Hello! My name is Daria Dane. I have been scaling businesses using contextual advertising for 10 years.
-
Yandex training expert; author of courses in Yandex, Yandex Workshop, MaEd, Elama, Skypro.
-
I’m leading telegram channel about attracting clients through Yandex Direct, Google Ads, Facebook, Telegram.
In advertising, 2023 was marked by a flurry of spam requests on the websites of many advertisers. In 2024 the situation gets worse. With the advent of auto-strategies and pay-per-conversion came a huge number of savvy webmasters and competitors who invented new ways to get people to click on ads. I told who and why clicking is needed Here.
In this article we will talk about how to deal with fake applications in Direct. Suitable for all advertising sources.
-
What are untargeted leads?
-
What determines the % of non-targeted leads?
-
Six methods for cutting off non-target applications
According to a study on lead quality conducted by the Callibri service
on 3,200+ clients, approximately 50% of the budget is wasted: real leads – a little more than half. The rest is spam, missed calls, requests from partners, repeated calls.
What leads are called targeted in relation to paid advertising?
-
bots (+incorrect number, offline number, etc.);
-
“I didn’t leave a request”;
-
wrong number (they called the wrong place, a common problem when connecting call tracking);
-
I submitted an application, but it does not meet the criteria (another region, does not meet the requirements).
It would be nice to get rid of all categories.
Fake leads depending on the niche
You need to understand that it is impossible to completely get rid of fake and non-targeted applications. A certain percentage is everywhere and will always be, it’s about up to 20%.
To avoid being caught off guard, take this into account at the media planning stage. Set aside non-targeted leads and budget for them. How to calculate a media plan for contextual advertising is in this material.
How many specific spam and non-targeted requests there will be depends on the niche and size of the advertiser.
For example, I struggled with fake leads in the financial sector for several years and was not able to completely defeat them. To understand the scale, 90% of applications from YAN belonged to one of the 4 non-qualified categories listed above. But, thanks to the business model, the remaining 10% of targeted leads covered everything else and gave a good profit. But this case is rather an exception.
Moral – know your enemy by sight and consider in advance what you will have to face.
Fake leads and Direct auto strategies
Special attention should be paid to Yandex Direct auto strategies. In itself, I think this invention is not bad. But there is BUT.
If you are already caught in a meat grinder with fake applications, the auto strategy can turn this into a real disaster.
How?
The optimization mechanism works on the LAL principle – to bring users similar to those who completed the specified target action.
If your action is already performed by a large percentage of bots, using an autostrategy will be tantamount to s*** on the fan.
Once I went on vacation for a week, and when I returned I found that the client had received a flurry of fake applications for 120 thousand rubles. 52 leads, 0 of them picked up the phone. Of course, I returned the money through my colleagues at Yandex, but the aftertaste was unpleasant.
How to deal with fake applications: Elimination methods
1. Text-level clipping
The first and most obvious method that comes to mind for the vast majority of clients. “Let’s change the USP so that only one million leads come to us?”
Essence: change ads / creatives / texts on the site
To be honest, the method doesn’t work. More precisely: all this needs to be done at the stage of castdev and competitive analysis. Without USP, selling texts and functional website design, advertising will not work at all. But the root of the problem of untargeted leads is different. Bots don’t read USPs.
Therefore, a small part of the audience is cut off here, many non-targeted ones pass on, and bots are not cut off at all.
2. Transfer of offline conversions to Metrica and payment for qualifications. lead This is the most interesting and working method.
It allows you to take the analytics of advertising campaigns to a whole new level – it appears ability to build any reports on target leads in the Report Wizard. And with enough data – Pay only for targeted leads.
Essence: When a lead moves through the stages in the SRM, the goal is achieved in Yandex Metrica.
And as we know, Direct campaigns can be optimized for goals from Metrica. Those. – set a qualified lead as the target action for the auto strategy.
Problems:
-
Technical component. You need to configure the transfer of UserID / ClientID / Yclid from the site to CRM, and from there to Metrica. Usually with the help of developers, or no-code solutions. Many people struggle with this point: it is difficult to bring together site developers, CRM integrators and advertisers.
-
There is not enough data for the strategy to work. Even if you correctly transfer data about qualified leads to Direct, autostrategy may not have enough data for training. If there are less than 10 leads per week, it’s definitely not enough. A question of business scale.
But everyone should bother with the settings. Even if you have 1 target lead per week, in Direct reports you will see exactly where it came from and you will be able to manually adjust something. The quality of analytics will increase in any case.
You can read a separate material about this method in more detail.
3. Complicating application forms + Google reCAPTCHA v3 The point: prevent bots from filling out the form
Essence: prevent bots from filling out the form
The first step is to install Google reCAPTCHA v3 (invisible form protection against bots).
Next we complicate the shape. This can be implemented like this:
-
Two-level (when there is a “Next” button).
-
Decoy field (a simple question that needs to be answered, for example 4+2 =)
-
Color captcha. Less annoying for people than entering letters/numbers.
Start with the first form on the first screen – bots often fill out this form. And bots don’t know how to answer questions.
4. SMS verification
This is a variation on the previous point, requiring technical implementation and additional. costs for the SMS sending service.
Essence: confirmation of phone number when sending an application
It cuts out bots and various “I didn’t leave” messages, but can reduce conversion due to people’s reluctance to confirm the number.
In my experiment, the conversion to a qualified lead increased 3 times:
5. Quiz with dead-end questions
We do a quiz with questions that classify leads. 3-5 questions that a script manager usually asks. For example: minimum order amount.
If the answer to some question does not suit you, we issue a page a la “Sorry, we can’t help you.” Thus, we relieve the sales department and cut off non-targeted ones.
Problems:
-
some bots can give all the correct answers;
-
in my practice, “I didn’t leave a request” are still present.
In my experiment, adding dead-end questions reduced overall conversion by almost 1%, but increased conversion to qualified leads.
6. Clipping at the Direct level
There are several techniques that can be used directly in your advertising account.
-
The simplest one is a copy of an advertising campaign for a new goal.
We create in Metrica a copy of the original goal for which the campaign was optimized (the same one but with a different id). Create a copy of the rk and change the target in the settings. -
If the main requests come from calls – As an optimization goal, we use only a unique target call. These are unique calls lasting more than 45 seconds. Call tracking connection required.
-
Multi-level goal values. We set optimization for several goals in the campaign settings. We assign a low value to those that are achieved easily and simply, and high value to real conversions. If we transfer offline conversions, we simply set the leads to a low value and a qualification. leads – high.
More information on learning strategies and choosing different optimization goals is available in a separate material.
Summing up the material and answering the question “How to get rid of fake applications on the site?” I will say that:
-
It is impossible to get rid of fake applications 100%. Factor this into the economics of advertising.
-
The best method, close to 100%, is optimizing for qualified leads. Get confused with setting up offline conversions.
-
Test different methods, get rid of fake applications by 30-40% – this is a huge budget savings.
More interesting materials in my telegram channel.
Acknowledgment and Usage Notice
The editorial team at TechBurst Magazine acknowledges the invaluable contribution of the author of the original article that forms the foundation of our publication. We sincerely appreciate the author’s work. All images in this publication are sourced directly from the original article, where a reference to the author’s profile is provided as well. This publication respects the author’s rights and enhances the visibility of their original work. If there are any concerns or the author wishes to discuss this matter further, we welcome an open dialogue to address potential issues and find an amicable resolution. Feel free to contact us through the ‘Contact Us’ section; the link is available in the website footer.