I can’t help but feel that almost every PPC advertiser manages their paid search based on instinct more than logic. This is problematic for two reasons:
1. Instinct can be highly derived from a “survival mode” or “stop the bleeding” mentality, and
2. A natural instinct of humans is to take the reasons from what most would describe as common sense, but sometimes defy logic completely in their course to solve a problem.
“So what’s the problem?” asks the advertiser. “If I see my account wasting money on a keyword or ad group, I pause it.”
Yes, you will stop the bleeding, but you’ll also stop all of the sales, leads, and phone calls that go along with it. You actually restrict other parts of the campaign even though they drive good volume, some of which is profitable.
“Well if it’s a keyword, I’ll just bid it down. I won’t pause it.” he insists.
The same things happen. You will reduce your risk for that keyword, but you’ll also reduce the volume of clicks, which directly reduces the volume of conversions or calls you get. You haven’t really solved the problem. You’ve only reduced your risk in the problem.
And then there’s the industry-old software plague. I’m sure you’re familiar with some of the big players in PPC software. These software solutions primarily do two things: highlight and automate the inefficiencies in CPC bids and cost per conversions. To be fair, they do a great job of it. But this cannot be the full scope of the management because it completely bypasses the “human” management element. There are things you have to take into account, such as who the target market is, what the product or service actually is, and what people search for when the target market is looking for that product or service.
So the advertisers that use this software or SaaS (software as a service) solutions end up with PPC managers that are thinking more about at the software and less about your target market, products, and services. The problem-solving capability becomes limited to what problems the software can solve alone and not what the informed advertiser can solve.
WAYS TO LOWER CPA (CONVERSION COSTS)
Think about fixing a high CPA keyword, ad group or campaign in two general ways. You can either modify things to decrease the exposure for those areas OR you can do things that decrease the CPA without decreasing the exposure. In the end, you might end up doing both, but you want to fully exhaust all of the non-exposure-decreasing solutions first. Then, if the CPA is still too high, begin with changes which lower the exposure the LEAST.
Take a look at the first set of charts. Notice that the “Cut Cost” (Fig. 2) philosophy shows revenue decreasing when costs are being cut. Now look at the second set of “non-exposure-killing” chart (Fig. 3). The costs are much higher but so are the clicks, revenues, and presumably the profits.
Here are a few basic examples of different management activities that can lower your CPA.
Lower CPAs without decreasing exposure on good impressions by:
- adding negative keywords that stop bad or irrelevant queries and query themes
- improving ad copy, CTR, and quality scores = naturally lowering CPC
- improving conversion rate with onsite optimization
Lower CPAs while decreasing exposure on good impressions by:
- reducing bids or making negative bid % adjustments
- reducing budget on high CPA campaigns
- restricting keyword match types to stricter settings
- reducing scheduled hours of ad serving
- reducing geographical territory for ad serving
Eliminate CPAs while stopping exposure on good impressions by:
- pausing or deleting keywords
- pausing or deleting campaigns
The activities in the above lists aren’t by any means earth shattering. They’ve been around since the beginning of PPC and we’ve all done them. So it’s not a matter of “if” we do them or not but more so “when” and “in what order” that really matters. By no means am I saying we should never pause keywords; I pause keywords all the time. But I only do it if I no longer need that keyword to bring in those same sales (i.e. I have another keyword that covers those same queries) or if I have exhausted all other possibilities in lowering the CPA first.