With the latest 100% not provided news, I have been asked more times in the last few weeks for PPC training than in the last year! (Albeit jokingly from many SEO’s)
So I thought that now is as good a time as any to put together a PPC series to help refresh your memory if you’ve been out of the loop or for those that are short of time and patience, cover the basic things you should be looking at in order to make better use of that budget and time.
Of course you can just set up your account and let it run, but do you feel lucky punk? Do you? When was the last time you checked your client account manually?
My top 10 PPC Improvements are as follows:
As with SEO, before you can start work on optimising a PPC campaign or setting one up, you need an audit. Jumping in and fiddling with a few settings for a quick fix can work but given that there have been many changes to the PPC interface in the last 6 months that require a different approach (hello enhanced campaigns!), it is going to be better long term to start from the ground up and if there are a lot of changes to make, a potential re-build of the account in a more efficient way is the way forward.
It is easy to miss things or just let them run as ‘if it aint broke don’t fix it’, but as with SEO, you should be constantly monitoring positions, bounce rates, keywords and looking for potential golden gaps. Even if it’s doing well already and you can just set up and leave it – why give any more money to Google than you have to?
An in-depth SEO audit for keywords, marketplace, links can take a few days to a few weeks, so why wouldn’t you spend at least few days looking at your PPC campaign? The last time I used a free software audit – it gave me some pointers but didn’t really help me to understand the problem (this may have changed), so if you are looking for a more comprehensive manual audit – a fresh pair of eyes or even an outsourced service can pay for itself very quickly with additional revenue or decreased spend as a result.
The first thing I look for is how an account is structured. Often campaigns are left in their original form then built up from there. Sometimes because users don’t want to lose history but sometimes because at the time it is the quickest and easiest way to make changes and sometimes just because it’s been left running and not been looked at for months or even years!
I have always favoured a hierarchical structure that mirrors your website purely because it makes the most sense. An account that closely matches your campaign also has the following benefits:
Google have a handy image on their help files to explain:
When was the last time you really looked hard at your Analytics? ‘If’ you have your account linked correctly this helps! (I’m not joking – 50% of the accounts I look at aren’t linked effectively).
Analytics & tracking is a huge area but for the purposes of this article, I am assuming that you have Adwords conversion code in place, e-commerce tracking set up to show you sales values and that you have some goals in place.
If not – WHY NOT?
I spend most of my time in Analytics these days looking at the Matched Search Query Report – not only does it give me an insight into which keywords are performing – it shows me pretty quickly which keywords or areas are wasting cash.
Below are just some of the matched search queries for the term ‘romantic break’. If you don’t already have them set up, there are plenty of spin off Ad Group ideas from a handful of keywords and plenty of keywords that could be easily excluded.
Go through the matched search query report and add everything that is not relevant to a negative keyword campaign. Some words you just don’t think of until they come through as a matched keyword string. For example, I recently added the negative word ‘donkey’ and ‘donkey sanctuary’ to a holiday group for a well known beach area – who knew? – we do now and it has made us check areas more thoroughly before adding keywords to groups.
If you are being clever and exact matching out keywords all over the place to drive traffic to the actual exact match groups you will be adding individual negatives at Ad Group level. If you have keywords that are relevant to the whole campaign – use the shared library and negative keyword lists to exclude unwanted keywords at Campaign level.
You know the ones, over and over again Buy Stuff Now, It’s cheap & Quick for every single product or too much automation (looking at YOU Amazon & Ebay).
Ads like this do work for branding and many products if no one else is advertising there. Also, Ebay and Amazon probably don’t need to be as careful with their pennies! [disclaimer alert].
However, if you’ve had the same Ads sitting there for a decade with no testing, no modification, and no additions – how do you know for sure that another Ad won’t work better?
I’ve had clients tell me that they know their customers and these Ads work the best with no testing. They may be right, but it’s always worth a test.
Poor Ads damage your CTR, Relevant Ads increase your CTR, which in turn increase your Quality Score which in turn lower your CPC – what’s not to like about that?
I know it’s nice to appear for everything your heart desires, but do you really need to be there? Before switching off low performing keywords or keywords that have had no conversions in 60 – 90 days, consider switching down, then monitoring your overall revenue.
Unless you have state of the art super whizzy tracking software, you may inadvertently switch off some brand traffic returns or lose sales that may be made on a different device (search at work, buy at home).
Switching down should be your first call, then switch off once you are pretty sure that the keyword was not converting elsewhere.
If you have had keywords sat there for months and months, it could be that they have a much better overall quality score than you think. Which means that you could still appear in the same place for a little less.
Don’t start cutting of £’s or $’s randomly, shave a few cents or a few pence off either the groups or individual keywords (depending on how your account is set up). It could save you enough to ramp up some of your higher performing keywords without affecting your pocket.
Try to avoid the dreaded below first page bid score notification. If in doubt, watch closely and take of a few cents/pence every other day.
Endless keywords in the same group is almost as soul destroying as 500 Ad Groups, each with just one keyword in. If you are testing this can be a quick way to find out which keywords win but once you have researched your marketplace, you will know the kinds of keywords and Ad Groups you want to try.
There is debate with how many keywords you should put into a group but I would say no less than 5 and no more than 15/20. There will be exceptions where every single keyword in a large keyword group is being targeted effectively, is totally relevant, has a brilliant quality score and a click through rate to die for… but I have yet to find them (please share if you have found perfection in large keyword groups). With all the additional match types available and proper planning, you should find that less keywords are required to get the same effect.
Using less keywords/Ad Groups on a lower budget also means that the system is not spreading itself too thin to try and target all 10,000 keywords that you are asking for on $5 a day
Broad Match, Modified Broad Match, Exact Match, Phrase Match, Negative Embedded Match. As with DJ Names and types of dance music, it wouldn’t’ surprise me if in the next few years we start to see creatively named Broad House Bling Modified Britney Match with a Twist, but for now, we have 4 (well 5 if you include embedded negative options but it’s not really a type, more of a technique).
Many consultants suggest that you separate all of your different match types in different groups and I would agree in many instances – BUT if you are manually creating and managing your account, one group for each type on EVERY SINGLE KEYWORD GROUP is just unnecessary – especially for groups that you’ve made that get absolutely no impressions (see above).
Questions you should ask yourself:
Quality score is the name of the game but if you have no impressions or clicks because you have been too complicated or too specific, you are not going to see any benefit either. I have also seen some exact and phrase matches cost more than broad so as with everything – test it.
Be creative but also be smart. Don’t just upload a host of formulaic spreadsheets for the sake of it. Use separate match groups for high traffic terms, where you can spend the time adjusting individual keyword bids and put the time in to specific relevant targeting as otherwise you have a load of groups with similar keywords potentially ‘fighting’ against each other.
Broad match with plenty of negative keywords can be just as effective in niche areas without giving you a reporting headache. If you really want to get specific with your matches – check out this post from SEJ
If you’ve made a major change to your account you also need to let it settle for a little while so you can get some data, gain some history and work on that Quality Score.
I don’t necessarily mean set it up and let it run as you should be monitoring, but the more you fiddle in the early stages because ‘things aren’t working’, the less you are ‘teaching’ the system about how you want to run things and the less you are giving it a chance to tell you what is and isn’t working.
Within reason is the phrase that springs to mind. Obviously you don’t want to be spending $20 a click for 3 weeks to prove it doesn’t work so decide on how much ‘test’ budget you can afford before you just let the campaign run away with your money.
Sure you can get results in a day or a week/month but thinking long term means you can monitor and tweak effectively and really get to grips with what the data is telling you.
I have turned income around in a matter of weeks, but conversely there have been some smaller spending accounts that take months and months to develop because there is no test budget, money is tight and improvements are thought of on a daily basis i.e. that didn’t work yesterday so let’s switch it off today…
I tell clients that realistically it takes 3-6 months to implement changes and refine things and even after 6 months there are still further refinements and areas to target and definitely more wins and cost savings if you just give the changes a chance to kick in and take some time to nurture what is already working.
Everyone has a different way of setting up accounts so I am sure there will be disagreement on some of the above. There are hundreds of other things you can do to increase your reach, increase your conversion and lower your costs, but not all of them are quick fixes.
Things like conversion improvement and landing page creation take time, effort and a lot of testing, so I have kept this post as a guide to the top things I look for when optimising accounts and the things I have discovered whilst working on different sized accounts.
If you have any tips or things you have discovered while working on accounts big or small – feel free to feedback, comment and share.