Segmenting Your Paid Search Program For Improved Manageability

Most direct response marketers either need to meet a given efficiency goal (CPA, ROAS, etc.) or get the most of the budget available — or both combined. The budget and efficiency goals typically vary over time for lots of different reasons, such as seasonal trends and promotion periods, and you need to quickly toggle from one strategy to another over the course of the year.

While you can simply activate/pause some campaigns or channels, update the daily budgets, and bid up or down across the board, I would like to suggest other ways to seamlessly scale up or down your online marketing effort by identifying more subtle segments in your program.

1. Identify Actionable Segments In Your Account

From my experience, these segments can be defined as different pieces in your account which can be turned on or off, with no or limited overlap, just like different investments in a portfolio. The point is that you want to make sure these segments are actionable, in the sense that you know how to easily update a few settings and anticipate the impact.

Generally speaking, most campaign dimensions (device type, day of the week, etc.) can become a segment as long as they are actionable. Once you have identified and measured each individual segment in terms of both ad spend and revenue, you can easily toggle different settings on and off to always meet your goals by picking the most profitable scenario out there.

Just to name a few, you could come up with the below segments in your paid search account:

• Branded vs. All Campaigns: Temporarily pause all non-branded campaigns in order to improve efficiency while securing a big chunk of the revenue.
• Top vs. All campaigns: Only keep your top product campaigns when scaling down, or have all campaigns active when scaling up. This only makes sense if your account structure was designed adequately.
• Active Match Types: Pause broad match type across the board (or even phrase + broad) in order to decrease ad spend and improve efficiency.
• Search Network: The search partners tend to account for roughly 20% of the ad spend and 10-15% of the revenue, so you might want to temporarily disable the Google search network.
• Device Targeting: Opt out from mobile impressions, or set conservative/aggressive mobile bid modifiers depending on your mobile strategy and goals.
• Location Targeting: Opt out from some poor-performing locations, or set conservative/aggressive location bid modifiers depending on your budget and goals.
• Day-Parting: Opt out from poor-performing days or times of the week, or bid up at the best-performing times of the week.

Maximum Efficiency vs. Maximum Volume

2. Analyze Segments’ Performance

Now that we have identified multiple segments in the account which can easily be toggled on or off, we can further analyze each segment’s performance in terms of both ad spend and revenue so you can determine the best strategy given your budget and efficiency goals.

More specifically, each segment accounts for a given percentage of the account ad spend and revenue, so you want to calculate those percentages by pulling a couple of reports: by campaign, by match type, by hour and day of the week, by search partners, by device, by location, etc.

For each segment, you want to determine the impact when tweaking different settings, given that 100% indicates maximum ad spend and revenue. In the below example, exact match type accounts for only 65% of the ad spend and 87% of the revenue — so you could pause all phrase and broad keywords to save 35% of ad spend and see a limited 13% dip in revenue in the meantime.

Match Type Breakdown

Another example would be the search network analysis. In this instance, Google.com accounts for 89% of the ad spend and 92% of the revenue. So you could turn off the search network and easily anticipate the impact on ad spend and revenue.

Search Network Breakdown

The same type of analysis goes for each segment, hence the table below with multiple scenarios.

Segment Examples

3. Simulate & Predict

We now have broken down the account in different segments and calculated how much ad spend and revenue they account for as a percentage of the account. We can now simulate and compare multiple combinations using the above table, factor in seasonality and pick the most-profitable scenario accordingly.

Say your ad spend was $25k and the revenue $300k last week, and you are currently running at full speed with all layers set to the most aggressive set-up (100%). You’re expecting a 9% increase in ad cost and revenue due to seasonality, and you also need to keep the ad spend under $15k next week. In that case, you’d probably pick the following set-up (based on the chart above):

• A2: All Campaigns
• B2: Active Match Types = Exact + Phrase
• C2: Search Network = Google + Search Partners
• D2: Balanced Mobile Bid Modifiers
• E2: Balanced Location Bid Modifiers
• F2: Balanced Hourly Bid Modifiers
• G3: All Week
• H3: Branded & Non-Branded Campaigns

As a result, the simulated ad spend in percentage would be 100%*72%*100%*92%*90%*90%*100%*100%= 54%. You add 9% for seasonality and your predicted ad spend now reaches 58% of $25k= $14.6k. The associated revenue would be 86%, hence $257k.

In this case, all you need to do is pause all broad keywords and adjust your mobile/location/hourly bid modifiers, and you should achieve your budget goal while maximizing revenue.

Conclusion

Each paid search account possesses unique characteristics that can potentially be broken down in slightly different segments, or in a more granular way, as long as it is actionable. This post really just opens the conversation about how to leverage historical data in a multi-dimensional analytical environment, and easily pick the right strategy for your online marketing program while making it more actionable and predictable.

Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.

About The Author

Benjamin Vigneron

Benjamin Vigneron is a seasoned digital marketing strategist with experience in Europe and the US and was listed as one of the best eCommerce PPC Experts by PPC Hero in September 2014. He currently works as a Senior Business Analyst in the Digital Marketing team at Adobe, where he provides advertisers and account management teams with data-driven and actionable insights on strategies to optimize their search, programmatic, and social initiatives.