Feb 162009

Budgeting Search? What's the point?

Investment Banks call me just about every week these days, asking whether our clients' search budgets are growing or shrinking given the state of the economy. Their interest is in valuing Google and Yahoo stock, but I'm not sure my responses are terribly helpful to them.

Very few of our clients have set budgets for search. The vast majority of our clients are retailers, and almost all of those view search as a direct marketing channel with well understood ROI expectations. How much they spend is therefore a function of market opportunity, not board room decision.

Granted, many of our clients have been forced to think about top line - bottom line trade-offs, but they're thinking about those trade-offs as direct marketers do: what is the incremental ROI on the next X dollars spent in this channel, and how does that compare to other channels? This is very different than deciding: "We're going to spend Y dollars on search this quarter".

I get the sense that our clients are in the minority on this, but I don't really understand why.

Why would a company keep spending money when the ROI turns south? "By golly, we're going to spend this last $10K even if it doesn't generate a sale!"

Why would a company stop spending money when the ROI is good? "I know that every time I give you $10 you'll hand me back $11, but I can only spend so much..."

Neither of these statements make any sense to us.

Targeting ROI with undefined budgets is a fundamentally different approach, and it informs search practices. Algorithmically, our bid management system was designed to generate the maximum sales/margin or leads without exceeding the client's efficiency target(s). Other systems are designed to get the most sales/margin/leads they can from X dollars in advertising. That's a different algorithm, and we think the wrong one.

The problem with budgets generally is best illuminated by the specific example of campaign budgets. One obvious question is: why do I only want to spend X on this collection of ads? Let's assume there is a reasonable explanation for that, there is still another question: if I'm actually hitting these caps am I managing the campaign properly? The answer to this is pretty clearly "no".

Suppose the campaign can only spend $1,000 in a day. At $1 per click you hit the campaign budget after 1,000 clicks. Wouldn't you rather have 2,000 clicks for $0.50 each? In all likelihood, you'd double your sales on the same spend. Granted, maybe the ads are in position 7 all day, rather than position 3 for a few hours, but who cares? Is there any reason to generate less traffic and sales on the same spend?!? Any time caps are hit, by definition you've lost opportunity. Lower bids to get maximum traffic for the dollars spent.

With that tuning however, you run smack into the first question again: now that the campaign is more cost effective, why shouldn't I spend more on it?

Those who can't shake the budgeting habit often try to pick a budget that will generate the ROI they need, but that's still backwards. Why guess at the market opportunity in advance, when you could instead seize the day with flexibility?

A number of our clients have some component of branding in their program, and budgets there certainly make sense. The more divorced the goals are from profits the more budgets make sense.

However, for those of you looking to make the most of search in these trying times, consider eliminating fixed budgets as a first step.


26 Responses to "Budgeting Search? What's the point?"
Nick Skislak says:
This is great, I am doing SEM for clients, small business mostly, and you hit the nail on the head. I am also having problems acquiring new business due to the fact that many small business owners don't know anything about Search. There is only so much teaching I can do on a 5 minute phone call. Any suggestions, shoot me an email. Good post.
Jim Novo says:
Let's make the puzzle even stranger. The people who use budgets for Search are also generally the ones evaluated on goals related to Sales. The people who do Search without budgets are generally the ones evaluated on goals related to Profit. Is this paradoxical or is it just me? "Your goal is Sales but we are holding you back with a budget because, well, we don't want your Sales to get too big, you know. Don't want to much success, because it might be too expensive." To me, that attitude literally means what the people setting the goals are really thinking about is Profit. They are using the budget for controlling expense to hit Profit goals. Wouldn't it be a lot simpler and more valuable to the company to just set Profit goals in the first place?
Thanks for the comments Nick and Jim. Jim, I agree with you, it comes down to a fundamental misunderstanding of the channel. I've heard the argument made this way: "If I tell you it's okay to spend 25% of revenue on advertising, you will. But what if I could get those same sales for less than 25%? Heck I'd rather have a 5% cost to sales ratio!" Of course the reality is that 1) yes, you should keep an eye on the marginal ROI in addition to the average. If the marginal ROI is so bad that a 20% A/S generates the same top line as 25% then by all means, trim the target; but 2) In most cases you're going to get a much greater sales volume if you can afford to spend more per click. As we tell our clients and prospects: we can hit any efficiency target you want, we just don't know what the sales volume will be. It's kind of like Heisenberg's Uncertainty Principle: if you know the efficiency you can't know the volume; if you fix the volume, you can't know the efficiency.
It's surprising how rampant this is among advertisers and agencies alike. I definitely agree that maximizing sales/leads/etc for a particular budget is a backwards algorithm. The optimal budget is always a moving target. If you substantially overshoot it, then that algorithm will essentially bid you to position number 1 for everything, because you'll get the maximum traffic. The damage isn't as bad if you undershoot it, but the missed opportunities are still huge.
David Camma says:
Great post. This why client scare me who I think are mortgaging their house to do PPC thinking it is going to save their business. We can have a good idea, but we won't know how effective it will be until we run the campaign. I usually do use budgets over the first three months of a campaign while we are building quality score and gaining research.
Johan says:
Great post! The theory is so simple, that it is very hard to get clients to understand the trueness of it :) This reminds me of a similar post by Seth Godin: http://sethgodin.typepad.com/seths_blog/2009/01/do-ads-work.html
Brian Combs says:
Great article, George. The majority of our clients are B2B, but our B2C clients are behaving in exactly the manner you layout.
Thanks for the comments folks. Johan, what's frustrating to me about Seth's post is that he makes the same point I do but only uses 1/4 of the words. How does he do that? David, we often rely on the campaign budgets as guard rails when we launch a new client because we don't know what to expect. However, like guardrails, they're helpful for preventing a disaster, but they're a poor substitute for steering :-)
Mark says:
George, well written & can I please use your quote in the future! "like guardrails, they’re helpful for preventing a disaster, but they’re a poor substitute for steering" That is pure prose! Overall great subject and as more and more are getting online with no clue on how to do PPC campaigns companies RKG are a godsend. I commiserate with Nick. With so many new adopters getting into the game now as PPC has become a difficult terrain to navigate it does take some hand holding. This from your post sums it all up “I know that every time I give you $10 you’ll hand me back $11, but I can only spend so much…” No guard rails needed if the campaign is optimized properly and adjusted as the spend occurs. In my opinion it's about the metrics tracking and adjusting to the changes (steering is what George called it). George my question to you would be what metrics are you tracking and/or tools that you use (of course if you can without giving away your secrets!) to optimize your campaigns?
Chris says:
George, Great post. I've actually been reporting CPA to my executive leadership for the past several quarters now. As long as we remain within our targets, there is no budget. I suspect as CFO's are making the spending decisions, they will ask, why set a cap? We get $7 in profit for every $1 spent on PPC. Why on earth would you hold that back?
rvb says:
The crux of alot of large-ish companies who spent far too much time and money moving around fake $$$ within groups at the same company: once you're allocated a budget for something, you spend it....otherwise, your budget will be decreased the next time budget planning comes along.
Great comments, folks, thanks for sharing! Mark, you're welcomed to quote me any time you like. I might recycle that line myself :-) To your question, we built all our own software for managing search. It's a massive project, and folks just getting into the game today would have an awfully difficult time trying to catch up. Renting a bid management platform might make sense for those folks.
What about businesses that don't get a financial return for their spending? Would you say that the same "budget-free" approach would apply to a business that has a limited budget (overall, not just for marketing, etc.) if they can't expect to see a financial return?
Hi Tanner, The harder it is to "see" the ROI the more budgets make sense. As I mentioned for branding, if you don't know what your advertising does for you, budgets are essential. Indeed, because perfect understanding of what marketing drives what sales is impossible it's imperative that the overall marketing spend stay within some reasonable fraction of overall top line sales. That notion allows for flexible spending in areas that are easily tracked, like search, but also can involve fixed budgets for programs that aren't so easily tracked, or that require a great deal of time to germinate. George
yeah I think the three letters "ROI" should not even be in some of the conversations around search engine marketing. I mean at the end of the day nobody has control over Google and Yahoo so outcome is sometimes unpredictable.
Bill Cook says:
I really enjoyed this article. We often have trouble communicating this concept to clients. It can be difficult getting people to place trust in their conversions rates and keyword bids to focus on the ROI. They always want to set a budget. It seems they are convinced it will be like other forms of advertising - which can be hard to determine the ROI - and they are trying to limit their losses by setting a cap on the budget. To help with this problem, we built this AdWords ROI calculator that can help get them focused on the ROI, not the budget: http://www.bluewatermedia.com/tools/roi-calculator.html (if anyone has feedback on the calculator, I'd love to hear it.)
H town says:
Everything that involves money should be thought twice. Say for example, budget. The company that has a limited also has limited capability.
DM Taylor says:
At this moment in time I prefer using image and banner ads for this reason as there is less competition and you can reach and you get more for your money. The only downside is that the traffic is less targeted. Great post.


Check out what others are saying...
[...] having a search budget might not be necessary Really interesting post from The Rimm-Kaufman Group on why you shouldn’t have a search budget, maybe it’s time to change our thinking in [...]
[...] a direct marketing program (as distinct from a branding campaign) budgeting search doesn’t make sense. Particularly because there isn’t even a cash-flow consideration — sales are generated [...]
[...] search is different in every respect. Budgets don’t make sense for advertisers with shopping carts on their sites because the sales happen before they pay for the [...]
[...] search is different in every respect. Budgets don’t make sense for advertisers with shopping carts on their sites because the sales happen before they pay for the [...]
[...] A téma bennem volt már régen (nem egyszer írtam róla), de most ismét beugrott, sőt egy-két hasonlatot innen kölcsönöztem: searchengineland és a büdzsé dologhoz pedig innen. [...]
[...] value. That way we maintain constant efficiency, and generate the most sales possible. (Folks who budget search instead look to spend the budget at the lowest cost to sales ratio possible. This is a totally [...]
[...] value. That way we maintain constant efficiency, and generate the most sales possible. {Folks who budget search instead look to spend the budget at the lowest Cost to sales ratio possible. This is a totally [...]
[...] Is Competitive Paid Search hitting efficiency targets? If not, go no farther, the program is wasting money and heads should roll. Spending the budgeted amount for media regardless of the ROI is the wrong way to approach search. [...]

Leave A Comment