Here is what they have done and what we are doing about it.
What have they done ‘exactly’?
Before, we could specify a precise phrase and trigger the ads and measure the results.
Now, we can no longer isolate words and measure the value in this way.
>> If we are not really careful, Google will show any ad to any keyword
The short version is that we have to rebuild all campaigns, for all customers in a better way than our competition do.
Luckily for us, it is really hard to build good campaigns now (and we worked out how to do it).
Cost Per Lead EXAMPLE:
Most things you see in Google Ads are an average – the ‘average‘ results of a campaign show that the cost per lead is £57.60 below (for a lucrative product)
But for very closely related keywords, the difference in cost per lead for KEYWORD A and KEYWORD B can be 300%. I’ve seen differences much higher.
Before, we could just switch off KEYWORD A
Now, if we do that, KEYWORD A will just show up in KEYWORD B group.
Unless we stop it, we are not measuring anything properly, and results are ‘averaged’ on our behalf. The water is getting back into the sponge!
The above chart is for very very closely related keywords, the difference in cost per lead can be 300%
Before, we could just switch off KEYWORD A. Now, the keywords we switch off, will just show in another closely related group.
Unless we stop it.
This change means Google can match any search phrase they like to any of our ‘exact’ keyword phrases. It makes it way harder to manage and control campaigns.
We can’t easily switch off poorly performing keywords (we can, it just takes longer)
It’s like keyword whackamole. We switch off the keyword and ad – but now, Google merely shows it on other campaigns. We have to go an negate it.
Rather like the ‘frog in the saucepan’** Google ‘seem’ to be turning up the heat and hoping we didn’t notice. We first properly felt the effects in April and May.
Why does it matter?
If we couldn’t measure things precisely, we would end up paying for both high and lower performing keywords.
To be clear, we still have the power to avoid keywords we don’t want. But, if we can’t separate ‘ok value’ from ‘great value’ because we can’t measure it, then we will slip backwards. We always have more clicks available than budget.
Thankfully, we worked out how to fix this, and quicker than we first thought.
Making it harder to ‘not bid’ is likely why Google did it. They think it is in our interests to bid on everything.
Smart, they call it.
You have to be kidding?
You would think so, but no.
“Exact” is now the biggest ever misnomer. It’s a bit like saying Google make “quite a lot” of money.
Exact match actually means – ‘we can show whatever we want’.
But by showing the ‘wrong’ ads for the wrong keywords, they are affecting the user experience for the worse, so it is a cynical move. The normal response of it ‘being good for advertisers’ might ‘possibly be true for badly run campaigns, but that is small comfort for us.
What’s the Answer?
Do we actually need to redo everything?
We think the better we do the more likely Google will switch up their fuzzy meter. Just because we are paranoid doesn’t mean they aren’t out to get us. We have to be able to trust the campaigns.
The process to fix this has not been easy. It was just plain confusing for a few weeks.
Onward and Upwards
Let’s do it. Roll up the sleeves!
After a bit of complaining and moping around, we are now confident we can eliminate the effects of this change and have run tests.
On the test campaigns, the performance seems better than before.
- Locked down keywords, restored cost per lead
- Higher Click Through Rate (CTR) (more market share) better click price.
- Campaigns will be even more granular – which means more controllable in the long term.
Let us hope this change has made it harder for the competition to be profitable.
Our new system is ready.
It means consistency; we can all contribute to campaigns and understand the complexity. It includes;
- Some standard naming conventions that turned out to be extremely important
- Some new software to help us with the process
- Our system of reviewing that ensures we don’t repeat the analysis endlessly
These new structures will be extremely positive for us and the campaigns going forward.
The problem with that is
There are some downsides – other than the swear jar being full.
Wow, this is complex.
- Implement a highly fiddly configuration involving cascading negatives to separate words
- It’s like, instead of switching on hundreds of keyword taps, we now have to pay attention to the entire waterfall and switch off a drip at a time.
- Take way more time than before
- Sign up for new software and learn how to use it
But, by following the procedures, we should quickly recover performance overall.
What does it mean for you?
We hope it means we will be more competitive once we have rolled out. We are doing this without passing on the cost of our hard work, new people and the software we need. We will sort this as soon as we can, please be a little patient with us in the meantime.
How long will it take?
We have to be smart about this. We need to focus on the best performing and the highest click campaigns. The majority of progress will take a month or two, and then we get into the long tail.
Is there a silver lining?
Yes. Many linings.
There is always so much opportunity to improve.
We believe wholeheartedly in the aggregation of marginal gains. The Brits used this philosophy to completely dominate the Tour de France for the last few years. Lift lots of things by 1% and you drastically improve and compound the results.
In campaigns, we have so many percentages to choose from, and most of them deliver way higher than 1%.
What happens if Google changes it again?
Every couple of years the goal posts change. And so will we. It’s not just the traffic that gives us the results. It’s the analytics, decision making, product strategy and messaging. We will never run out of marginal gains.
We have to start looking at other elements that give ‘our clients’ Competitive Advantage– and start plugging the leaks in the bucket.
** the tale of the frog in the saucepan is an urban myth. Not true, but it shows the point.
It says that if you dropped a frog in hot water it will jump out right away. But if you instead put it in cool water and slowly heat up, then it will sit there till it cooks.
Perhaps we weren’t supposed to notice Google cooking our