Report: 40 percent of mobile ad clicks are fraud or accidents | Hacker News Reader

Report: 40 percent of mobile ad clicks are fraud or accidents

gigaom.com

19 pointsthedoctor10 comments

I had mistakenly activated the mobile ad option in my company's google ads account. We had about 5-10 calls a day coming from mobile ads, and 100% were accidental. In fact, we had to track down the source for all these calls for weeks, because it didn't make any sense to anyone (neither us or the callers).

After that I carefully digged through all of our google ads activity. I finally deactivated absolutely all automatic placement and authorized only the main search page and a few selected sites, because none of the placements made any sense. Most of the clicks came from video pirating site and the likes.

My impression now is that 80% of all web ads are slightly disguised scam, and completely useless unless you're selling stuff targeted to the most clueless, naive consumers (ringtones, phone games and the like).

Honest question. How does accidentally clicking a mobile ad lead to accidentally calling the vendor in question?

> How does accidentally clicking a mobile ad lead to accidentally calling the vendor in question?

    n("AFMA_PlaceCall", function(a) {
        var b = i;
        0 == a.indexOf("tel://") && (a = a.substring(6));
        0 == a.indexOf("tel:") && (a = a.substring(4));
        if ((Ma || Na) && !navigator.userAgent.match(/3_0/i)) b = confirm(a);
        b && Z("open", {
            a: "call",
            u: "tel:" + a 
        })  
    }, h);

http://media.admob.com/gmsg.js

More reason to switch exclusively to CPM. Impression fraud is not particularly easier or harder to detect than click fraud, so if a typical CTR is generously estimated at 3%, then switching to CPM would knock the fraud rate down to less than 1%. And the concept of "accidental" simply goes away.

I understand why CPC is favored; it gives an economic incentive to make ads more prominent. But surely there's a better way to handle that which doesn't also offer an economic incentive to cheat. After all, direct conversions are a tiny part of an ad's effectiveness; otherwise television and radio advertising wouldn't work for grocery stores and soft drinks. Getting the brand into people's heads ("message association" as marketers call it) is the main impact of ads, and that is better measured by CPM than CPC.

One golden rule of business is to commoditize your complements, to kill off any competition they might pose. CPM display ads are complementary to search ads, hence Google commoditized them by making AdSense primarily CPC.

It will be extremely hard to reverse the effects of this action.

That's an interesting argument, but I don't think it quite adds up. Complements don't compete, they complement. The reason you commoditize them (like releasing them as an open source project) is not to kill them off as competition, but to let them thrive cheaply so that they build up the product you are selling. For example, Google's AdWord/AdSense product is complemented by monetized blogs, so Google commoditized monetized blogs with Blogger. Many other free Google services like Gmail can be seen in a similar light.

In addition, CPM ads are a substitute for CPC ads, not a complement. So commoditizing CPM or CPC ads would be a very, very bad idea for Google.

Furthermore I don't think that making AdSense CPC can be said to commoditize CPM ads. Nor did it really commoditize CPC ads; AdSense has been close to monopoly status for a very long time, whereas the goal of commoditizing is to increase competition with no regard for controlling the market.

While I don't agree with a lot of what twitter has been doing lately, they seem like the only company out there that has a non-gimmicky way of delivering mobile ads. If they can actually tune their ad delivery to be relevant to the user--which I think they have a chance of doing better than most--they may actually pull off mobile advertising. If Google can retain the search result screen real estate they also have a chance here--but delivering at most one ad per search.

I think that expectations related to advertising revenue will undergo a big adjustment due to mobile. You simply don't have the same amount of screen real estate to display ads on mobile without resorting to gimmicks, and so the overall pie is shrinking rapidly.

Email marketing is the one type of marketing that I think still works just as well on mobile--each (opted-in) marketing offer in my email still gets 100% of the screen real estate... and my full attention for a few seconds.

I don't think this would surprise anyone who has seen how mobile ads are implemented, especially in games.

What would be more surprising is if the accidental rate isn't actually much higher than reported here.

I was just talking about mobile ads yesterday saying they feel like they're designed to trick me into clicking them.

This article confirms my suspicions :-/

it sounds too low mainly the accidentals, I thought it was more around 50-60% and another 20-30% fraud. All in all 70-90% BS clicks

Reply on news.ycombinator.com