Articles, Blog

AdWords Location Optimisation: T-Time With Tillison

AdWords Location Optimisation: T-Time With Tillison


Hey guys welcome to another T-Time show, or
rather a cup of coffee today. I wanted to share with you a bit more advanced tips on
Google AdWords, specifically around optimising for locations. Follow for the next 5 to 10
minutes, its all about AdWords Location Optimisation. The first thing that we need to understand
is where to look for the data in your AdWords account. The first place we can look at this
is in the dimensions tab and then picking, you can do days of the the week, this kind
of stuff. What we are interested for today, what we are interested in today is Geographic,
so this will break down all the data for your campaigns and tell you where your traffic
is coming from, how it behaving ok. So lets just add some columns in here. So lets just
do town and city and obviously depending on what data you have got set up in your account,
if you have got the cost per click its obviously in here, the click through rate those sorts
of things. Hopefully you have conversion tracking set up, if your don’t ARE YOU NUTS? Get it
set up you NEED it, just don’t run and AdWords account without conversion tracking. If you
are an eCommerce store, or you have attributed values to conversions if you are trying to
generate leads then the cost per conversion is here as well, for each city so we can optimise
based on that data for each individual campaign or your whole account. So you can see here
a whole bunch of countries and territories and all sorts of stuff. In this view we are
going to filter so that we have got anything which has cost us more than £100 for any
given city. The thing that I should point out in a way, is that you really need a lot
of data to make decisions about this. There is no point in excluding or changing bids
on one or two clicks, you need a lot of traffic from any given city, town or location to be
able to judge whether this is working for you or not. Otherwise you could be excluding
some places which could work out to be quite good but you will never know. So make sure
you have got lots of data..FIRST TIP! The second one is, lets do our cost per converted
click is, let have a look at, lets say it is more than £20 and then we will order by
cost so we can see the highest spending locations. So a few immediately pop out here. This one
here and also this one here. So this one here Milton Keynes, we have had lots and lots of
clicks here, click through rate is kind of in keeping with the rest of the account, not
too high could be better. The cost we have spent £150 in quite a long period to be fair,
but we have only had one conversion, we have had one click converted that actually bought
something. So I might want to think whether I want to continue advertising in the particular
location. Lets just look at, lets just modify this a little bit. So lets look at a bit more
data here, so costs for these locations, so the Milton Keynes one again stands out like
a sore thumb. Durham isn’t performing too great either, so there is a few there we might
want to think about going “do you know what, do we need those locations? Are we getting
enough sales, can we optimise, is there something else we can do there? Or shall we just exclude
it?” which you can certainly do. So that is one way of looking at the data and as you
saw in these views that we can look at these different columns and change the sort order
and look at different things. We have identified a problem, so what do we do about it, how
do we modify this? So we come into setting go to locations and then we want to “what
triggered your ad” geographically. What we are looking for is, lets order by cost, and
then again we are looking for cost per converted click. Now United States, clearly there is
some campaign that has been running and pointing there and to surprisingly this is a UK company,
but the United States has wasted £940 of our pounds and we have got one sale to show
for it. My guess is this is historical data in this particular account and we are not
advertising there any more. Good! We certainly don’t want to pay £900 for one sale, not
unless we are selling something a lot more expensive than this website is. So next, moving
on, so there is a campaign here which is just looking at again United Kingdom, that campaign
is just not working. Lets have a look at United Kingdom and lets just look at county level
for United Kingdom. So, now a bit deeper into this data. Now Greater London £26 a conversion
for this particular campaign, Hampshire this is working well. Some of these are very low,
working really really well. What we are looking for is high spend, Greater Manchester so there
is a few there. Now I think in fairness just looking at this data, remember we have got
campaigns mixed up with the different locations. Now it certainly looks like this particular
campaign isn’t working very well at all. So its not the fact that it is Hampshire or Kent
or Greater London or whatever that is producing the high cost per conversion, it’s the keywords,
maybe the landing page, maybe the product price. There are any number of things it could
be thats actually kind of screwing this up. So in this particular instance its not Hampshire
that is causing a problem, this is just that particular campaign. So you need again just
to be careful about making decisions with this data. The fact that it’s not converting
well doesn’t mean to say it is the location, it could be other things. So lets dive a little
bit further through this, lets see if we can spot some others. Ok, so here is another one
Greater London £59 a conversion, £51 a conversion and here’s TSB1012 again £50 a conversion.
These are high. OK so we have identified a few of these and you will be able to go through
your own campaigns, remember you need A LOT OF DATA to make these decisions or you could
end up missing out on a heap of sales. So what do we do with this data? Lets assume
we are thing Greater London for that campaign, we just don’t want that. If we scroll up to
the top here we are going to get some options here to edit. Now what we can do is, the first
thing is, you know just exclude them. I don’t want to advertise in those locations for that
campaign, it just doesn’t work. You might put some deeper though into it and you might
think why doesn’t it work? Can I change something? Can I fix the problem or should I just stop
it? Quite often, it depends on the campaign, but quite often it is just a case of going,
do you know what? We haven’t got enough budget anyway why would I spend it in these locations
that aren’t converting? Let’s advertise in locations that do convert for us. So that’s
the first thing is just exclude it, take this option and turn it off and don’t advertise
in those locations. The second is that you could add these locations as specific target,
so remember you campaign could be targeting the UK but we can then add additional targets
in here and then say “do you know what, these locations they don’t convert well for us so
I don’t want to be in top positions, or I don’t want to be as high as I am generally
so I want to decrease those by 20%” for example. Or you might want to be more aggressive and
go 30% or less aggressive and go 10%. So you can keep monitoring these and you can see
here Oxfordshire already has an adjustment of 20% of our bids. So, as it says here if
we are bidding £2, for example then this is going to come down to £1.80. That could
mean the difference between being in position 1 or 1.5 to maybe position 2 or position 3.
It depends on the market and how competitive it is. I would recommend that you kind of
monitor this and make adjustments, or if you are more aggressive just turn of those locations
that aren’t working for you. BUT, do be careful that it is the location that it is the location
that is causing the problem. Always look at the campaign overall stats to start with.
If it is poorly performing everywhere your problem isn’t a location problem. If your
campaign is performing ok, moderately ok, then you might find that there are certain
locations for certain services where they just don’t work. Why that is, is going to
depend massively on the product or service that you are selling. If you are trying to
sell inflatable dinghies my guess is that it is quite possible that the sales are going
to be better near the coast than they are in the middle of the country. There could
be any number of permiatations why a product sells well in one location than it does in
another. It might be that the particular area, the product you are trying to sell is more
of a middle class or upper class product and your are targeting lower class areas with
lower class incomes and that is why it doesn’t work. There could be any number of reasons
why particular areas don’t convert as well as others and sometimes you just don’t know.
But, you DO KNOW from this data that it just isn’t working. So definitely spend some time
going over your AdWords account now. We cover this quite a lot in the AdWords training that
we do one on one online with people. i will put a link up to that for you, but this stuff
comes up and it is really handy. It could save you heap on your AdWords spend, or if
you want to be more positive about it areas that are working really well increase the
bid so you are in higher positions for those occasions where your sales do come much better
and your cost per conversion is lower and you will increase your conversion rates overall
over the course of the campaign and of course your sales and profitability as well. I hope
that has helped. As usual find links, find me on Twitter, I am @tillison. If you have
questions about this or anything to do with Google AdWords that you are stuck on then
ask in the comments and maybe I will do another T-Time show just for you to answer that question.
Until then enjoy your cup of tea.

2 comments

Awesome video and sorry if I you mentioned and I missed it but when you are bid adjusting based on CPA, how much data is enough to make an accurate decision?

I am using search campaigns and location only based targeting, no 'interested in' stuff.

It's a pity you can't use a calculator that considers statistical significance when adjusting for 'costs' like you can for 'through rates'.

Awesome video and sorry if I you mentioned and I missed it but when you are bid adjusting based on CPA, how much data is enough to make an accurate decision?

I am using search campaigns and location only based targeting, no 'interested in' stuff.

It's a pity you can't use a calculator that considers statistical significance when adjusting for 'costs' like you can for 'through rates'
.

Leave a Comment

Your email address will not be published. Required fields are marked *