Hacker News new | past | comments | ask | show | jobs | submit login
Chase had ads on 400K sites then on just 5K and got the same results (2017) (nytimes.com)
125 points by Ibethewalrus 45 days ago | hide | past | web | favorite | 53 comments

>“It’s only been a few days, but we haven’t seen any deterioration on our performance metrics,” Ms. Lemkau said in an interview on Tuesday.

The ephemeral "brand awareness" provided by most advertising takes more than a few days to fall off. Chase isn't going to get someone to open an account just because they saw a banner ad, their goal is to familiarize people with their name so that when someone wants to open a bank account Chase seems like an option. People are not going to forget about Chase bank...

Yeah, I'm kind of surprised that the editor didn't kill this story.

The cynical thought would be that this story exists to encourage other big brands to be more choosy in the sites they advertise on, and what better site to advertise on than the NY Times?

NT's editorial line these days is to push for stories without much peer review if I trust how poorly most of their features are written.

This! With a well focused product, you may get customers on very small but focused websites. Also on most ad networks, on which site, the user is matters less and less due to heavy user tracking.

Other key facts:

> Surprisingly, the company is seeing little change in the cost of impressions or the visibility of its ads on the internet, she said. An impression is generally counted each time an ad is shown.

I take that to mean that the tail didn't cost very much, and didn't increase ad visibility very much.

I think it's the exact opposite - the CPM didn't change, so the tail is just as expensive, at least the way Chase was using it.

OK, I see that. They were spending on stuff that produced few impressions.

> The ephemeral "brand awareness" provided by most advertising takes more than a few days to fall off.

Marketing metrics in general move slowly: think weeks, months, or even longer for significant moves. A lot of short term moves are noise generated by imperfections in sampling, particularly with daily collection. This applies regardless of the metric: awareness, consideration, affinity, penetration.

I don't know what she means specifically by "performance metrics", but it would be a pretty poor showing for the CMO of a global company to expect significant moves in any marketing metric over the span of a few days. I'm therefore guessing she means something else - like impressions, clickthroughs, or sign-ups - and that the article is simply of poor quality.

I'm straying off topic here, but generally the only time marketing metrics make significant moves over a short period of time is if something major happens that hits the news or goes viral. Even here it may only be changes in media metrics, such as buzz, that aren't reflected by changes in more substantive areas.

E.g., I can see a gigantic spike in negative buzz for KFC in the UK in late February 2018 because they ran out of chicken nationwide (I kid you not). I can also tell you that whilst over the next several months their penetration dropped by a few percent (significant at the levels we sample), no long term harm appears to have been done to their business: penetration quickly returned to its pre-crisis levels.

The reason nothing has changed could at this stage be attributed to the fact that it's too early, or simply that nobody cares Chase has stopped advertising: they're so ubiquitous it doesn't matter.

And if you're still looking for more reasons why this isn't news, this might be the killer: it's pretty well understood that you're going to get a different response to online advertising depending on where you place it. Some sites simply give you more return than others on your advertising investment, and the differences can be stark. I'd bet Chase has limited their ads to the 5000 sites that give them the best ROI so, again, no surprise that nothing has changed.

How do you measure brand awareness? It’s certainly either bullshit or trivial to measure.

The most believable sense in which brand awareness is a thing is in the number of people who are passively aware that your brand exists. For example, if you wanted to visit a grocery store, you probably could think of a list of stores in your area that you might want to go to. If I were going to build a new store, I would want to put myself on your mental list of stores, so that if you ever needed something I was selling you would know to visit. That's more or less the quintessential example of advertising. It's also easy to measure in an objective way, because any given individual has either heard of Tesco or hasn't.

There's a spectrum of marketing that ranges from "how can I communicate the existence of my cheese spray product to fans of Cheese Whiz," to "how can I get every American to spray cheese wiz directly into their mouths three times a day as a meal substitute." Techniques for doing the first are about as well tested as Newton's laws, but techniques for doing the second are virtually guaranteed to be a scam where you're the mark.

The thing brands get most excited about is "spontaneous" or "unprompted" awareness. E.g., I ask you a question such as "Thinking about womenswear retailers, which names come to mind?"

Brands want to appear on that list of names, and the closer to the top ("top of mind") the better. Of course, for most brands the aspiration to be top of mind is ludicrous since there can be only one. For UK womenswear that one is Next and, barring the company suddenly going under (possible but not likely), it ain't changing any time soon.

Prompted awareness, when you're presented with a list of brands and asked a question to elicit your awareness of them, is much less valuable because if you see anything often enough on the high street you're going to remember it when prompted.

For me, "brand awareness" comes from news coverage, not ads. I know that ads are biased, so I ignore them.

LOL, news coverage of corporations (except for covering some public scandal) is heavily biased as well. In this particular case, the source of the article is Kristin Lemkau, the bank’s chief marketing officer. So keep trusting the "unbiased" news. And guess what: you might think you ignore ads. You do not.

OK, so I don't "ignore ads".

I actively avoid seeing them. I use ad-blockers, and have for over a decade. When I commonly read print media, I used to go through first and tear out all the ads, seeing them as little as possible, and pasting blank paper over the ones that had desired text on the other side.

When driving, I avoid looking at billboards, just as I avoid looking at oncoming lights at night.

I also don't trust news coverage. I especially don't trust favorable news coverage, because that's often advertorial. That does create bias I know. And if I'm actively interested in something, I read a variety of media, from different countries.

> When I commonly read print media, I used to go through first and tear out all the ads, seeing them as little as possible, and pasting blank paper over the ones that had desired text on the other side.

To each his or her own, I guess. Do you close your eyes when venturing outside as well? Also, as I am sure you are aware, much of what you read has been pitched and influenced by PR agencies: that’s what they do. When you read about a product in some publication, that product didn’t just randomly find itself as part of a story, a PR firm helped get it there. I might suggest that you might be a bit naïve about how advertising, media and PR work. You can’t escape it even if you think you are.

I know all that. I don't trust reviews, because >90% of them are advertorials. You can often detect that by searching for key phrases, and then finding them in press releases.

I don 't close my eyes outside, no. But I do trash ads, whenever I can. Back in the hardcore days, I carried stickers with moisture-catalyzed adhesive. Once those suckers cured, sand-blasting was the only option. Now I just carry black markers ;)

The irony here is that, it seems, ads are having a bigger impact on your life than if you weren’t taking these steps.

So does gravity.

Most people, probably including you, don't have conscious control over their awareness of things.

Sure. And I do make an effort to stay conscious of how my awareness is changing, what's been influencing it, what agendas may be involved, and so on. I don't take my consciousness as a given. It's just something else to be questioned.

You might be surprised at just how much news is little more than slightly rehashed press releases.

Yes, I know that. I didn't mean that sort of "news". I meant the results of insider leaks, prosecutions, fines, vulnerability discoveries, investigative journalism, and so on.

This is from March of 2017. I couldn't find any follow-up articles talking about results almost 2 years in.

What they conveniently leave out is what % of the total traffic on those 400K comes from the 5K remaining. It's a safe assumption that they stuck with the most popular mainstream sites so this could realistically be a <10% reduction in the number of eyeballs.

Yes, TFA says that impressions didn't decrease.

That should be the end of this "story" (oh yeah, Pareto/power distribution, nothing of interest here).

The existence of Patreon and similar services gives me hope that we can one day have a viable way to fund websites besides advertising. It seems like at present, one or two person teams (e.g. small podcasts and YouTube channels) can survive or even thrive on Patreon donations, mid-sized entities (e.g. major podcasts like Serial, most websites) rely on advertising, while large organizations like newspapers use some combination of advertising and subscriptions with mixed success. It would be great if Patreon-style donations could become a major revenue source of the mid-sized content producers as they have for the small.

Plenty of websites don’t rely on ads for funding: e-commerce, product listings, yellow pages, price comparison websites...

Flattr is a good system for smaller content producers, who in turn could pay large content distributors like Youtube for being listed there. No ads needed, but who would want to burst the web advertising bubble?

There’s actually a company doing automated “pay to remove Google ads and support the site” — http://subscriptions.publir.com — uses Stripe and proceeds go to the publisher.

Actually Google itself does it: https://contributor.google.com/v/beta

I'd like to use Google Contribute, but to work, it would require me to be logged into Google all the time, and I don't want to do that...

In addition to the obvious disadvantage of the Google login requirement, Google Contribute doesn't remove all ads –– just Google ads.

Is there any research on the effectiveness of ads? I can’t ever remember intentionally clicking an ad on a webpage. I would imagine ads are more relevant or effective on an image feed like Instagram, but does anyone interact with ads on actual webpages? Or, worse, mobile games?

I review some of the statistical/experimental papers in https://www.gwern.net/Ads . For example, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2367103 observes that given plausible effect sizes / variances, you can't run an advertising experiment which will pay for itself; https://www.kellogg.northwestern.edu/faculty/gordon_b/files/... demonstrates that even using absurd amounts of detailed tracking data from FB, you still can't recover actual causal effects.

I started shopping at American Apparel because I liked their banner ads back in the day. Same with this other apparel brand that used to be around in Richmond (Need Supply I think.) I’ve also bought things from Instagram and things advertised on podcasts, so maybe I’m the kind of person willing to buy things from ads?

In any case, the ads that appealed to me sold an aesthetic.

Older people and children feed the industry

Citation needed. The prime demographic for ads is 18-34. And people in those age groups are most definitely responding to ads even if they aren’t clicking them.

° And people in those age groups are most definitely responding to ads even if they aren’t clicking them.

Could you provide a source for me to read more?

We are approaching the situation where 51% of Internet traffic is fake.

Let's hope the outcomes bode better for society than blockchains.

I wonder if it's advertised-selective.

That is, a mainstream product like chase works with mainstream sites. While the fake news / conspiracy sites might be good for the kinds of things that end up in my spam folder: get rich quick, penis enlargement, tax scams and the like?

> An intern then manually clicked on each of those addresses to ensure that the websites were ones the company wanted to advertise on.

why is it always the interns

Because grunt work can be a valuable way of learning the fundamentals of a craft.

“Surprisingly, the company is seeing little change in the cost of impressions or the visibility of its ads on the internet”

This is pretty significant.

I bet you there's a power law in the distribution of effectiveness.

How much of the results are driven by 1 property (aka Facebook) alone?

Wonder if this says more about the effectiveness of ads or the dominant position Chase holds in the US financial system.

Given how a lot of New Yorkers feel about Chase, is this really anything to crow about?

Branded ads on the Google extended network are incredibly cheap.

Mods: please add [2017] to this article.

Online advertising is toxic. Everything from the execution (code), the push, the presentation, the stalking, the need for isolation, and the horrid insecurity. I understand that ads are the life blood of much of the web, but so long as that is the case there are many sites I won't visit and just about everything else I will block ads.

The only site I frequently visit where I don't feel dirty and violated by ads are the video ads on YouTube because I know they are screened for in appropriate content and aren't littered with 30 spyware packages. Of course YouTube makes up for that with horrid ad/stalking behavior everywhere else.

I even bet Facebook would be an ethical company if not for all the behaviors associated with chasing ad revenue.

Except for the folks paying content makers to do ads, as far as i can tell plenty of YouTube content makers have no problem with targeting kids with stuff that is effectively gambling.

That isn't the advertisements. That is the actual content people intentionally request. I wouldn't confuse a casino for its billboard.

>That is the actual content people intentionally request.

Maybe I'm misunderstanding you here, but I'm not really sure any viewers are requesting content makers do ads that target kids for gambling like apps and site.

Either way they are ads, and they are on YouTube. It's hardly a clean place advertising wise.

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact