Understanding The Google Display Network

1. While a lot of people know about Google AdWords, not many are familiar with the Google Display Network (GDN). Here is an overview of the GDN.

The GDN is a collection of over 2 million websites worldwide, from massive sites like Oprah.com right down to tiny blogs and forums that most of us have never heard of.

Each website has decided they want to monetize their site. By allowing Google to put ads on certain pages of their site, the website gets a cut of the revenue those ads generate.

You have the option – from within your existing AdWords account – to show ads on any of those pages!

Most importantly, you can show ads all over the web with the same precise level of control that you have with your normal search campaigns. This means you can choose to only show ads to people within 50 miles of London, UK on a Tuesday between 3&7pm (if you want to) even though they’re looking at an Australian blog – if you think visitors to that blog are your ideal target market.

It’s important to note that it’s a totally different type of person you’re getting in front of. With the Google Search Network, you know exactly what that person is looking for, you serve them the most relevant & precise ad and take that visitor to your website.

With the GDN, the people aren’t actively searching for your products and services.  They’re browsing a page on the web and you’re essentially interrupting them and hoping that they find your offer interesting enough to click on it.

So in order to improve your chances of success, the first thing you’re going to do is target webpages that have some relation to your product or services.

2. What types of campaigns can you set up on GDN?

There are a number of targeting options available – again all through your existing AdWords campaign. Some of the most popular are:

Managed Placements – You simply give Google a list of websites that you want to show ads on. If those sites are in the GDN & your bid is high enough, your ads will appear there within hours.

For example, perhaps you’re interested in showing ads on Oprah.com or nytimes.com, or you want to target a section of a site like theage.com.au/business.

Be aware that there is no published list of the 2 million sites and that they’re being added to all the time. You can get some info on the top sites from Google’s AdPlanner tool, but for the most part you need to know that a site is in the GDN before targeting it.

As such, many people start with other types of GDN campaigns, where Google does the heavy lifting.

Contextually targeted – Instead of a list of sites, you give Google a list of themes. Themes are about the content of a particular page on a site. So these campaigns are about asking Google to find pages that match your list of themes & getting to show ads on those pages.

For example, you sell wine, so you might choose cheese as a theme. This is a great example of something that doesn’t work with Search, because a wine ad isn’t relevant to someone searching for cheese. But on the GDN this is perfect. Someone is browsing a site about cheese, you can show your wine ad & only pay if and when someone clicks.

In this way, GDN gives you the ability to market on tangents that you just can’t do on Search.

Topic Targeting – Every single page of the 2 million sites has been categorized by Google and organized into around 1,750 categories.

This includes big general topics like ‘business news’ down to specific topics like ‘cookware’ with many pages categorized into multiple topics.

You can give Google a list of Topics & get it to find all the pages matching that topic and place ads there.

In short – contextual targeting is more work and gives you more precise control, whereas topic targeting is easier and quicker to set up but needs more management to ensure great results.

3. A lot of people (even AdWords pros) struggle to get the Display Network to work for them. What are the 3 biggest mistakes you see people making when managing Display Network campaigns?

#1 Not setting up a specific GDN campaign.

Typically in the past, when you signed up for a new AdWords campaign, Google showed ads on both Search and Display Networks  – so many didn’t realize they were actually opted in for the GDN!

While this wouldn’t harm your search campaign, not having a specific GDN campaign gives you no opportunity to optimize it. Given it needs a completely different set up and management approach, that’s not good!

This is less of a problem now as Google is (finally) changing the way you set up campaigns. But many people still don’t understand the differences.

#2 Limiting reach by not having enough AdGroups

This is particularly relevant to Contextual Targeting, which most people use, and results from not giving Google enough themes to work with.

For instance, many advertisers will set up their GDN campaign with just one AdGroup – effectively telling Google there is just one theme. For best results you need to give Google multiple themes, so it has lots of ways of finding the types of pages most likely to work.

Limiting your potential reach by not having enough themes means you’re not showing your ads to enough people and limits your ability to test and find success.

#3 Not managing your GDN campaigns

A GDN campaign is definitely not set and forget.

Regardless of the approach you choose for a campaign, there is going to be a lot of junk traffic along the way which you constantly need to be getting rid of.

Due to this, a big, big mistake is not managing your campaigns on an ongoing base to get the best performance and ROI.

The combination of these three mistakes is definitely why many believe that GDN isn’t profitable.

To begin with, they didn’t set it up properly, then didn’t use it to its full potential and finally, they didn’t put in the time to manage it properly so they’re not going to get the best results.

4. What types of businesses/campaigns is the Display Network particularly well-suited for?

Because there are so many different variations on the approaches to setting up a GDN campaign, essentially any business could be suited to the GDN – it’s definitely worth testing.

Image ads (banner ads) work particularly well when you’ve got a very visual product, or you’ve got a great offer that will grab people’s attention. Particularly if it mentions the word free! Be it a free DVD, a free chapter of your book, a free webinar etc.

In terms of product vs service – the general rule of thumb is that the GDN works better for services than e-commerce however, done well, it can be very profitable for e-commerce too.

5. Are there certain businesses/campaigns that the Display Network probably isn’t a good fit for (and, if so, what are they)?

GDN doesn’t work for businesses with really, really small budgets because in order to manage it, you need data, and to get data, you need clicks.

In addition to clicks, you also need to be able to track conversions so it’s also not good for businesses that don’t have any way to track conversions on their website.

There needs to be some way of measuring what success is on your website – be it a simple form prospects complete or a product added to your shopping cart – because otherwise you have no way to find out what’s working & what’s not.

6. Conversions obviously have a lot to do with the landing page, but can using image ads have a big impact on conversion rates? If so, can you share an example or two of how changing an ad has boosted conversion rate?

The biggest impact that your image will have on your conversion (other than the offer that it contains), relates to the consistency between the imagery in your display ad(s) and your landing page.

The image essentially pre-sells the person, so if they click on it and then come through to a page with the same or similar imagery, they feel that little bit more at ease.

If this is the case, they’re more likely to stick around for that extra few seconds to orientate themselves and hopefully do what you want them to do – ie. buy a product, fill out a form etc.

If there’s a jarring difference between the image in the ad and the image on the landing page, alarm bells are going to start going off for the user, which will lead to high bounce rates and wasting a lot of money.

The other big benefit of using image ads is the ability to test multiple images to see which attract the most people to your site.

This is rarely done – most businesses just run 1 ad (maybe 2) that the CEO or graphic designer has chosen. But that’s typically NOT the ad that performs the best.

Take this to the next step… Once you’ve tested 10 ads, now take your 2 winners and create 2 new landing pages that specifically match these images to see which converts better.

It’s a great way to truly test the market – it’s market research, not opinion research.

Because you’re testing your prospects, but they don’t know they’re being tested!

7. Would you recommend businesses have different landing pages set up for Display Network campaigns vs Search Network campaign in AdWords? (If so, can you share an example of how a new landing page for a Display Network campaign boosted campaign performance?)

What you need to remember about the GDN, is that you’ve distracted this person, which is why for the most part these Ads contain more hype than a search ad.

Your landing page needs to be consistent with your ad, so there is certainly an argument for tailoring the landing page for display vs search.

Going back to the wine & cheese example – if you’re selling wine via search, then you’re most likely going to have your imagery and text on the landing page revolving around wine.

However if you’re targeting cheese lovers on the GDN, you may find it works better to include wine and cheese on your landing page to tie the theme back in, and perhaps use a different headline.

There might be subtle differences, however for the most part I don’t recommend clients initially set up separate landing pages if budget and resources are limited (which is 99% of the time!).

But of course – always continue to test!

8. When you’re running a Display Network campaign, what are the top 3 metrics you’re looking at to gauge how well the campaign is performing and what are you looking for in each?

#1 Click through rate (CTR)

I never judge the overall success of the campaign based on CTR as I don’t know what a good CTR should be from one industry to the next.

However, where CTR rate is important is when you’re testing a raft of different ads, as it gives you information on what ads appeal the most to your customer – what ads do they want to click on.

This is only relevant if your ads have big enough differences in them. It’s not ‘testing’ if you have 10 ‘different’ ads but they’ve all got the same hero image, headline and there’s only a few pixels difference!

CTR% = total clicks / total impressions

#2 Cost per action (CPA)

This can also be known as ‘cost per lead’ or ‘cost per sale’ and it represents ‘how much each lead costs’

You can measure this at many levels: campaign, AdGroup, ad creative & these days even the keyword that caused the ad to show. It all depends on how granular you want to manage it, but keep in mind that the more that you manage it, the more profitable your campaign will become.

CPA = Total cost / Number of conversions

#3 Profit per impression (PPI)

This metric doesn’t exist anywhere in your Google AdWords – you have to pull your information out and work it out in a spreadsheet.

However it’s an important metric to know, because it’s the only true way you can truly judge whether one ad is better than the other as it gives you the complete picture.

The following example will show how using CTR or CPA alone can lead you to draw an incorrect conclusion about which ad has given you the best bang for your buck.

Which of these ads do you think is better?

#1 – Looking at either Clicks or CTR rate alone, the obvious answer is A.

#2 – Looking at additional CPA (or Cost/Conv) data, I imagine you’ve now swung towards B as it has the lowest Cost/Conv.

#3 – To get the final profit per impression (PPI) figure, you do need to know your breakeven point (BE) for your conversion.

Your break even amount is the point at which sales no longer become profitable.

So if you sell a $100 item that physically costs you $80 to get from your supplier, your breakeven is $20, assuming no other costs.

If your cost per conversion is higher than $20 you won’t make a profit running that ad.

If however you sell an ebook for $27 and your actual cost to provide that is essentially zero, then set your breakeven at $27 (as long as you make sales for less than $27 you’re making a profit).

Now you know your breakeven (BE) you can work out Profit Per Impression (PPI) using the formula below:

PPI = break even – cost per conversion / total impressions (in thousands)

As you can see, depending on your BE, the question of which ad is better can look remarkably different to what it did using CTR or CPA alone.

Here Ad C is the most profitable, but just how profitable changes depends on the break even point.

For example, with a break-even point of $2,  advertisement A’s arithmetic would look like this:

(2 – 6) / 80 = $-.05

but advertisement C would look more favorable with the following:

(2 – .56) / 20 = $.07

9. What resource(s) do you recommend for someone wanting to learn how to set up and manage a Display Network campaign?

If you want to learn it yourself, then I can recommend my own PPC Masterclass which is available at websavvy.com.au and gives you all the information you’ll need to get started with the GDN & the other AdWords features.

However, there are pitfalls to doing it yourself.

As I’ve already mentioned, an incorrect set-up is one of the biggest mistakes people make with GDN, which leads to poor performance and worse, wasting your precious marketing dollars and paying what’s been called ‘AdWords stupidity tax!’.

So if you are keen on managing it yourself, at the very least I’d recommend outsourcing the set-up to a professional agency such as WebSavvy.

Once you’re set up, keep in mind that it’s not just set and forget.

You’re going to need to invest a considerable amount of time into training so that you know the AdWords system well enough to make the best possible use out of it, or else you’ll fall victim to another one of the biggest mistakes – not managing your GDN campaign properly.

And again, this will lead to poor performance and wasted money.

I’m a strong believer in using your internal resources in the wisest way possible, and due to the time and effort it takes to become proficient and remain up to speed with the ever changing AdWords landscape, I think your best bet is to use a professional agency for the set-up and ongoing management.

How do you pick a good agency? Look for a professional who has expertise in AdWords, a proven track record and if possible is recommended by others you know.

Any agency that tells you that GDN is too hard and/or doesn’t work … run away from them and fast!

WebSavvy would obviously be delighted to help you with your next GDN campaign, and I can also recommend Rocket Clicks or Cardinal Path.

10. Do you have any favorite software/tools (not including Crazy Egg) that you use to help optimize Display Network campaigns?

In addition to your AdWords account and analytics tools such as Crazy Egg and Google Analytics, we use 3 main tools:

• AdWords Editor – a freely available application you can get from Google that not enough people know about and will save you a lot of time.
• Excel –  yes, the humble spreadsheet! Pivot tables in particular can be scary to some, but if you’re running AdWords properly, you need the information this will provide.
• WebSavvy’s very own beta testing tool – as this wasn’t available, we had to make one our self! So if you are interested in finding out more, feel free to contact us.

3 Big Takeaways

This was a longer interview with lots of great information packed into it. Here are three key points to takeaway from it…

1. You can use the GDN to reach prospects who you may not reach using the Search Network and are totally different type of person you reach via Search (which requires a very different strategy).

2. Don’t just run 1-2 ads in your campaign. Run 10 and then take the top 2 performing ads and create new landing pages highly targeted to each ad.

3. Metrics like CTR and CPA are helpful, but can be deceiving. Take the time to calculate PPI to get the complete picture.