What do we really know about the effectiveness of digital advertising? • Hacker News Board

Re: What do we really know about the effectiveness of digital advertising?

#361

Post by pushrax » Sat, Nov 09, 2019, 3:24 AM UTC

lsc wrote: post #292

pushrax wrote: post #254

Earlier quoted context omitted.

i.e. you want to see the facets you'd be looking for when choosing a product to purchase. For many products and people that will just be the visual appeal, and advertisers target their widest audience. This is a case where tracking can actually be seen in a positive light - I too think I would be more likely to convert if I saw a datasheet as an ad, but that's definitely not majority. At the same time, obviously fash…

>For many products and people that will just be the visual appeal, and advertisers target their widest audience. I... think this is usually unwise. If you attempt to appeal to everyone, you appeal to no one. I personally think one of my advantages when I ran prgmr.com was a consistent brand? At the very least, it was very recognizable. (I don't run it anymore, and I'm not claiming to have done a great job when I did.…

Attempting to appeal to everyone by serving a different ad based on tracking is precisely what I meant, but I'm thinking more from a platform standpoint than as a brand. An ideal marketplace would prefer to sell a spot to brands who advertise in a way that appeals to the buyer.

Re: What do we really know about the effectiveness of digital advertising?

#362

Post by graycat » Sat, Nov 09, 2019, 4:19 AM UTC

vbsd wrote: post #360

graycat wrote: post #172

Earlier quoted context omitted.

> there are plenty of books I would be interested in which I don't know exist ... > The problem there is I'm generally only going to be interested in long tail items, ... Yup. For that problem, did some original applied math based on some advanced pure math prerequisites and wrote the software for the math and for the corresponding Web site -- software appears to run as intended. Need to add some data and do an alpha…

I’ve first read posts about your website many months ago, and I’ve been intrigued ever since. Could you share a planned release date, even if approximate and preliminary? Also, it’s seems that your product has been “almost ready” for some time now, just out of curiosity, what are the main blockers?

I got interrupted by unpredictable exogenous events. Somehow when I get the last one handled, some more come along.

None of these events have anything to do with the project. The project is in good shape, and the work on the project that is uniquely mine has all been fast, fun, and easy.

I had a motherboard fail. I moved. And many more.

I better understand the advice to entrepreneurs -- "Never quit". Well, I don't want to quit, and I don't have to. Actually now my bank balance is growing nicely.

Most recently I've been setting up my new office -- it was a lot of work but looks good, efficient, functional, etc.

I'll get the nonsense out of the way and get back to the work.

The work to do now is reasonable. What has not been reasonable are the unpredictable exogenous interruptions.

Thanks for your interest.

Re: What do we really know about the effectiveness of digital advertising?

#363

Post by vivekd » Sat, Nov 09, 2019, 2:33 PM UTC

I do not work for google facebook or any other internet company. I do however purchase online ads for my law firm. I also tried turning off the ads for a few weeks to disastrous results.

I can say the big problem in the examples in this article is the keyword selection. For example ebay advertising the word ebay to customers who are googling the word ebay is probably fruitless for the most part. Those people were already looking for ebay and would find it with or without the ad.

However if you are a lesser known online auction and you use the keyword "internet auction" or something similar, it would probably help.

This article, to me, seems to be more about the importance of experimenting and critically evaluating the data rather than an actual indictment on online marketing. With online marketing there is no clear answer: what keyword you bid on, how you target, how you write your ad copies, how your competitors are behaving all come into play. It's difficult for anyone to parse through those factor and really come out definitively as to whether online marketing is effective or ineffective in general. We can only do the smaller measurements and see if our particular ads are effective for our particular business, which in the end is all any of us who buy ads really care about.

Re: What do we really know about the effectiveness of digital advertising?

#364

Post by Marsymars » Sun, Nov 10, 2019, 12:41 PM UTC

tempestn wrote: post #123

leoedin wrote: post #7

I've always wondered about those adverts for the website I'm searching for. If I type its name into Google, I'm already 100% committed to going there. Sometimes I click the advert (it generally depends how much I like the company). The other aspect of online ads that strikes me as ingenious but probably not sustainable is the way Google inflates the auction system with fake money. Anyone who's run a website will have…

I had the same thought. Our PPC consultants strongly recommended that we do branded ads and I pushed back against it using this same logic. Ultimately we performed a test, but the results were very different from in the article—we did find that a significant percentage of those who clicked our branded ads would otherwise click the organic results... but far from 100%. And because cost per click on branded ads is so l…

If I had a brand, I'd be tempted to bid it up with ads directing to ad-blocking extensions. Does Google allow that?

Re: What do we really know about the effectiveness of digital advertising?

#365

Post by niknikson » Wed, Nov 13, 2019, 4:27 AM UTC

onion2k wrote: post #316

josephjrobison wrote: post #170

A very easy way to break this argument is to not look at huge known companies like eBay and Amazon, but brand new companies 1-2 years old that have grown almost entirely by online ads. Eventually word of mouth kicks in, but there are plenty of smart operators that built their entire brand and revenue off of ads. Works best in online direct to consumer brands.

...there are plenty of smart operators that built their entire brand and revenue off of ads. Can you name one? I find it very hard to believe there are any companies that have scaled up only using online adverts. Why would the founders of a business do that when they can grow in their local market by talking to people as well as buying online ads for wider reach?

Wayfair - you can listen to the NPR podcast "How I built this" where they talk about building their entire business around domain names and pay-per-click advertising (7 billion market cap).

Re: What do we really know about the effectiveness of digital advertising?

#366

Post by JetSpiegel » Thu, Nov 14, 2019, 11:43 PM UTC

jefftk wrote: post #336

belorn wrote: post #317

Earlier quoted context omitted.

The free internet can be split into two groups, websites operated for non-commercial reasons and websites operated for commercial reasons. The group that operate under commercial reasons can also be split into two groups, those whose primary revenue is from advertisement and those whose primary revenue is from non-advertisement sources. Hacker news for example belong to the later and gain revenue from investing. To a…

HN has ads: startups advertising for employees. But I don't know how YC thinks of the effect of HN.

It has ads, but those are not commercial. Only indirectly.

Re: What do we really know about the effectiveness of digital advertising?

#367

Post by jefftk » Fri, Nov 15, 2019, 1:45 AM UTC

JetSpiegel wrote: post #366

jefftk wrote: post #336

Earlier quoted context omitted.

HN has ads: startups advertising for employees. But I don't know how YC thinks of the effect of HN.

It has ads, but those are not commercial. Only indirectly.

What do you mean that the ads aren't commercial?

Re: What do we really know about the effectiveness of digital advertising?

#368

Post by JetSpiegel » Fri, Nov 15, 2019, 10:32 AM UTC

jefftk wrote: post #367

JetSpiegel wrote: post #366

Earlier quoted context omitted.

It has ads, but those are not commercial. Only indirectly.

What do you mean that the ads aren't commercial?

The companies don't pay for the ads (that's my understanding). It's not dissimilar from an artist mentioning a patron. It's more an endorsement than an ad, it's not exactly the same thing as AdWords.

Re: What do we really know about the effectiveness of digital advertising?

#369

Post by Breza » Fri, Nov 15, 2019, 1:26 PM UTC

FussyZeus wrote: post #241

seem_2211 wrote: post #232

Earlier quoted context omitted.

There's a difference between performance and brand marketing. Bart advertising by Twitter/ Facebook is mostly brand marketing (not really tied to anything). A lot of marketing is about reminding you that a company exists.

Despite concentrated effort, I have yet to be able to forget that Facebook exists. Brand awareness marketing for companies as ubiquitous as that seems like lighting money on fire for the sake of it.

What about Coca-Cola? They spend billions on advertising even though everybody already knows who they are. They want to seem current. No brand wants to be thought of as yesterday's news. Facebook advertises to remind people that they exist and to seem like they're just as hip as Tiktok.

Re: What do we really know about the effectiveness of digital advertising?

#370

Post by Breza » Fri, Nov 15, 2019, 1:27 PM UTC

seem_2211 wrote: post #232

komali2 wrote: post #198

Earlier quoted context omitted.

There's twitter and facebook ads in the san francisco bart system and I just don't fucking understand who in SF isn't aware of twitter and facebook or what they're about.

There's a difference between performance and brand marketing. Bart advertising by Twitter/ Facebook is mostly brand marketing (not really tied to anything). A lot of marketing is about reminding you that a company exists.

This is an important distinction that the article doesn't cover. The article focuses on a specific subset of advertising. Brand awareness ads are even harder to measure.