Why PPC Granularity Will Be Your Best Friend

January 7, 2016

This is a guest post from Johnathan Dane, Founder of KlientBoost.

Have you ever thought your Google advertising account should be performing better?

You may be following the advice of many that say that the more time you spend in your account, the better.

But what if it’s all backwards?

What if it only takes you 10 minutes a week to improve your Google advertising performance?

If your Google campaign performance hasn’t been improving month over month like the table below, then keep reading.

01-GoogleCampaignImprovement



It’s about to get interesting. Let’s get started.

Automatic Placement Extraction


If you’re running any type of display or remarketing campaign, you might find that your display ads are showing up on websites, apps, or even video overlays that aren’t performing well.

Overall though, you might be decently happy with your display performance, but always wondered if it could do better.

[caption id="attachment_7076" align="alignnone" width="500"]

02-DisplayPerformance

Take a look at your Automatic placements under your Dimensions tab[/caption]

To start the “performance pruning”, see which Automatic placements either have a cost per conversion that’s too high, or better yet, which placements are actually bringing in sales (not just conversions) by equipping your Google advertising Final URLs with ValueTrack parameters.

This will then help you get more conversion volume out of those specific placements when you extract and target them exclusively through a new campaign.

Search Term Extraction


Search term reports are such an important part of regular Google advertising maintenance that it’s not uncommon that some people do this more frequently than brushing their teeth.

When looking at your search term report, get as close as possible to making sure your search terms and keywords have no discrepancies between them.

In other words, your Added / Excluded column from your search term report should have the green “Added” label going down the list for as long as possible, just like this:

[caption id="attachment_7077" align="alignnone" width="500"]

03-SearchTermExtraction

This gives you a much stronger control of what you’re paying for[/caption]

When that happens, you can make your ads specific to not your keywords, but your search terms and see higher click-through-rates from your efforts.

Let’s say you look at your search term report and find your search terms and keywords don’t match. The first thing you should do is extract your search terms with the most impressions and create what are called Single Keyword Ad Groups (SKAGs).

Just like the name implies, SKAGs are ad groups that only allow one keyword per ad group, that then have corresponding ads that are extremely specific to that keyword.

Time Lag and Attribution Reports


Did you know that the last keyword and/or ad clicked always gets to lionshare of conversion credit?

What if there were seven other touchpoints (impression and ad clicks) that happened before the final conversion? Wouldn’t you want to know what helped assist that conversion?

I know I would.

If you don’t care, there’s a good chance you’ll pause keywords and placements that don’t get the conversion credit. But, when you do, you’re strangling your account at the same time, without even knowing it.

Let’s take a look at your Google advertising attribution.

Inside your account, go to the top of your Google advertising interface and click Tools > Attribution.

Once you’re there, take a look at the Time Lag report on the left side. Here, you can see how long it takes people to convert from either first impression, first click, or last click.

[caption id="attachment_7078" align="alignnone" width="500"]

04-ImpressionConversionDelay

Here’s a look at first impression conversion delay of 6.19 days[/caption]

This will help you make your nurture and/or retargeting campaigns more of a priority to test.

Geographic Granularity


Are you a local, statewide, nationwide, or even an international advertiser?

No matter how big an area you’re targeting, every geographic hill, slope, mountain, and valley performs differently. The same thing goes for individual states and cities.

And, because you can’t target people who live on just a hill (yet), the next best thing is to understand the performance of each state or city that sees your ads.

[caption id="attachment_7079" align="alignnone" width="500"]

05-GeographicGranularity

To see this report, go to your Dimensions tab, then View -> User locations[/caption]

As you can see above, the state of New York may be costing more per conversion than others. So, you may want to add in negative bid modifiers at the state level, like this screenshot shows.

06-States



You can then drill even deeper and create new campaigns with state level campaign targeting, and give bid modifiers to individual cities within that specific state to get your closer to your cost per conversion goals.

You can take it even further and start utilizing city specific ad copy and landing pages with area code specific phone numbers, to appear more local to visitors and increase your conversion rates.

Device Targeting


As I’m sure you’re already aware of, Google advertising doesn’t allow you to separate devices in their own campaigns like they used to.

These days, you have to group desktop and tablets together in the same campaign. And while Google may say that both those devices perform similarly, there are thousands of Google advertising accounts out there that say something completely different.

Here’s the truth: Desktops and tablets will never perform the same way.

I’m not just speaking from a conversion rate standpoint, but also from a sales standpoint.

When Google told the world that devices don’t matter, but user context does, they certainly never thought of every single industry, but more so of a blanket band-aid that would apply to “most advertisers”.

Believe it or not, there are some workarounds you can use to get desktop, tablet, and mobile campaigns in their own campaigns and still target the search and/or display network.

But first, let’s look at how we find current device performance differences within your account.

[caption id="attachment_7082" align="alignnone" width="500"]

07-PerformanceDifferences

Here’s how you find that info[/caption]

First, go to Segment then Device in the dropdown.

[caption id="attachment_7083" align="alignnone" width="500"]

08-SegmentDevice

This will expand your view with three extra rows[/caption]

As you can see in the screenshot above, our mobile devices are giving us the lowest cost per conversion while tablets are sucking it up and being the most expensive.

Now let’s say for a minute that your tablet performance is just as good as your desktop performance (like Google says it is), but your mobile performance sucks.

You can quickly add in what’s called a negative bid modifier between 1 and 100%.

[caption id="attachment_7084" align="alignnone" width="500"]

09-NegativeBidModifier

Go to Settings -> Devices and increase or decrease in the red square[/caption]

If you never want to target mobile devices, then you can set a negative bid modifier of 100%.

Day of Week Targeting


Just like keywords, ads, and landing pages perform differently, so does Monday compared to Thursday, and Saturday compared to Wednesday.

Inside your Google advertising account, you can see this day of the week granularity in a snap. Just head over to Dimensions ->View: Day of the week.

[caption id="attachment_7085" align="alignnone" width="500"]

10-DayOfWeek

In this case, Saturdays and Sundays are doing really well[/caption]

Having these kinds of numbers doesn’t mean that you should stop advertising on Thursdays (because it has the highest cost per conversions). But, it could mean that you should start considering “day of the week” bid modifiers like we did for our devices earlier.

Some industries tend to be very predictable in their weekly trends. If your company falls into a category like that, then take advantage of the control you have and get more aggressive with your bids on great performing days, and taper back on the not so great-performing ones.

Time of Day Targeting


Just like we saw how your days perform differently during the week, so do your hours within the day.

11-TimeOfDay



And, just as we can create bid modifiers for 24-hour day targeting, we can also take advantage of the same thing with bidding blocks of hours within a certain day of the week, to break it down even further.

[caption id="attachment_7088" align="alignnone" width="498"]

12-BiddingBlocks

In this screenshot, late mornings and afternoons tend to perform better than mornings[/caption]

If you already have the data and insight that allow you to use this type of granular bidding, then definitely do so.

You might even find that Google or other bidding platforms are restricting how many bid modifications you can make on a daily basis. If that’s the case, I suggest you try using Brainlab’s 24 hour bidding script that allows you to take it one step further, and then some.

In Closing


Now before I let you go, please keep this in mind:

“With great control, comes great responsibility.”

Having access to all of this data is great, but only if you can be actionable with it to improve your performance.

I see time and time again that people spend countless hours trying to tweak and prune things with modifiers, rules, and even scripts that change bids depending on the weather.

While all of this is great, most of it becomes entirely obsolete as soon as you have a landing page test that improves your conversion rates by 50%. When that happens, all the things you’ve put into place need to be redone.

One thing that will always help you out, no matter your goals, is to extract and target things in a granular fashion that makes sense.

Use the dimensions tab and its reports to your advantage and keep on making progress :)

Johnathan Dane

Klient Boost
By submitting this form, I am agreeing to Marin’s privacy policy.

See why brands have relied on Marin to manage over $48 billion in spend