William Charlwood

To make AdWords work most profitably you need to beat Google’s system or you need to beat your competitors. Preferably both.

This article describes an advanced technique that can give you an edge and in certain circumstances can boost your ROI by several percentage points. It concerns time-slicing your campaigns to find out when they are most profitable.

Nearly everyone using AdWords learns the hard way that some keywords are more profitable than others. More experienced users learn that different ads are more profitable too. Advanced advertisers start targeting their campaigns on a geographic basis because it is often easier to sell to local customers than to ones a long way away.
But what about different times of the day? Take yourself for instance. Are there particular times of the day when you are more likely to spend money than at other times?

There’s probably no hard and fast rule here but I know that I am more likely to spend money for my business in the afternoon. I’m not sure why, it’s just that looking back on my recent purchasing history, I’ve reached for my wallet more often in the afternoon than in the morning.

For personal purchases, I tend to buy things online in the evening. I may do some initial research during the day, perhaps in my lunch break, but my real spending goes on in the evening. I’ve also noticed that I get more sales enquiries from a campaign I use to promote my AdWords optimisation site in the afternoon than in the morning.

So if you are an advertiser, how can you use this information to boost your own profitability?

The first thing you need to do is find out when your pay per click / AdWords campaigns are most profitable. It’s quite easy to do this on a daily basis by running the appropriate report using a few weeks’ data at the day resolution level. This will tell you what days in the week are most profitable for you provided you’ve set up conversion tracking or some other suitable analytics system.

But to fine tune the process and gain competitive advantage – something that’s vital if you are operating in a hotly competitive area - you really need to identify what hours in the day are most profitable too. This is not so easy because at the moment Google doesn’t report conversions on an hourly basis.

So we need a workaround.

The way to do it is to use the AdWords Editor tool to create mirror copies of your campaigns and then use the Scheduling facility to run them at different, non-overlapping times. For example, you could run one copy from midnight to 6am, a second copy from 6am to 12 noon, a third copy from 12 noon to 6pm and the fourth copy from 6pm to midnight.

After a while you’ll be able to see from the conversion tracking data which time period is most profitable. At that point you may wish to adjust the bid prices you set in each campaign to reflect their relative profitability. If in your business you tend to get higher conversion rates in the morning, it would probably be worth your while boosting your keyword bid prices in the morning so that you get more of this profitable traffic and then dropping them back again during the rest of the day. This is something you can do quickly, easily and automatically by using the advanced settings within the AdWords scheduling facility.

You can set the system up to increase or discount your keyword bid prices across a whole campaign for your chosen time periods. For example, you may decide you want to boost your bids from 9am to 11am by 20%. Cut them by 25% in the afternoon, and then take them back to 100% the rest of the day. Once you’ve set this up Google will automatically adjust your bids on this basis according to the time of day so you don’t have to fiddle about making loads of changes all the time.

The time dimension is easily overlooked in AdWords campaigns but effective time-dimensional analysis can boost your AdWords ROI by several percentage points. It takes effort but if you are spending a lot of money, you should try it.

I do a lot of managing of other people’s AdWords accounts and for most people it makes sense to sub-contract this function simply because the system has become very complicated.

Part of the sales process in running other people’s accounts is to take a look at what’s going on right now and point out some errors that make intuitive sense to them - once I’ve highlighted them. Typical problems I keep seeing are that keyword lists are too short, conversion tracking is not used at all and that very few ads have been tested out.

Recently I took over a client’s account in a service industry and by doing some sensible experimentation I managed to double the click through rate on his account simply by making some copy changes in his ads.

It was an iterative process involving trying out various combinations of adjectives until we hit a clear winner. Interestingly, we then had a discussion about what was going on. The problem both my client and I suffered from was feeling that some copy was just too “American”. In other words, “in your face”, gung-ho etc. But the fact is that copy that from a British angle can look just too damn punchy does seem to work even in the UK.

We were advertising in the UK and my client wanted UK-only clients.

The adjective that seems to do the trick best at the moment is “power packed” so if you are looking for an adjective to add to your own AdWords ad copy, try using “Power-Packed” and see what happens!

Here’s a quick, cheap and simple way of doing some online market research.

Suppose I want to sell widgets (they seem to be quite popular after all) then should I stock red, green, blue, and yellow widgets or just red ones? Which widget is most popular?

What I can do is set up an AdWords ad promoting my widgets and then run a campaign using the keywords “red widgets”, “green widgets”, “blue widgets” and “yellow widgets”.

After a while, I’ll start to get three sorts of data back from Google.

First, I’ll see how many searches there were for each keyword and in general the number of searches done relates to the market interest. So if more people search for red widgets than blue widgets, I’ll know that there is more interest in red widgets.

Second, I’ll see which keywords generate the most traffic which will give me a better idea about the potential demand for my widgets: people who click on ads are more likely to be interested in buying widgets than people who just search for the things, especially if I include the word “free” as a negative keyword in my campaign.

Finally, I’ll start to see where my ads rank for each keyword which, if I’ve set the bid prices to the same level for all my keywords, will tell me some interesting things about what my competition is doing.

For example, if the keyword “red widget” causes my ad to rank lower than for the keyword “green widget” then assuming my competitors are smart, it is likely that red widgets are more popular and profitable.

What should I do about this? Stock red ones.

If you are thinking of opening a store selling branded goods such as motorbikes, a quick bit of market research like this can tell you which brand to run with: Honda, Harley, BMW etc. You clearly need to take into account local factors here and that is something you can also do with Google.