April 2018 AdSense Earnings & Observations

[Ironside]

---

System: The following 18 messages were cut out of thread at: https://www.webmasterworld.com/google_adsense/4889486.htm [webmasterworld.com] by martinibuster - 6:42 pm on Apr 1, 2018 (utc -5)

---

I am up to 14 clicks with a 12p CPC so far today. But it's still early afternoon in the US so hopefully, things will get slightly better. At least marchers earnings were a lot better than January's.

@May Is it possible to work out how many clicks were disabled.

[Travis]

ha, ha :)

[papux]

Are Stats delayed?

[nomis5]

You (everybody) shouldn't look at your Adsense stats all day long. This is not going to change anything, plus, stats might be delayed, or inaccurate. Better use your time to work at your site, and for a walk.

Checking my Adsense earnings (I just timed it) takes me 25 seconds. So why not check as often as you want? I'm not going to even get my shoes on to go for a walk in 25 seconds!

[NickMNS]

@nomis

So why not check as often as you want?

Because the number you see (intra-day) may be misleading.

[Ironside]

Like it could be another decent day. At the moment I have a page RPM of £3.74 with the CPC at 26p, although I've only had nine clicks so far. But it still morning on the East Coast of the US so I still have the rest of the day to earn more.

@nomis5 yeah, easily said than done. Sometimes it can be a bit addictive checking your stats to see how much you are earning :-)

[frankleeceo]

Checking stats is almost like checking stock market quotes to me :), highly addictive and volatile haha. I have been day trading stocks and doing decent amidst market volatility.

It's good to check it frequently for a day or two in my opinion, that way one can get a good feel on how budget is spent throughout the day. I have a rough idea on how much I should have at any given time in the day, so I don't check as often. And if something is wrong, I would personally know right away.

I am seeing RPM's trending at this month's high by almost 70% higher compared to beginning of month.

[KaseyM]

The only thing dropping faster than my RPM is the bombs on Syria. (sorry)

[MercuryHero]

I check often, I get that it can be quite addictive.

I understand that revenue doesn't come in smoothly throughout the day and stats have delays. It comes in bunches, with many more coming in between 10pm and midnight than any other time (my local time, which is not Adsense billing time). Typically I have a bunch of non-US clicks in the morning, then the US clicks tend to bunch up in the evening. But it's not hard and fast.

The unreliability of intra-day stats still doesn't stop me checking it often.

[Travis]

And don't forget that each time you check your stats, Adsense deducts a 5 cts access fee from your earnings, for the bandwidth and resources you are consuming*. This is why, the more you check, the more you see one value one time, and a lower value the next...


* old internet legend :-)

[Runfun]

Checking several times a day is sometimes interesting. Like the monthly clawback that is actually excluded the daily clawback. A few years ago there wasn't a reducting of revenue during the day but I notice for about 10 to 20 percent clawback during the day and probably still missing few euro's. For example when I notice at the end of the morning €25 and an hour later it's €22. Same with late in the afternoon and begin of the evening. So the monthly clawback is much more than most of us think.

[NickMNS]

@Runfun how is knowing that there was roll-back beneficial. All it really does is make you feel like they took something from you.Which isn't the case, it is actually a correction in reporting. The same holds when you check your stats early in the day and you see a high RPM, then over course of the day you see that value drop. Not a fun feeling. But this is normal, as it is simply the result of a reporting bias. The issue is not that numbers are wrong per-se, but psychological impact of the perception of loosing something.

To Franklee's point, this is very much the same thing as checking your investment portfolio. When thing are going up, you are feeling great. When markets turns, your stress level goes up and mood goes down. There are number studies that document this and also show that making decision in such an environment is not optimal.

That being said, there is no doubt that one needs to watch one's sites performance such that one can address any issues to minimize any impact. But to make decision with biased information is not going to help. (intra-day stats are always biased). So there needs to be a balance between too much checking and not enough.

Now for my big admission. I typically have an Adsense window and an Analytics window open all day.

[tryingtogrowsite]

Are three ad blocks too many? Would it make sense to remove the last unit and replace it with a large Media.net unit?

Does the Adbalance really make a difference? Are we leaving money on the table by lowering it to 80-85% (while supposedly maintaining 99% revenue)?

I do like the fact that for a good number of the impressions the ads collapse? Has anyone found they've increased page views by creating links to other pages to show when Adsense has nothing relevant?

Has anyone successfully asked for a donation on their sites?

[MayankParmar]

Has anyone successfully asked for a donation on their sites?

$5 donations or $5 per month subscription sounds bogus to me.

[NickMNS]

Are three ad blocks too many?

That depends, is there sufficient content on the page to justify three ad-units? Are all the ad-units and specifically the last ad-unit being seen by users?

Would it make sense to remove the last unit and replace it with a large Media.net unit?

In my opinion no. If you want to show ads from two networks it is best to use DFP such that the networks can compete for the placements, otherwise you are certainly "leaving money on the table"

Does the Adbalance really make a difference?

In what respect? I use to help block unwanted and spammy ads, in that respect it is the most effective tool available. I also use it to promote other projects, so again yes effective. Does it come at a cost? Likely, but it is not measurable. You can setup an experiment and test it.

[Mentat]

Another Monday, but the payment is still not picking up :(

[MayankParmar]

Still bad. 16th April it is :(

[Maximum44]

Had a great start the 16th of April, but it died out later in the afternoon. Traffic was stable all day. People are probably focusing on filing their taxes instead of clicking on ads :D

[MayankParmar]

Does Sucuri affects AdSense earnings?

[csdude55]

Terrible month so far, but I guess I'm not shocked with it being tax season (small businesses aren't advertising), Trump's trade wars (big businesses aren't advertising), and the Zuckerberg Trials (tech businesses aren't advertising).

Comparing 4/1-4/16 to 3/1-3/16, traffic is down by 1% but earnings are down by 24.8%. 4/15 was the only "near normal" day, but then 4/16 was down again.

I finally got Auto Ads to work, but they're hardly worth the effort... generating about 4% of my overall revenue. I mean, I guess that's still better than nothing, but hardly worth the effort it took to implement them.

I'm hoping that after Tax Day things will improve, but I'm not really counting on it.

[Maximum44]

@MayankParmar Everything affects Adsense earnings :D

[Runfun]

NickMNS, those during the day clawbacks are introduced a few years ago. So when people are happy with a low clawback at the end of the month but don't mention the daily clawback it's not the complete figure. A drop in CPM/RPM during the day is logical, the best paying advertisements are showed the first. During the day those ads are through their impressions or clicks and will be alternated by less paying ads.

Does anyone see more blanks than regular? Look at this topic: [webmasterworld.com...]

[MayankParmar]

Yes, Runfun. Seeing a lot of blank ads. I thought something is wrong with my PC and internet.

[Travis]

What about backup ads ? are they showing, or everything is blank ?

[NickMNS]

A drop in CPM/RPM during the day is logical, the best paying advertisements are showed the first.

The main reason for high RPM early in the day is due to a bias of how the data is collected/reported. RPM(page RPM) = earnings / page-views * 1000. Say you get a click on the first page view of the day at midnight plus 1 second and the CPC is $1.00 your RPM will be $1 / 1 * 1000 == $1000 RPM. Wow! that is awesome. After the second page view $1/2 * 10000 = $500 RPM, less than before but still Wow!, by 8am you have 1000 page views now your RPM 1/1000*1000 == $1, not so wow.

Now lets change the scenario. Midnight + 1 second, 1 PV, no clicks RPM = 0, midnight + 2 seconds -> 2PVs no clicks RPM = 0, >>> 8am 1000 PV, 1 click 1$ RPM $1 not so wow. If in scenario 1 you would check you PV shortly after midnight you would go to bed thinking you would wake up a wealthy person but instead you wake up to check your stats feeling like you were robbed. In scenario 2 you go to bed feeling like your going to need to work another day, and wake up to find confirmation of that. The two scenarios are economically equivalent, the difference is the bias caused by how the data is collected and reported.

Once you understand this you clearly see that there a no market forces causing better paying ads to be shown first.

There is a theory that suggests the opposite (along with plenty of anecdotal evidence). The idea is that if AdWords were to consume budgets early in the day (or what ever the budget horizon is) high prices could be achieved, eg: advertiser has a $10 budget for the day, 1 click at 8am for $10, budget is filled and there are no more clicks for the day. But then in the afternoon another advertiser appears in the market with a budget of $20, but now since all the big bidders have spent their money, demand is now low and high prices can no longer be achieved. So the afternoon advertiser ends the days with $10 unspent of its budget. So Adwords needs to ensure a steady level of demand during the day as this will ensure that budget are fully consumed. Thus the result should be that prices go up towards the end of the budget horizon. This is why anecdotally there appears to be a bias towards higher earning at month end, and quarter end. I have my doubts whether this theory really holds in practice, but that is the theory...

[Mentat]

Still, there is no real explanation why my revenue is 40% lower this month vs March.

[MayankParmar]

April 2017 was much better! And AdSense support always says it is a seasonal change lol

[Runfun]

Once you understand this you clearly see that there a no market forces causing better paying ads to be shown first.

The whole point of RTB or header bidding is the best paying ad will be shown first. Those ads are also likely to be clicked the most at the beginning at the day.

[NickMNS]

The whole point of RTB or header bidding is the best paying ad will be shown first.

Not really. If one use a typical cascade and sets ceiling (or floors) for each ad-network what can happen is that the ceiling will cause bids above the ceiling level in one network to be ignored (as the ceiling is reached) and the bidding switches to next network in the waterfall. But if there is low demand in that next network the price of the bid will close at the ceiling of the last level despite there being higher bids available. Header bidding prevents this from happening by allowing bids in other networks compete therefore maximizing the price paid.

So header bidding ensures that you get the best price for ads. But, "the whole point" isn't to get the highest paying ads to be display earlier in the day. The point is to allow as many ad networks to participate in the auction together on the same level, thus making the auction more efficient for all parties.

Whether the publishers uses or does not use header bidding does not directly influence the consumption of the advertisers daily budget, that is typically managed by the ad-network. For an optimal outcome for the ad-network (ie more profit) it is best not to consume an advertisers budget in a short period of time. But obviously, depending on actual market condition this may still occur.

[balex25]

I only one what see about 35-45% drop in RPM?

The drop starts on 8th April 2018 and is still down.

[Cloudbaseflyer]

Yes, everybody else seeing over 50% drop. Lucky man! ;-)