Why can targeted advertising be smart enough to show me ads at home for something I searched on my PC at work, but not smart enough to not show me the same ad 10x in a row or for services for which I’ve already subscribed?
In: 128
This is a symptom of poor marketing/comms/customer experience strategy, and tech limitations
Think about it like a store. You walk in, storekeeper asks you if you want to buy a shirt. You buy a shirt. The storekeeper knows you bought the shirt, so they don’t ask you to buy it again.
In the digital space (ie online stores), the storekeeper can (and should) leverage data integrations to tell the advertising platforms that you have bought the shirt that was advertised to you.
However, this doesn’t always happen, usually due one or a few of
Ignorance – don’t know it’s possible
Ambivalence – they don’t care
Tech limitations – they don’t have the right tech in place for the integrations
And a 4th reason that’s coming up more and more is privacy. Browsers in particular are limiting the ability for advertising and marketing technologies to accurately identify buyers, which means it’s harder to notify the ascertaining platform to “turn off” the advertising after the purchase has happened.
Source: I lead marketing technology strategy for global enterprise organisations.
Because being better at it might require they were MUCH MORE creepily up in your business about it, and it would be a lot more expensive (at scale) in order to make it work for them.
To not show you the ad for the thing to which you subscribed (say, a cook on OnlyPans.com), FaceGoogle would have to know (from OnlyPans) that you actually subscribed–FaceGoogle would have to have some standing agreement where OnlyPans would send them huge (demographically valuable) data-dumps of all the FaceGoogle accounts that had created OnlyPans accounts, so FaceGoogle could avoid sending them repeat ads (but use that valuable info to sell BETTER ads, too!) This way, when FaceGoogle is considering what ad to show you, instead of just doing the work of figuring out that you like cast-iron cookware, they’d ALSO have to cross-check your subscriptions (OnlyPans, SetPlix, YouVideo, etc) to see which things NOT to show you… Or (more likely) they’d use the OnlyPans subscription info to figure out BETTER ads to send you, in order to defray the cost of that extra data handling.
But honestly, OnlyPans probably only pays $2/adclick… And OnlyPans doesn’t mind paying that $2 for an existing customer very much. Maybe it’s not worth the extra work. And maybe after FaceGoogle realizes you didn’t click on the OnlyPans ad (for whatever reason), they understand better that it’s not worth showing it to you anyway, and they downvote that ad in their auctions next week.
In fairness, there are “conversion fees” that FaceGoogle can ask for (basically “we’ll only charge you x% for showing your ad, if you give us an extra y% if that ad-click turns into a sale), and for some of these things it might be useful… But it’s a bunch of extra work, a bunch of extra expensive software and data… So it’s not always obviously worth it to the companies involved.
I purchased a thing after a day or so of searching and I’ve had ads for the exact same device suggested to me for over 3 years now. It’s super awesome.
The short story is: the best and most-coordinated ad campaigns would likely try to avoid serving you ads as an existing customer, but doing so might not be possible, practical, or cost-efficient compared to simply serving you the ads!
The longer story is this:
When an advertiser targets an ad, they do so based on many different things about you – AKA demographics.
The available demographics to choose from differ based on the advertising platform, but they can include or exclude everything from your physical location, your hardware, interests you have shown via your browsing habits, and if you visited their site in particular.
On many ad platforms, those things are associated with you via a unique user ID. That user ID is often anonymous (sometimes by law). They don’t know who you are, exactly. They don’t necessarily know your email address, either. They know you as a collection of demographics to which they’ve assigned an ID. That may also correspond to the ID of your specific hardware device.
Many ad platforms can keep track of you as a unique user across multiple devices thanks to some shared logins to sites you use across those devices. Other ad platforms cannot do that, and are just getting lucky on other demographics you share in your browsing between work and home.
So: you may or may not be known as the same person at work and at home. Based on your example, let’s say they *do* know you as the same person.
To avoid showing you an ad for a service you already bought, the advertiser needs to identify you with NEGATIVE demographic – one that should be EXCLUDED.
Except… how can they achieve that? And, is it worth doing?
For many advertisers, they might have an email address associated with your login. That’s the best negative demographic! But, does the ad platform have an email address associated with your unique user ID? And, is that the same email you signed up with?
Also, depending on how many subscribers the advertiser has and how often they change, doing that might take *a lot* of manual effort and coordination with their ad platform. It might be cheaper to simply serve you some ads.
The advertiser COULD use visits to the logged in portion of their site to exclude you. But, that takes more work, including editing their site. And, they might be happy to sell upgraded services to existing subscribers or to capture recently-lapsed subscribers, making this impractical.
The best and most-coordinated ad campaigns would likely try to avoid serving you ads as an existing customer, but for many advertisers the 10x you saw the ad is merely a rounding error in their advertising metrics.
If you think about it from the ads company’s perspective, they are trying to put an ad for a product that you are most likely to buy/click on. Even if they have shown you this product before, you might still be most likely (based on their data) to click on it compared to all the other products that they could show you an ad for. Usually the probability of a user clicking on an ad for a product decreases as more ads are shown to them but it is not as drastic as you would expect.
This is the high level reason but as you can expect, it is actually much more complicated in reality so I will not go into more details.
Source: worked on such algorithms for many years
No monetary incentive to show you variety.
THE CONSUMER does NOT pay them to present ads. The SELLER of the product pays them.
I am assuming that they charge them per period or per commercial, in which case, there is no incentive on their end to make a more complex algorithm.
I am not good at explaining, good luck.
It’s important to know that quite often, ads are managed by agencies or people external to the business that the ads are for. There are a lot of ways to advertise so it’s a pretty big job, and there are also a lot of ways to either optimise or overspend your ad budget based on whether the person who set them up knows what they’re doing or not. Being done externally means it’s usually up to that person’s personal opinion about who sees the ads.
There are definitely ways to stop you from seeing an ad for something you’ve subscribed to, such as through your email address you’re signed up with to the subscription service, or a cookie stored on your devices if you’ve logged into that site from there. Depends on the network doing the advertising but it could be the case that your Google login email address is different to the one for your subscription so there’s nothing to match it up with and take you off the targeting list. In the case that it knows what you searched at work and shows you ads at home, you’re likely logged into a service or browser on both devices and it knows your info through that (commonly either Google or Facebook networks or both).
I obv don’t know the ads you’re seeing for the subscription service but ads aren’t always made with the goal of getting you to sign up. Often it can be used to be a subtle reminder to people to keep using it, or showcase new features or things to try out etc. Or deals etc.
Also, unfortunately I’d say a lot of ad revenue comes from people who don’t know what they’re doing and it’s not I’m the advertisers interest short term to tell them they’re overspending most of the time.
Source: work in advertising (sorry)
Many brands have a gap between their owned purchase data and the other data points they pay for in ad targeting. Mapping this customer journey is actually big business for some data/tech companies.
In theory you might think you would want to “fall out” of the targeting “pool” after a purchase, but that capability/level of targeting is more expensive.
And, some brands like to target you just after a purchase *intentionally* for purposes of “cross-selling,” as in, buying more shit to complement the shit you just bought. Socks for your new shoes, for example.
Whether intentional or not, brands’ data might not always super accurate or recent. Take social media for example. Facebook has kind of a shitty “match rate” to customer files, so even if your purchase was tracked, Facebook might not be able to detect you in the brand’s customer file. Or maybe the brand only refreshes their files once a month.
So yeah, op, adding to what some of the above comments say, over-saturating isn’t hurting the companies’ bottom line. An impression served to you costs a few cents (or a few dollars) and in some cases is less expensive than the technology to take you out of the pool.
Latest Answers