Setting Up & Optimising a PPC Campaign - Day 2

Cool, so let's get right down to business today. Quite a few of you wrote to me to say that the landing page and ad optimisation techniques would come in handy, so I'm quite excited to be bringing you this today.

I'll start with techniques of how to optimise your ads for your PPC campaigns. In the second case study we covered techniques and rules for writing your PPC ads. I'll have to presume that you have already mastered / used those and are comfortable that they work well. In this part, I want to give you methods you will be using to make your ads better after they've been running for a few days/weeks.

Firstly, the way I usually operate is that in the first week after setting my campaign, I will not change anything (as tempting as it may be sometimes), but I make a point of monitoring it very closely to be able to get a good understanding of its behaviour.

Your campaigns will vary in cost and click/impression volume and that will require you to treat them slightly differently as you move into the optimisation phase, but in the first week, I believe it is important to leave it undisturbed. Let it gather momentum.

Side Note: All of the above will apply correctly if you've followed the proper way of research and campaign setup. But please don't leave the campaign unattended!!! It can be very costly. Like, I am pretty sure that if I had to setup a campaign for the "Burn the Fat feed the Muscle" book, we will go through a lot of money very fast - as this is a very competitive, high search volume niche.

So what I'm really saying to you is to use your judgment - unfortunately there isn't a "one size fits all" solution here… If you're not spending beyond your means, and the click volume is good, you should leave the campaign to get some data under its belt. Or at least let it gather 50 to 100 clicks before you change stuff.

Handy Tip: If you had conversion data from any other source for the same product then you could use that as a guideline in terms of how many clicks should give you a sale. So for e.g. if you've had the pre-sell page to the product you've just setup a PPC campaign for, on your site. You'll be able to tell that for every 50 unique visits to that page, you make 1 sale. Now you can use that data to estimate that you should wait for at least 50 clicks to decide what you will change in your PPC campaign.

Technically, your PPC campaign traffic should convert a lot better than organic traffic, but that may not always be the case - only a test will prove the truth :o)

Right - so let's say you've ran your campaign for a week, and you have a set of stats for how it's performed. Get your report out of AdWords into Excel and let's look at how to about analysing and optimising your campaign.

This is a screen shot of a campaign I'm running, and will use for illustration purposes:

campaign-stats.jpg

I've given you basic keyword-wise stats in there - you will get these for any campaign although sometimes your keyword-wise stats will end at the clicks and CPC levels, because as an affiliate sometimes you may not have sufficient control over how you track stuff.

So, there are (in my opinion and experience) 2 stages of influence you can look at and 3 variables within those two stages. How well these perform will define what part of your campaign you need to optimise. I find this method puts some structure into what can other wise be a very haphazard process.

The 2 stages:

  • Your CTR (click through rate) = clicks / impressions in %-age
  • Your CR (conversion rate) = conversions / clicks in %-age

The 3 variables:

  • Your keyword match type
  • Your ad copy (including the display URL)
  • Your landing page

So - if when you come to optimising your campaign you see that your CTR is too low, you'll need to assess two factors:

  1. Are the keywords in that ad group tightly enough targeted to warrant a close match to the ads (I sincerely hope you have at least 2 ads running for each of your adgroups) of your AdGroup?

    arrow If this is not true, than you can play around with either changing the match type from broad to exact or phrase match - that should lower the amount of impressions you're getting and hence increase the chances of a higher CTR.

    arrow You could also add negative keywords to your campaign. This is very often overlooked as a method of optimisation, but it is one of the quickest way to give a failing campaign a new lease of life. If you're selling a product that has a name that also has a lot of searches for a similar free version, adding the negative keyword "-free" would make a huge difference to otherwise irrelevant impressions to your ad. Think of any other words that if searched for will not mean your product is needed. Then add them on to your campaign as negative kwds (keywords).

    arrow Lastly, you could take the phrases you think are not working well with that ad, and put them into their own adgroup, where you can write them a separate better targeted ad that will deliver better CTR and hopefully conversion…

  2. Are your ads tightly enough targeted to the keywords in that group?
    I.e. are you keeping to the rules we spoke about before - like having the search term in the heading of the ad? Then have a look at both your ads and see which one has a better CTR. Keep that one and slightly amend the other one. Make sure you only change one thing at a time, though, otherwise you will not know what it is that made the difference.

    There are far too many books and tips on what parts of your ad you can optimise and how - so I won't go and repeat the whole thing, but good places to start from are:

    • the headline - it's most visible…
    • the display URL - similar reason…
    • include more benefits…
    • give more specific pricing details - but only if that pre-qualifies your prospects in a beneficial for you way…
    • you can also give specific numbers in terms of results generated by the product you're advertising - people like numbers.

Righty-o, that's all I have for optimising your ad for better results.

Tomorrow we move on to discuss the rest of the influence factors - the stuff you can fix by changing your landing page and the stuff after the click.

Side Note: Please let me know if you have any questions at all on any of the above - I can elaborate before we move on.

Also - SHARE - if you have techniques I haven't mentioned, but are particularly successful for you, share them with us all :o)

Great! See you tomorrow.
Have a lovely evening…

Anita

<< Previous Step: [Case Study 5 - Day 1->Setting Up & Optimising a PPC Campaign - Day 1]

Next Step: >> [Case Study 5 - Day 3->Setting Up & Optimising a PPC Campaign - Day 3]




Share This:
  • Digg
  • Technorati
  • del.icio.us
  • Facebook
  • StumbleUpon
  • LinkedIn
  • Sphinn

{ 0 comments… add one now }