5 Tips for Scaling Paid Search

So, you’re an agency that just landed a huge multi-location brand.  Or maybe, you manage PPC for an established brand experiencing huge growth across several geographies. In either case, congratulations!

But how are you going to manage and optimize paid search and display for hundreds or thousands of locations without pulling in lame results? Hire more people? Probably. Build complicated spreadsheets for importing/exporting into your existing platform? Maybe if you’re desperate.

The problem is, neither of those strategies accomplish the one thing you need right now and in the future: SCALE.

We’ve compiled the most important tips for scaling up your paid search efforts across locations and brands, because you’ve got better things to do than worry about spreadsheet versioning and new hires.

Data Housekeeping

Before taking on paid search efforts for a multi-location brand, you must get your data in order.  What data?  All of your locations/agents, their physical addresses, phone numbers, hours of operation, etc. These are all pieces of data that you need, for each location, to build successful campaigns. Bonus points if you can gather data such as product lines carried, performance, category, amenities, and promotional periods.  This type of data, when compiled in a single file,  will help you crank up the relevance of your ads and build quality scores that reduce cost.

Automation

Now that you have this data, one option is to use concatenation to build a series of campaign and ad group spreadsheets to upload into your bid management platform.  This will take quite a bit of time, but it will be totally worth it for about…24 hours.  What happens when a data point changes?  Or when a location opens or closes? At this point a human must open the correct spreadsheet(s), change the data point without messing up anything else, and upload it.  Maintaining these data points, along with launching and maintaining campaigns, takes all of your time away from optimizing.If this sounds like what you’re doing, you have already gone past the tipping point of needing automation.

When you can no longer optimize EACH location as if it was your ONLY location, you’ve gone too long without an automation strategy.

An automation platform (such as Balihoo) will grab the freshest data each day, trigger and deactivate campaigns based off those data points, and ensure the ads always reflect the latest offerings and locations.  This frees up your team to optimize like crazy.

Account Setup – Subaccounts

Before you set up anything in AdWords or an automation platform, you must be prepared for each location/territory/dealer to be isolated in their own account or sub-account.  There are so many reasons why this will help your campaign optimization, including:

  • Sub-accounts allow your ads to run using the correct timezone. Unless your locations are open 24/7, this can dramatically improve results.
  • Sub-accounts allow each location to have it’s own monthly budget that paces independently. Without this, a handful of locations typically eat all of the budget quickly.
  • Sub-accounts allow you to easily report and diagnose problems on the location level.

Segmentation

Using new columns in your master data file, break your locations into groups, or “segments”. Try to think of as many elements as you can that could affect the campaigns you run.  Do you have a group of premier dealers that need additional marketing spend?  Do you have a group of kiosk locations with limited offerings? Or underperforming locations that are on your radar?  Whatever it is, using this data along with automation will allow you to quickly customize and launch hyper-relevant campaigns specifically for these groups.

Optimization

You can successfully optimize 1000 locations in 1 hour.  Don’t skip this – you’ll be throwing money at your competitors!

Interested in Balihoo Paid Search automation?  Let us give you a demo!
What tips do you have for scaling paid search and display?  What’s working for your team?  Have your KPIs seen a difference?

Comments