Reducing CPCs (Cost per Click) for PPC Campaigns

Imagine you have a set monthly budget for your PPC campaigns e.g. Google AdWords or Microsoft AdCenter and everything seems to be running smoothly- how can you get more bang for your buck?

General trend for PPC Campaigns
With lower bids your ads will appear less and in lower positions thus the campaigns will spend less (and vice versa).

Bidding strategy to maximise clicks*

• If you have a set budget which you reach daily, lower the CPC bids 
• Reduce the bids cautiously and continuously over 2-3 weeks, maybe 4-6 times
• Wait at least 3-4 days until you change the bids in the same Ad Group
• Set yourself reminders in your calendar
• Monitor spend and av. ad positions
• While reducing the bids continuously you will notice that after a certain time the daily budget will not be spent anymore
• Find the “sweet spot” (see the star below) where your bid is just enough to spend your daily budget
• In the example below it is €0.25 which is just enough to spend the daily budget
• You may have to raise your bids again to hit your daily budget

PPC Bid Strategy

*Btw, for ecommerce etc this may not be applicable as there may be different bidding strategies

How to check if Google Analytics Event tracking is working

Imagine you have a button on your website and you want to know how many people click on this button. It may be an important conversion button. You may want to track it so you can see which of your marketing campaigns are more successful. For PPC Campaigns you want to see which keywords are driving more conversions.
So you set up Event tracking in Google Analytics, change the website code and now you want to test if everything works smoothly.
There is different ways to check if Google Event tracking works, the below is my favourite; quick and easy and reliable. There is no need to wait if you can see Event Tracking data in Analytics by the way, you can test it rightaway.

1. First download and install Firebug: https://getfirebug.com/

2. Open the page you want to test, then open Firebug. You should select the ''Net'' tab and ''All'' tab.

3. On your website, click the button you want to test.
You will see a ''GET__utm.gif"" appearing with the Google Analytics domain (see picture).

4. Click the +sign to expand this.
5. Click Params.
6. Now find "utme" where you will see the category, action and label you specified. Check if they are correct. I would recommend to have a spreadsheet where you keep track of all your event tracking codes.

After 24 hours you might see the first data appearing in Google Analytics. In the Analytics menu, go to "Content" and click "Events". You will see the event data with the category, action and label information. Don't forget to use a good naming convention.

Real time Bidding Platform Google AdWords

Google AdWords is a real time bidding platform.
The ad auction is done within a split second right at the moment you hit the search button.
But do the stats you see in the AdWords interface reflect real time stats?

Before I joined my current employer, I was applying to a job as head of search in a very well known betting company in Ireland.
Funny enough, during the interview I learned that the role was only a Search Account manager position, not head of search (the recruitment agent messed things up).
But nonetheless I enjoyed the interview and learned more about this company.
The head of search with whom I had my interview said that he let the team work late until 11-12 o'clock on horse racing nights, where there is increased search activity.

This was very interesting-because the Google AdWords stats you see are not 100% real time. They are at least 3 hours delayed! I know it well from my times in Google but anybody can see it on the bottom of the Google AdWords interface:

What makes it even more difficult is that not everything is delayed exactly 3 hours. Part of the data is more up to date than the rest which can make the performance look really good when it is not, or really bad when in fact it is very positive. This is dangerous.

Imagine part of the costs are updated earlier and the conversions are updated later, you will see a very bad cost per conversion which doesn't reflect the truth. If you try to counter this by reducing the bids, you will ruin your well performing keywords/adgroups.

This is also the reason why you may see a clickthrough rate of 200%. Some of the clicks are updated earlier than the impressions, so that you may have 1 impression and 2 clicks for a keyword which will result in a 200% CTR. In reality there may have been 10 impressions and 3 clicks.

So, the poor Search Engine Marketing team had to work and 'optimize' until late in the night on racing nights, this activity was only good for the Head of Search to impress his boss.. it most probably harmed the AdWords account performance even..

The solution is not to intervene so quickly, give the campaigns enough budget, let them run and check back after at least 24 hours. Remember also that the Google Analytics data may be delayed even more. After big optimizations I recommend to let the optimized campaign run at least 3 days before you intervene again.

Looking up one's own Google AdWords Ads

Imagine you run Google AdWords ads, either yourself or through a Search Engine Marketing Agency and you want to check if your ads appear...

1. Scenario.
This is what happens in 99% of the cases. You want to check if your ads are showing and if they are high in the rankings so you search a keyword you know is in your keyword list. Your ad appears and you are happy...
what happens in the background though is:
Your ad gets impressions but no clicks. This lowers the CTR (Click through rate) artificially, which in turn lowers the Quality Score (QS). A lower Quality Score will result in higher av. CPCs (Cost per Click). In other words you make your own ads more expensive ! And in the worst case Google might stop showing your ad totally as the system might think "a lot of impressions but no clicks = irrelevant ad".

2. Scenario
You search your keyword, see your ad and you click the ad. This is different, you pay for the click but it may increase the Quality Score. If you do it too often it will count as click spam though..

So, the correct way is:
If you want to check if your Google AdWords ads are showing, do not perform a live search, use the Google AdWords Ad Preview Tool.


This Google tool simulates the search, you see exactly what you would see in a live search but it does not affect the Quality Score in any way. Another advantage is that you can simulate being in different countries.
Further recommendations are:
Setup IP filters in the Google Adwords system (Settings) for your office IPs and even e.g. supplier companies' IPs.

Google AdWords Ad Preview Tool

Cost per Sale for AdWords?

Google AdWords is maybe the best trackable media; you can track every single click, conversion and Euro.
You can perfectly link Google AdWords and Google Analytics and get data beyond the click as well.
In certain cases AdWords falls victim to this transparency.
Companies are often too strict and Google AdWords is put under the microscope because it is so trackable. If the cost of conversion is higher than what they expected it seems like "We are losing money with AdWords". And often the conclusion might be: "Oh Google AdWords is too expensive lets reduce the budget or stop it!"

This is not quite the case.
What is the cost of conversion of radio, magazine, TV, cinema, outdoor etc?
"Oh but that's branding, Adwords is not for branding it is only for sales/driving ROI."

What better branding is there than bringing a potential client who is interested in your products to your website and let him experience the full effect of your website branding, with logos, banners, messages etc?

My advice on this is:

• Don't be too strict with the conversion rate/cost with AdWords if you are not with other media
• Use attribution models to see if keyword groups which don't lead to sales drive sales indirectly (last click vs first click e.g.)
• AdWords can be a strong brand awareness driver as it brings your target audience to your website which is highly branded. 
• ROPO (Research Online Purchase Offline) will definitely occur in most industries. AdWords will increase Offline sales but it will be hard to track 
• AdWords has a positive effect on organic searches

CTR goals, the higher the better?

Let's say you are running a PPC campaign on Google AdWords and you are wondering what CTR is good...

Google AdWords Account-Campaign Level Screenshot 

What is a CTR?
The Click-Through-Rate (CTR) shows what percentage of the people who see your ad actually click on it.
The CTR is an average and can be found on different levels, Account level, Campaign level, Adgroup level and Keyword level. It is a good indicator of how relevant your ads are to your keyword.

Why should I optimise my Google AdWords Campaigns for CTR?
The CTR is an important factor in the Google Quality Score. Google rewards relevant ads by giving them a higher Quality Score. A higher Quality Score might mean that you pay less per click for your ads, or alternatively your ads get a higher ranking for the same cost per click.

Search CTR vs Display CTR
First of all one should distinguish between GDN (Google Display Network-formerly called Google Content Network), and Search. In Google a Search CTR of about 2% and above is regarded as good and for the GDN, 0.10% is regarded as normal. Anything less than these values should be investigated further.

Brand CTR vs Generic CTR
It is also important to distinguish between brand terms (e.g. Adidas) and generic terms (e.g. runners). A brand campaign might have typically 20%-30% CTR. Non-Brand campaigns (Generic) should have 2% at least as stated above. And by the way, if you haven't separated brand vs generic keywords into different campaigns then please do. It is not easy to assess and manage a mixed campaign.

CTR and Traffic
When I was working in Google, one inexperienced industry sales manager said: "I don't regard any CTR less than 4% successful!"
Hmm, ok... let's analyse this further.
Through optimisations you can influence the CTR. The CTR is in negative correlation with traffic. Below you see a very simple graph showing this correlation.

The more specific you become, the higher the CTR, but you will sacrifice traffic on the other hand.
Well, through optimisations you could even achieve an average CTR of 15% on generic terms, but traffic would suffer. So it is not advisable to set rigid CTR goals. Striving for a too high CTR is not advisable.