As a finalist for the Google Premier Partner Award, I recently had the chance to visit the Google campus in New York. There, I spoke to a googler of the state agencies and he lamented the fact that a big wave of sponsored link agencies (PPCs) having started in the last economic downturn in 2008 are still managing their accounts as they did a decade ago. success when we risk being on the verge of another economic downturn.
Given the importance of the changes made to Google Ads over the past decade, it goes without saying that successful account management should have also evolved dramatically.
My friend and pioneer of the industry, Andrew Goodman, recently proposed the idea of agencies not corresponding to the time in slightly different terms and wrote that self-proclaimed CFC experts do not follow the nuances of all. This has changed in Google Ads because they choose rather well than looking for greatness.
The following is therefore my reflection on how to put your sponsorship management skills into good shape for 2020 and beyond. The basic idea is that the use of Google's automatisms is inevitable. If you want more control, you can recover it by superimposing your own strategies on those of Google, thanks to a concept that I call layered automation.
It's not a question of whether, but when automation would do better
In 2007, Google had launched Conversion Optimizer which allowed advertisers performing at least 300 conversions over a 30 day period to achieve a CPA goal in relation to their ads. Today, this strategy calls CPA target and requires 20 times fewer conversions (15) to achieve similar results.
There are now about 11 types of bid management strategies. Some, such as Target CPA and Maximize Conversions, seem almost different, showing that things have progressed far enough that the differences in strategy can be reduced to nuanced goals. It takes serious know-how to understand all of these strategies and their interaction with manual controls such as bid adjustments
The fact is that the automation of bids has come a long way. And it's not even about automation in creatives, targeting, etc.
The infiltration of automation in all areas of the CPP will continue unabated because of two driving forces:
- A mine of data in constant expansion. Whenever a search is performed, Google learns a little about the correlation between users and about the ads that interest them. The more data we have, the better the signal.
- Moore's Law and the seemingly endless cycle of computing power doubling every 18 months. Since computers can do more calculations with the vast data in point 1, they can make better predictions. And although Gordon Moore believes his law will reach its limit by 2025, teams are already working to ensure sustainable growth in computing power.
So, try to understand how automation fits into your business plan. Too often, I hear that this is something that people are too busy to understand right now. "We could get there another day," they say … well duh! If you've used automation, you may have the time to demonstrate strategy for your own business and perhaps even for customers who pay your bills.
Now is the time to get ready
Even if you do not live in California Like me, you probably read that PG & E, the utility company, cut off power to 800,000 customers to avoid catastrophic fires due to aging and poorly maintained transmission lines, in order to avoid the type of fire that destroyed the city of Paradise in 2018. does this have to do with PPC ? It showed me how much people thought and planned in the long run.
Now that electricity is off, everyone is upset about PG & E because households are in the dark. And although the utility company is certainly not flawless, it has been warning its customers for months that this would most likely happen. But until the warnings materialize, few consumers have been slow to prepare.
Google executives, bloggers and industry peers all said that automation would disrupt us. Yet, too few organizations consider warnings and wait until it's too late to change.
Complement rather than compete with automation
Thus, while it is inevitable that automation will be permanent, a modern account manager will know better how to make it a part of their routine. But if, for whatever reason, you still believe you can compete with automation and win, listen to this advice from Hal Varian, chief economist of Google
. you are looking for a career where your services will be in great demand; you should find something where you provide a rare and complementary service to something that becomes ubiquitous and cheap. So, what becomes ubiquitous and cheap? The data. And what is complementary to the data? Analysis. "
You know what's becoming ubiquitous in PPC? Automation. Most of these automations are not only cheap, they're free and accessible to all users in Google's own ad management interface. the kind of thing that Varian was talking about. "Taking into account his advice, it seems more appropriate to become a great complement to automation than to fight it by continuing to do the same things by hand.
Automated Tools do not guarantee success – this is the case
The reality of PPC automation is that results vary from one advertiser to another, so that everyone does not see results worthy of case studies when deploying Google's latest smart feature, but on average, automated tools deliver better results with less effort. As an aberrant person who has not succeeded, you must first determine why. Then you can plan a strategy to put you on the right side of averages.
For the moment, much of the automation we see solves very narrow problems. There is a system to automate bidding, another to optimize ads, and another to match those ads with likely leads. The role of the agency is to bring together the right solutions and ensure that they work well together. For a closer look, check out my article on how the Wrong Combination of Automations can destroy an account.
By the example of the error committed using an attribution model as the last If this does not work with the other automatisms, it is clear that the results of the automation can be extremely dependent on the knowledge and skills of the account manager. If you do not see the results obtained by others, consider that the problem may not be due to the tool but to the person who uses it.
As a person who builds PPC tools to make a living, I know that even if the user is to blame for the bad results, it is still the problem of the creator of the tool. A great tool for automating the test of ads is useless if it is too difficult for an average advertiser to use it without committing an error. But the best advertisers do not leave out the gaps in technology and will strive to produce amazing results with what's on the market.
Automation does not mean that we rule and we forget
There is this interesting cycle that Google's tools often seem to go through. They start manually, then become automated, and eventually new controls are added. This turns automation into a new type of manual tool.
Bid management is a good example. First, we managed the CPCs and bid adjustments manually. Then, it became automated as a smart bid strategy as the target ROAS. But then, Google added the ability to set group-level ad goals, mobile bid adjustments, and seasonal auction adjustments. If you wish, the automated target ROAS bid strategy is rather manual if you take into account all the parameters that you can now control.
The simple fact that there are several parameters and controls for automated tools should be a flawless gift so that automations can be optimized if advertisers are sufficiently motivated.
I recently asked advertisers the last time that they had an account with only one CPC offer. Or the last time that they had manual CPC bids that they had never changed. The answer for most was "never!" Target CPA and target ROAS targets should also not be the same for the entire account or static. They must be managed for better results and certainly should not be treated as an unstable tool.
Humans Can Make Meaning Of The Data
I am curious why I say that the target APC and the target ROAS need to be monitored. and managed rather than left alone? We should do this because there are factors that affect conversion rates, which Google's forecasting systems may not be able to control. Automation only detects a change in metrics and its narrow scope of what it can do can very well use this data to make the wrong choice for the company.
Here's an example that I've personally encountered … automated auctions. One day, we found that conversion rates had dropped significantly and that bids had therefore been reduced in order to maintain the target CPA. Due to the much lower bids, the advertiser's conversion volume was not recovered automatically.
Here is what happened. The auction automation has correctly seen a decline in the conversion rate and adjusted the bids downward. But as silly automation, he never asked why the conversion rate had dropped. Humans knew that it was because a new landing page had been launched. Humans would have known that the correct answer to this event was to return to the previous landing page rather than lower the bids.
The machine learning is bad, very bad to explain why and how. Show him a picture of a cat and he knows that it is a cat. But good luck telling him why he is a cat.
In my book on the future of digital marketing, I explain that one of the roles that humans must play is that of PPC driver, someone who monitors automations and can make course corrections if bad data leads to bad actions.
Understand what the machine does and make sense of the data it uses to make decisions. That's about what Hal Varian said in the quote I already mentioned.
If automation does not work or if you do not believe it, consider the overlay of automation
If you think that automation will become even more ubiquitous in PPC in the future, it is therefore logical that we should learn the operation of the automations provided by the engines in order to optimize them by managing their parameters.
It's the simple principle that humans + machines are better than machines alone. But I think in PPC, this premise has extra meaning. The equation should perhaps be:
"The man (account manager of my company) + machine (automation created by Google) is better than the machine alone."
The concern of an advertiser is not just that machines take over our jobs, but it's because these machines are built by Google, which also collects a lot of our advertising dollars. As a former Google, I hope that Google is trying to do what is right, but there is nothing wrong with wanting collateral and setting up a surveillance, especially when even the Googlers can not really explain how their machine-learning automatisms are coming to an end This is why CPM managers want to be part of the equation rather than letting the machines do it themselves . Humans can monitor machine decisions and provide corrections and advice when these decisions seem suboptimal.
Make the process of your PPC work for better results thanks to automation
However, adding human labor to the combination is detrimental to efficiency and economic growth. The automation would be better. Account managers should therefore strive to have their own automatisms to monitor and optimize engine automation.
So, instead of trying:
Man + Machine
We Should Strive:
Controlled Automation by Advertisers + Engine Controlled Automation
But Builders Are Also Enthusiastic that builders want to build theirs automation (thanks to all who downloaded my scripts over the years ), the truth is that it does not always come naturally. Third-party tools can be useful.
They help advertisers control how their accounts automate. I've explained how to use overlay to automate the monitoring and control of close-up variants that can confuse the meaning of exact match keywords. The graph below attempts to explain it in a more conceptual way, but check out my previous post if you want a more tactical guide.
Close variations allow Google to target keyword ads (the small circle in the middle). ) to a much wider set of search terms (the big outer circle). Google controls the size of this outer circle. If they want more revenue, they can literally change some code settings to enlarge the circle and make the ad bids more competitive. Advertisers want more certainty and control. Thus, thanks to the automated layers where they control automation, they can reduce the search terms (the dotted circle) to a level that suits them,
. The bottom line
The engine automation is already disrupting the sponsored link agencies and will continue to do more. All those who practice CPP as in 2008 must accept that this is not a long-term strategy. Determine how to integrate automation into the composition of your work. And if you do not believe it, know that techniques such as automated overlay can restore a certain level of control.
The opinions expressed in this article are those of the invited author and not necessarily those of the search engine. Associated authors are listed here .