Marketers love targeting. In search advertising, there are several different ways to reach specific audience segments with tailored messaging - one of them is geo-targeting. A geo-targeted campaign allows search marketers show ads to users only from a specific region as small or big as the advertiser wishes. This approach is most useful for businesses that cater to a limited geographic area. There are other advantages as well including:
1.Ad copy can be customized to incorporate
location-specific promotions or user behavior;
2.Keyword/ad group bids tailored to the region’s
ROI;
3.SEM spend adjusted based on availability of
local inventory.
As such, the average search advertiser loves the idea of geo-targeting
as much of the keyword inventory as possible. However, this tactic requires
careful analysis as there are trade-offs involved in the process.
ROI Trade-off
Advertisers could end up paying a CPC premium on
geo-targeted keywords. I have a few examples below that corroborate this in a
big way.
In the New York area, Keyword 1 paid 89% higher CPC and got
21% fewer clicks in the geo campaign when compared to the national campaign.
Data across other regions for both keywords shows a similar trend.
|
ACCOUNT
XYZ |
|
|
NATIONAL
TARGETING ONLY |
NATIONAL
+ GEO |
|
ROI:
4.15 |
ROI:
4.31 |
|
ACCOUNT
ABC |
|
|
NATIONAL
TARGETING ONLY |
NATIONAL
+ GEO |
|
ROI:
1.80 |
ROI:
1.64 |
Campaign Manageability
Geo targeting quickly multiplies the number of keywords in
the account. Creating a campaign for every U.S state alone grows the account a 50-fold;
going down to smaller regions multiplies the size even further. Campaign
management in this scenario can be painful, especially for those SEM marketers that
have poorly scalable bidding systems. Additionally, the incremental gains
realized (discussed in more detail below) are not always worth the extra human effort.
Another critical dimension to consider is geo search volume. At a few clicks per day, a low traffic keyword behaves almost like a speculative bet, with volatility being commonplace. In the chart below, revenue/click (RPC) has been plotted for a group of regions under campaign X—within a period of 2 weeks, average RPC goes down by 50% with other factors staying constant, suggesting greater volatility when the campaign is split further.
More importantly, low volume campaigns don’t move the needle much on
performance. For instance, in Account BCD, newly-created geo campaigns received
a mere 0.75% of additional spend
under performance-based bidding—the campaigns just did not have the incremental
volume to be able to spend more.
In summary, geo-targeting needs to be evaluated on a case by case basis as it
does not always lead to ROI gains. There are trade-offs involved, and the smart
marketer will consider both the upside and downside before making
implementation decisions.
Sameera Inapakutika
Business Analyst
