P&G Cuts More Than $100M in ‘Largely Ineffective’ Digital Ads

The only reason they have my email address was because I made an account so I could claim my house on Zillow, so they know exactly where I live, and that is 850 miles from Palo Alto.

digital ads are not always easy to track. a lot of malware is dedicated to filling your cookies with visits so that if you happen to buy something, they can claim you saw the ad prior and take credit. in many cases, a digital add campaign creates such a tiny blip in sales its hard to differentiate from randomness, were as a TV ad shows floods of sales.

> The best thing for ad industry is if traditional media dies and goes away and then they could focus their attention on "digital" rather than traditional/print.

traditional advertising, namely TV, is still the most successful way to increase sales for many companies. since digital ads leads to so few sales, even the platforms have often moved away from 'advertise to produce sales' to advertise for brand management.

That's the problem I've seen in the digital ad space -- lift is either hard to measure, or ignored completely. Causality to the sale is assumed on the ad.

All the dev tool companies run ads! CircleCI runs them, but the dozen or so other dev tool companies where I know the founders or early hires also run ads. You would think given comments like this on HN that they would be useless, but we all run the numbers. The numbers say that not only are they effective, but that usually we should run more of them.

It's like when people here say "oh I hate talking to salespeople". Yet, salespeople make the world go around.

On the other hand, there is a not-insignificant number of bot networks and shady clickfarm sites out there. Though, you can usually avoid being taken in by them by not being a cheapskate - quality traffic is more expensive. If you're bidding <$1 CPM for most things, enjoy funding a click farm.

That's the problem, though. You can tell me that the user clicked on the ad and then bought my product; but was the user going to scroll down to my organic search result, click on that and buy anyhow? The advertising sellers claim that they should be credited with the sale, when all they are providing is moving the last link up; at least in the industry I was in, a user, generally speaking, isn't going to sign up with a VPS provider they haven't heard of. Sure, sure, putting a special in front of them can move you from second or third place to first place, and there's value in that, but really some brand advertising needs to be done before you can sell at all, and as you pointed out, measuring the effectiveness of brand advertising is super hard.

I suppose there are some products where the user really is going to buy the product first linked; but I think there are a lot of products where the user doesn't make a buying decision for some time, and the last click... may not have a lot to do with what order the links are in.

I think this is the real problem with online advertising; tracking that last click into a purchase is fairly easy (though, as you point out, some people don't even do that) but figuring out how much of that purchase was because the link was first, and how much of that purchase was because the user had previous knowledge of you? that's hard - I understand that a lot of research is being expended in this direction, and that this is the value proposition that Facebook claims, but... it's a hard problem, and depending on the product, you could reasonably argue that the state of the art isn't as good as the people selling advertising say it is.

I spent a few hours coding up a script that would launch a VM and create a new account with a catch all email, then confirm the email.

Worked well also.

Your approach was much less work, though. ^^

If you're a consumer lifestyle or luxury brand, most of your sales are in-store and hence attribution due to influencers becomes incredibly difficult.

What many companies I've seen do is attempt to measure organic reach through a variety of internal or external scoring mechanisms, and using engagement as a proxy for reach. So a beauty influencer posting a screenshot of her travel bag with your brand tagged would have comments, likes, reshares/regrams measured, and then have that scored as some sort of an effectiveness index.

For example, we did multiple campaigns for major sports teams. How are they supposed to measure the impact on sales of tickets and memorabilia? The idea isn't to generate sales of product X right now, it's to energize and grow your team's fan base, which will hopefully lead to more people buying more things over time.

Another example: we did many campaigns for food and beverage brands. People don't tend to order those things right now, online. You're just trying to grow your brand in hopes that more people will buy that snack or drink the next time they're at the store.

So I amend my previous statement to agree with you that semi-precise metrics are possible if you're trying to get someone to buy a specific product online right now. That description covers a lot of companies, and they probably have good metrics.

But a big consumer-goods company like P&G probably isn't doing that kind of marketing. They're trying to get you to think of brand X the next time you're getting groceries. That makes impact of ads devilishly hard to measure, regardless of whether the ads are online or on TV.

The agencies that buy the ad verify it was placed in the right spot at the right size and that's about all they can do.

Why should the web be any different? As long as they insist on bundling javascript with their ad, I'm going to run an ad-blocker.

I disagree - for something at the scale of P&G, you can bet that they're doing a LOT of measuring for every campaign they run.

The article reads to me as if they're applying the same metrics to their online ads that they use already for traditional campaigns, and finding little impact.

Online ads promise perfect monitoring, but that is not only provided mostly by the advertisement channel provider itself (quite the opposite of independent) but also only useful for ads that aim for immediate reaction ("buy now!"), it does not tell anything about brand awareness effects.

I don't think that you are disagreeing with GP at all.

So maybe there's a better way.

Make things that work, sell your vision not the product. People will buy into your vision, and accept more product shortcomings and cheer you on as you improve.

Most newspapers could make old-style ad deals, like they do for print. A few newspapers also still finance their sites almost entirely through subscriptions, and those would continue to exist (e.g. Süddeutsche, SPIEGEL). As would obviously publicly financed sites (BBC, ARD, ZDF, NPR).

Also products financed via paying customers, such as imgur used to be, or possibly reddit if they change the way reddit gold works.

Also, most people's blogs would continue to exist, as many self-host or use services where the free users are subsidized by the paying ones. Even WhatsApp used to be profitable with that model.

We'd be going back to a 2010 era web, and I'm not so sure that's actually bad. We'd lose the corporate dominance over the web, Google, Facebook, etc.

Now, there is one part where paying user subsidizing doesn't work, but where ads are currently the only model of financing, and that's search engines.

But considering those act as a public service, maybe after advertisers leave the web and Google goes down crashing, a government could just buy Google (as the German government bought DHL, or major shares in VW, they could probably just outright buy Google).

Many sites will perish, and little of value will be lost.

And I'm ready to make a several-hundred-dollar purchase tomorrow and I see ZERO ads around these products. I see plenty of crap.

In my view, the founding hypothesis of the company- that people will be amenable to ads connected to what they are searching for- is incorrect. So we're seeing a pretty clumsy pivot.

I keep hoping Duck Duck Go takes it elsewhere. It could be very easy, maybe as simple as a little button I click that says "I am looking to purchase" and now let the advertisers make their pitches.

The point is, I would like to see if they got better results by letting me drive.

Paper media can sell a paper copy and (theoretically) run without ads. Digital media would have to sell subscription (when they can offer high-quality content), or run entirely on ads (then the content is secondary). Guess which type is prevalent.

I like the guarantee. That's why hacking in the conventional sense today-- which needs constant awareness, involvement, adaptation, experimentation to find out what works, non-codified intelligence, and dealing with the reality of a system -- can beat the process du jour. It might become the new canonical process if everyone else starts doing it.

Sometimes I feel like long-term or well-thought-out strategy is the hack in a sea of self-proclaimed hackers who focus on useless wins and optimizations.

What online ads allow you to target is "men between 28-32 from eastern Europe who like Formula 1 and Classical Music"

Previously, on HN... https://news.ycombinator.com/item?id=14002995

Noiszy is a browser plugin that creates meaningless web data - digital "noise."

It visits and navigates around websites, from within your browser, leaving misleading digital footprints around the internet. Noiszy only visits a list of sites that you approve, and only works when you turn it on. Run Noiszy in the background while you're working, or start Noiszy when you're not using your browser, and it sends meaningless data to these sites for as long as you let it run.

This meaningless data dilutes the significance of your "real" data, by creating a campaign of misinformation. You become more difficult for an algorithm to understand, market to, or manipulate. You can outsmart the "filter bubble".

Hyperbole much? That statement is just rubbish.

Personally I think it would be the most constructive thing to happen to the Internet. Hell, maybe even society in general. Advertisers are scum, they ruin everything they touch, and any 'content' producers whose businesses died if this was to occur would certainly not be missed by me.

(Also, I hate the phrase "content producer". It seems to imply that content is fungible - companies thinking in such terms are precisely those who should go under.)

It's simply detrimental to the overall ability for us to think.