the SME ppc landscape.

    Thanks again for the post!

  • http://twitter.com/bgtheory Brad Geddes

    Justin – while I agree with the optimization plan you laid out; it really does depend on account size. If you are spending $1000/month and you create 50 campaigns, now you’re trying to manage $20 in spend for each campaign which might take more work than the increased CTR gets you.

    For mid-sized and larger accounts – I totally agree with you. You just have to watch budget management as well. If Google were to just implement an account budget (ideally a search and display budget) that would make multi-campaign management and segmentation much easier.

  • Justin Sous

    I understand what you’re saying, Brad, and sometimes this changes how I look at spend vs budget with a particular client. Especially with service based businesses, some days are busier than others, so recommending a daily budget is derived from the question “what are you comfortable spending right now”, and then set the budget accordingly. I totally agree that for small/medium sized clients, $1000 monthly spend could feel like a lot of money to them. 

    The bright side of breaking down the campaigns in this fashion, is that you can pinpoint specifically where you’d like to add budget to. Going back to my example, if a company offers 10 different products/services, and they wanted to push a certain one, this campaign set up would allow me to do just that. This is just from my experience from working with smaller/medium sized accounts in the way you define them. I know there are ton of other variables as well. It’s always pretty interesting to hear how other PPC’ers set up their accounts and their thought processes, so thanks again!

  • http://twitter.com/dlukauskas David Lukauskas

    Great article. 

    Question:

    A third party is optimizing your small campaign and can only
    see the conversion cost, which is a sale online and due to technical
    limitations has to put all keywords in to one ad group and track the cost per
    conversion at the keyword level. The campaign has 1000 keywords, avg. conversion
    rate is 2%. So if we want to optimize the bids on this campaign we would have
    to get at least 1,000,000 clicks before the data is somewhat statistically
    valid (20 actions per keyword x 2% conversion rate x 1000 keywords).

    This niche may have 10-30,000 clicks available per month in
    Adwords …

    Would you say that somebody who tries to sell this type of
    PPC management for a small account like this is full of BS?

     

  • http://twitter.com/bgtheory Brad Geddes

    David – yes!

    Putting 1000 keywords in one ad group is almost always a terrible idea. There are a few exceptions; but that has to do with the long tail and the time vs possibility debate.

    However, for anyone who wants to manage accounts of any size (small or large);  they should be separating out keywords by ads and landing pages and creating multiple ad groups.

    If that’s there limitation – it sounds like a very bad system.

  • KFreiberg

    Excellent Article!  The trouble that I find with managing small accounts is that customers who have such a low ad spend, are rarely happy with the results they get, because there really aren’t any results.  

    Unfortunately, there’s such a huge gap in the market of people who are willing to manage small customers (and manage them WELL!), and I hate turning small clients away, but it’s hard to get “results” for these customers AND keep them happy.  They come saying “oh, I want 50 clicks a day, with 5 conversions and I only want to spend $50 day” –> and we’ve had people in the plumbing business, which from my experience seem to be anywhere from $7-$10/click and it just doesn’t make the budget go so far.

    Of course, a huge part of our job is setting expectations, and I will talk until I’m blue in the face, letting our customers have a good idea of what the reality is, but it almost always leads to an unhappy relationship  for both parties.  I’d love to see an article on how you suggest managing smaller clients, because they’re out there and they want help, but how do we fill that gap…  :)

  • http://www.facebook.com/profile.php?id=507536131 Michael Shostack

    It is definitely challenging with small Clients whose budgets prevent you from managing an account “properly.” And by that I mean breaking campaigns out to apply the correct settings, testing new areas, etc.  For small accounts with lots of locations that need different geo targeting settings, you can easily run into issues where campaigns may only be able to deliver 1-5 clicks per day because of how thin your budgets are spread.

    At this point, it is best for them to focus on really small and profitable areas, which may take a while to determine. And typically that means testing on AdCenter as well since CPCs are lower most of the time.

    The unfortunate reality is that Google has priced themselves out of the market for many small businesses in competitive categories–which it seems they are quite fine with.

  • http://twitter.com/bgtheory Brad Geddes

    Really small clients are very difficult to manage unless you’re setup to manage a large number of them – such as some of the resellers because you really need phone call tracking, leads tracking, fast setup, auto-optimization, etc. It can be done; but its usually a challenge that many people handling large accounts shouldn’t take on as its so different. 

    I find that one of the best things to do is add call tracking and then have all leads go directly into a CRM. That way you can say, “well, we got you 10 calls; but according to the CRM it took you 3 days to call any of them back. For every hour you wait to call someone back, your chance of making them a customer halves.” 

    It’s bit more complex than that – but I think that gives some of the overall thoughts. I’ll think about writing an article on the subject. Thanks for the suggestion. 

  • http://twitter.com/bloomarty Martin Röttgerding

    Terrific article, Brad! Yes, I’d like to read more about these things – not necessarily about big vs. small budgets, but what strategies work for which type of business in general. I think only few people can relate to the whole spectrum of PPC and articles like this one provide unique insights.

  • http://www.fuzzone.com/ Kunle Campbell

    Fantastic article Brad! What is your advice on the approach to negative keywords for both small and large accounts? Should they both have a similar approach? Small accounts need to be pruned in good time before the budget gets out of hand with no ROI but how about larger accounts? As you would be dealing a much larger dataset.

  • http://twitter.com/bgtheory Brad Geddes

    For small to mid sized accounts, I generally rely on a combination of stats and ‘gut feeling’ to add negatives. 

    For larger accounts, I generally use both adwords data and GA data. I find that by using GA data, I can see interaction goals and that if a word has no conversions but 4 minutes on site – I might keep it for a bit where if it has 100% bounce rates and no conversions then it becomes a negative.

    So, I guess in the end its patience. For small ones, since they can’t advertise on every word to begin with, its easier to add more negatives to find what does work. For larger ones, a negative means less traffic since they can buy most of what’s out there; so you’re more likely to try and make it work (assuming its not something that’s just junk and needs to be a negative). 

    For large ones, I’m also more likely to really silo ad groups/keywords (like some of these http://searchengineland.com/3-simple-strategies-for-organizing-your-match-types-70096) ; where for small ones unless its simple, I might not take the time to find all the little bleed-over areas to fix them. If it becomes a problem, then I will; but I might let some nit-picky stuff go assuming the overall account is doing well. 

  • http://www.fuzzone.com/ Kunle Campbell -Fuzz One Media

    Hi again Brad – thanks for your reply on negatives. 

    I have another question regarding device targeting. With the exponential growth of mobile traffic are you segmenting campaigns for specific devices i.e. creating campaigns dedicated to only desktops & tablets and then other campaigns only for mobile devices?

    I find it easier managing device targeted campaigns on smaller accounts but management seems to grow in complexity in larger accounts (especially if location targeted campaigns already exist in the account).

    Is this also the case with you?

  • http://www.jordonmeyer.com Jordon Meyer

    Great article, Brad. Working on all of the above types of accounts, I agree with your approach. From working with Massive and Huge accounts over the past few years, I have actually seen up to 7 Search accounts under the MCC, in addition to display and mobile. This is a typical practice of large companies. It works well, and scales perfectly with the model you laid out above.

    Foundation is everything. If you’re building a sky scraper, you better have the entire architecture figured out before you start building. You can only add so many glass boxes on the Sears (Willis) tower. It’s much easier to add a wing to a small house. But foundation is key. 

  • http://www.facebook.com/todd.bairstow.16 Todd Bairstow

    One of the major challenges we have with managing small accounts has to do with managing Quality Score. We run thousands of geo-targeted campaigns for our clients, but the click volumes are low on an individual basis. This makes it hard to move the Quality Score needle, as it takes tremendous patience and hands-on management to see an appreciable difference. Very frustrating.  We have been flummoxed by indicators like  Expected ClickThrough and Ad Relevance, because it varies so widely from geo-target to geo-target.  

    Great article Brad, keep up the good work! 

  • http://twitter.com/bgtheory Brad Geddes

    For large accounts, I segment by device and sometimes by carrier. This is why I find for large ones, it can be useful to create a mobile account from the MCC so you can easily see the mobile data vs desktop data. In some cases, I’ll also segment out tablets.

    For small ones it depends on the industry. For some accounts, I just won’t do mobile at all if the desktop is better and they can spend all the budget on tablets/desktops well. For others, I will use a mobile campaign with a mobile page; and often call extensions or click to call ads. The industry dictates a lot about the mobile strategy for small accounts. Some small accounts are only mobile with no desktop campaigns as well. 

  • http://twitter.com/bgtheory Brad Geddes

    Totally agree. That’s why with large accounts, its possible for me to spend a few days just mapping out accounts and campaigns without ever adding a keyword. The structure can determine so much about how you manage a campaign that if you don’t do this well you often mess up performance before you actually start advertising. 

  • http://twitter.com/bgtheory Brad Geddes

    Totally agree, and sometimes I find it best to just ignore QS for small accounts. If the results are there, then sometimes QS is a variable that doesn’t matter as much as it does with larger accounts with lots of data. In fact, I recently wrote an article on when to ignore Quality Scores here: 
    http://certifiedknowledge.org/blog/learn-when-to-ignore-low-quality-scores/

  • http://twitter.com/adriasaracino Adria Saracino

    Glad to see this came to fruition!

  • George Bounacos

    I *love* a mobile only account idea.  Thanks for that tip.

  • Erika McCarthy

    I would also like to see more on this topic. I run both large and small PPC accounts. For us, it has really been an issue of scale for larger clients. I was used to looking all my whole search query report twice per week for my small clients. And now with a large client, that is just not doable. I think the biggest thing to remember when working with ALL PPC clients is to have measurable goals in place. If both wan’t to increase online sales by 5% in Q4, then you have your marching orders.

  • http://devedge-internet-marketing.com/ Kevin Clark

    Thanks for the article. This was helpful as I’ve been collecting a few smaller accounts lately and had to rethink the setup. Not too mention finding a fair management fee for those sub $500 folks.

  • http://profile.yahoo.com/5QL3NDOXHL27YJPU75HLXGSCUA Tb

    This is an amazing article!  One question, when you say:

    For instance, it is a best practice to bid at the keyword level, and every large account should be doing that on most of their keywords.However, that is an unrealistic expectation for a small account that does not generate enough keyword level data to set keyword bids. Small accounts must bid at the ad group level based upon combined stats of all the keywords within an ad group.

    Is that simply because there’s not enough data to draw conclusions?  I manage ALL small, locally-targeted campaigns and I find it maddening the lack of consistency from month to month.

  • http://twitter.com/bgtheory Brad Geddes

    I find that if you track everything and manage at least $500/month in spend; that the results can be semi-consistent for most local companies. I do find small spend ecommerce can be a bit all over the place in terms of consistency at low budgets.

    If its a well built out local account (say 20+ ad groups) then at $500 where I usually see the inconsistencies is what ad group gets the conversions. While you might have 40 conversions/month – those conversions do often jump around in terms of what ads/keywords/ad groups are being credited with the conversions.