Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

TV Advertising Effectiveness and Profitability

onlinelibrary.wiley.com

Post reply on HN
Print view View original

21–30 of 197 posts

Re: TV Advertising Effectiveness and Profitability

#21

Post by yardie »

paxys wrote: post #3

Advertising is at its core a prisoner's dilemma. If every competitor in a certain space puts in $100 in advertising, they can all expect $0 in returns. However if a single company put in nothing they would be in a worse place because their competitors' returns would automatically become higher.

I feel that there is different types of advertisement. Brand advertising isn't so much about acquiring new customers but about reassuring existing ones. Like I honestly feel car makers don't need to advertise on the web or TV, yet they do it anyway. I have the car I have and I'm not looking to buy another for 5-7 years, yet Toyota and Ford will advertise to me as if to say "I'm still here in case you change your mind." I can't imagine the ROI is particularly high, yet not doing would be worse since competitors would silence you out. At the same time M-B isn't advertising to sell you a car but to sell you the exciting lifestyle you're going to have in their car. You already bought or planned to buy their car they are only reassuring your decision.

Re: TV Advertising Effectiveness and Profitability

#22

Post by mdoms »

paxys wrote: post #3

Advertising is at its core a prisoner's dilemma. If every competitor in a certain space puts in $100 in advertising, they can all expect $0 in returns. However if a single company put in nothing they would be in a worse place because their competitors' returns would automatically become higher.

This only holds if advertising can't grow the market, which it absolutely can and does.

Re: TV Advertising Effectiveness and Profitability

#23

Post by fighterpilot »

vikramkr wrote: post #12
yosito wrote: post #9

Earlier quoted context omitted.

Your logic doesn't add up.

How exactly does their logic not add up? This is actually a specific example of a prisoners dilemma taught in some intro to game theory classes, with the classic example being the tobacco companies

Well it assumes a zero-sum game from the perspective of the competitors. The combined advertising effort of all competitors could grow the entire pie that they each share. Even though their percentage of the pie stays the same, the magnitude of the slice could be bigger, and that improvement in magnitude could outstrip the advertising cost, thus making it worthwhile.

Re: TV Advertising Effectiveness and Profitability

#24

Post by 015a »

ohazi wrote: post #11
Animats wrote: post #2

TV ads. Where you can't measure response rate directly.

Also where "modern" "smart" platforms decide it's acceptable to chop up a video into a dozen slices and insert the same. exact. ad. in between each segment, causing half the viewers to swear off that brand for life?

Hulu is one of the worst at this, and it's astounding to me how seemingly willfully dumb advertisers can be.

It makes me think that no one in this industry actually cares. You've got marketing and advertising executives commanding eight figure budgets, inventing whatever metrics they need to showcase success and protect their jobs. They pay YouTube, Hulu, Samsung, etc millions, who provide back whatever engineered metrics they need to keep getting those paychecks.

You'd think there'd at least be some evidence in mapping advertising spend back to revenue, but I suspect that the platforms who can actually do this (e.g. Etsy) don't last very long or see worse advertising revenue, because it becomes startlingly obvious how poorly advertising works (especially when automated at scale).

It's a house of cards, with so many people deeply invested, spending their entire careers justifying their position, that it probably wont ever change.

Re: TV Advertising Effectiveness and Profitability

#26

Post by singhrac »

evancox100 wrote: post #20
lupire wrote: post #14

Title is critically incorrect. It should be: > The ROI analysis shows negative ROIs at the margin for more than 80% of brands, implying over-investment in advertising by most firms.

You are correct, but if you look at total instead of marginal the picture's not much better. While 80% of brands have negative marginal ROI, 67% (2/3) have negative net ROI, and so would presumably be better off with zero advertising.

Not necessarily. The thing to measure is the marginal ROI at zero advertising spend. Derivatives are not bounded by their average value across a range.

Re: TV Advertising Effectiveness and Profitability

#27

Post by hammock »

There is a fundamental misunderstanding (intentional or not) with these types of studies and claims.

Limited-time promotions, e.g. a 10% off sale, work on short timescales that are easily measured. You can easily say what the ROI is of an advertised promotion, and it is often positive. (Side note, these promotions have a more-difficult-to-measure detrimental effect on your long-term profitability, closely related to the point I am about to make below)

Brand advertising, in contrast, works on a spread-out scale of years or decades. When Coca-Cola runs a polar bear ad at Christmastime they don't do it to increase sales of Coke that week, nor should it be measured that way.

The actual effect that a brand advertisement has, is to add PENNIES (not dollars) to their sales every day, for the next 100 YEARS. It's a long-term investment. And for these reasons hard if not impossible to measure or control for confounders - but that doesn't mean it doesn't work!!

And without that brand investment, their brand value is continuously eroding away, at a rate of pennies (or more) per day.

Source: Been on the leading edge of marketing and advertising for over 12 years

Re: TV Advertising Effectiveness and Profitability

#28

Post by sfblah »

Animats wrote: post #2

TV ads. Where you can't measure response rate directly.

I work for a company involved in search engine advertising.

While it's technically possible to track conversions, very very few customers actually do it. Some don't have a great way to track it (health-care products where there's a privacy issue, brand advertising, informational advertising, services where there's no obvious immediate call to action, etc.)

Even among those who do track "conversions," I'd say under 10% of them actually define a conversion as something where money is exchanged immediately, which makes it hard to defraud. More than half of the conversion tracking we see just requires the user to spend a certain amount of time on the website, visit a certain number of pages or to provide an email address.

Not surprisingly, we see a massive dropoff in conversion percentage when money needs to change hands. It's more than 90%. Also, not surprisingly, when some action is required for conversion, we see more than half of the visitors taking that exact action and nothing more.

Our analysis of the above is that more than half of clickers on search ads, and probably more like 80%+ are just bots (or human "bots" in clickfarms) clicking the ads and then pretending to convert. Their incentive is that they are clicking ads on search pages where they get a share of the revenue.

Interestingly, when we alert customers to this dynamic and try to get them to shift their ad spend or conversion tracking to prevent this kind of behavior, they actively do not care and prevent us from doing it. Our analysis there is that marketing managers in general know that what they're doing is largely ineffective, but they don't want to admit that to their bosses because then it puts their jobs at risk.

So, what we do is help them maximize where we can, without pushing them too hard on the above: "Humankind cannot bear very much reality."

Re: TV Advertising Effectiveness and Profitability

#30

Post by mlyle »

singhrac wrote: post #26
evancox100 wrote: post #20

Earlier quoted context omitted.

You are correct, but if you look at total instead of marginal the picture's not much better. While 80% of brands have negative marginal ROI, 67% (2/3) have negative net ROI, and so would presumably be better off with zero advertising.

Not necessarily. The thing to measure is the marginal ROI at zero advertising spend. Derivatives are not bounded by their average value across a range.

The claim was: "67% (2/3) have negative net ROI, and so would presumably be better off with zero advertising."

If they truly have negative net ROI, they would be better off with zero spending than their actual spending. The claim holds.

> The thing to measure is the marginal ROI at zero advertising spend. Derivatives are not bounded by their average value across a range.

OK, so if we're going to get super pedantic... the derivative could be negative at the current value, and at zero, and there could be a positive ROI for some value of spending (between 0 and the amount spent, or even for some value more than what was spent). But it's not really likely, nor relevant to the point the person above made.

Post reply on HN