On the heels of launching Yahoo campaigns in three countries in eight days on an incredibly tight timeline, it makes sense to take a small step back and jot down a few thoughts about what went right, what went wrong, and how to improve next time.
Going global with SEM
I think this phrase is misleading. It’s not like you can take an SEM (or any kind of) campaign and globalize it. Going global really means going local in a dozen (or more) different markets. Sure, there are some elements of the campaign that are universal, but the real work lies in duplicating the franchise across a large number of diverse markets without deviating from the core message.
To go in-house vs. outsource?
Outsource. Don’t be a hero. Don’t even think about running this in-house. Unless you are an SEM agency promoting your own brand, definitely look for outside help. Even then, you might want to contract with an agency that specializes in this kind of work, as the sheer volume of details can bog you down. Your existing SEM agency may have the skills and experience to pull this off, but I wouldn’t take it for granted. This is not your father’s SEM. If you are considering using your existing agency, make sure they have case studies ready to show you. If they don’t, then don’t force it. Move on. There are a number of agencies that have the chops to pull this off. Find them, interview them and pick one that you like, because you’ll be spending lots of time with these people, and it gets pretty intense.
You need to look at your budgets in two different ways to make sense of them. First, take a top-down approach. Find out what your total media budgets are for the campaign and look at how they are distributed by market. You’ll probably have a big chunk of the media being spent on the US, and smaller percentages in EU and emerging markets. Take another metric that’s representative of SEM budgets in relation to other online (or offline) media. These days, you can use 40-50% as a placeholder for SEM as a proportion of total online media.
Now you have some budgets by market. Next, you want to take a bottom-up approach. Once you’ve worked with your agency on basic keyword lists, find avails for the list in each of your target markets. Then run three scenarios based on share of voice (share of search) in each of those markets. Now compare your bottom-up estimates to your top-down approach and see how they line up. Hopefully they intersect somewhere. If not, no big deal, you will either bump up against the share of voice totals (if so, congratulations!) or will be limited by overall budgets and will have to trim your share of voice to fit the limited budget.
If you find yourself in the latter group, you’ll be managing trade-offs, so here is a tip. Break your reduced budgets down into components like brand, product, etc. so you can spell these trade-offs out clearly and explicitly to management.
Building it out
So here’s how brand campaigns normally work for SEM. You’ve got a brand message. In our case the brand message is “It’s Y!ou.” So we have gobs of media out there—broadcast, radio, out of home, online display, you name it. Like all good campaigns, we have microsites and landing pages in all major markets that speak to this message, that invite users in to interact with our brand and guide them through the brand experience and the products that support the brand promise. So how does SEM play into this? Simple.
First, write as many ads as your brand message (and associated web assets) will support. Take your brand keywords (and misspellings!), attach the new brand ads to them, and point them to your microsites, landing pages, or whatever web assets you have to support your brand message. Put your brand keywords in their own campaign, so you can manage the budgets carefully (this is especially important for big brands, as high search volume can eat through a ton of budget). Now, build another keyword list of all the terms associated with the message itself. There won’t be too much search volume there, but you’ll want coverage on these terms to ensure if anyone (inside your company or out) searched on these terms. Attach the same ads to these keywords. If your have product offerings that support the brand promise, you’ll need to tailor some ads to these products. Make one campaign for each product, again, so you can fine tune your budgets.
Asking permission vs. forgiveness
At some point you are going to need to get approvals for your campaign builds. Depending on your organization, this could include brand and product folks in a number of countries. That’s a lot of back-and-forth. So a couple notes here. First, have your agency build approval templates that are easy to read. Second, review all the campaigns first and weed out the obvious problems. Now comes the tricky part. Depending on how highly-compressed your campaign is (think leadtime), you may or may not be able to get everyone to approve your builds before they go live. Don’t panic. This is where you can once again be thankful you are a search marketer. Go ahead and launch if you have to. Once you’re live, set up meetings with all the stakeholders and bring them up to speed. Get their feedback, and impress them with the speed with which you can make the necessary changes to your ads and keywords.
That’s all for now. Tune in next month when I’ll write about launching multiple campaigns in multiple markets, as well as measuring success.
Opinions expressed in the article are those of the guest author and not necessarily Search Engine Land.