Forcing In-feed, In-article, and Display ads to compete

[csdude55]

I've spent the last week or so running A/B tests on my different ad locations, to see if a display banner pays more or less than an in-feed or in-article banner.

So far the testing has been surprising. In some locations, the display consistently pays more, and in other locations the in-feed or in-article pays more.

My question, though... is there a way to set the ad spot to compete all 3 types against one another, and always only show the one that pays the most? Since there's no clear winner for all spots, I worry that the test could have had completely different results next week.

[Dimitri]

auto-ads ... but you no longer control the placement, so this is not the solution (in my opinion).

So, I would say that it's up to you to figure which type of ad performs better on which spot.

Also, keep in mind that, Adsense is mixing a bit everything more and more. In "display ads", you can have "text ads", "display ads", or even "link units" (I saw it myself). So, I suspect that, from time to time, Adsense may serve in-feed like ads within "display ads".

In a near future, I bet there will be no more difference, you'll have one ad code, and Adsense will serve whatever it thinks is best.

[csdude55]

Yeah, I tried auto-ads for a short period but they were a mess :-(

I really thought that Adsense would have them bidding against one another, but I had a responsive banner at the top of low-res monitors that made a few dollars a day. I changed it to an in-feed at midnight, and so far as of 4:30pm today it's made about 5 times more than it made all day yesterday! I don't know if that trend will continue over time, though, or if it's just getting more attention because it's new.

In an A/B test comparing a responsive banner that's at the bottom of the screen to in-feed and in-article, though, the display banner made about 50% more than in-feed or in-article. So there's no clear winner for all spots, I have to test and test and test and test.

Since we're talking about testing, I should note that in-feed and in-article seem to make about the same amount in any location, so I haven't found any reason to use one over the other.

I suspect that in-feed and in-article are going to grow a lot more. They're obviously easier for businesses to implement, they seem to load faster, they're fluid to whatever size the publisher wants, and right now they pay more. It's a win-win for everybody.

[Dimitri]

I see that, when you create an in-feed bloc, you have an option (checked by default), to allow display ads to appear too. So I guess this is what you want.

[csdude55]

I did that, but I'm still seeing a pretty big difference in RPM between them.

But I just discovered about 2 hours ago that my test is messed up, I'm gonna have to back up and start over :'-( I had set a CSS maximum height on the responsive, and didn't realize that the in-feed and in-article banners were going to overflow that height as if they had absolute positioning. So the banners were covering about 25-30px of my content, causing an abnormally high number of clicks from people trying to click a link that was half hidden by the banner.

So while the stats for today were quite lovely (reminding me of 2014 <3 ), I think they were skewed and I'll be lucky if Google doesn't take them all back.

I'll have to post back in a few days with an update on the new test results.

[Dimitri]

When you create an in-feed / in-article banners, you have a preview, and there is a handle to modify the width of the preview, so it gives you an idea of how tall the banner can be.

I never really paid attention to in-feed, but there are interesting settings! Especially for those who are mourning the progressive fading of text ads only.

I wonder if, advertisers have to opt in to appear in in-feed banners. In other words, I wonder if there is the same inventory than for display/text ads.

[csdude55]

I promised to update in a few days, so here it is. I'm afraid that I have to retract my earlier statement... after fixing the CSS on the ads, the average RPM on in-feed and in-article ended up being a tad lower than the RPM on display banners. So they weren't the panacea that I originally thought :'-(

Changing the display banners to responsive, though, seems to be bringing in a higher RPM, so maybe they load in-feed, too? I actually have a fixed width and height on a DIV surrounding the banners so I didn't think there would be any difference, but an A/B test going for the last 2 days has consistently had a higher RPM on the responsive. Not a HUGE amount, mind you, we're talking about $0.07. But it adds up! So I'll keep this testing going for 5 more days and post back with the final results.

[Dimitri]

Responsive ads "might", cover a larger inventory, because , they "might" display sizes , you didn't think about.

For example, where you are thinking of a 728 x 90 banner, the responsive ads "might" also serve 970 x 90 and 970 x 250 .

Now, I don't know which kind of variations you are observing from your A/B testing. From my experience, Adsense earnings can drastically vary for "no reason". This is why I stopped doing A/B testing , because results no longer made sense (in my case). So I am just doing as I "feel" it looks good, which is of-course totally subjective.

As for in-feed ads, I noticed that , you can remove the description! So, just a title and an Image. I wonder if this is producing better results, because it might be a better teasing ... titles sounding more appealing than description ...

[tangor]

Are these A/B tests being run long enough to get meaningful numbers? Through all hours/days of operation? During normal traffic periods, not some "slow time" or "abnormal time"?

I'd want *at least* three weeks run per test, if not three months!

[csdude55]

I'm desperate for money, honestly... ad revenue sunk like a rock about 2 months ago, so I'm trying everything I can to build it back up. I don't have 3 months to wait, I need it up soon or I'm gonna run out of credit cards!

So I've been running my tests for 7 days, midnight to midnight. It's not a perfect test, though... in PHP I'm just doing something like this:

$ran = rand(0,1);

if ($ran == 1)
 $ad = 'foo';

else
 $ad = 'bar';

So there are a lot of external variables that could affect the RPM, but I'm hoping that looking at an average over a few days will balance those other variables out.

[tangor]

Urgency understood! Fingers crossed you get enough data to make a decision that will improve things.

Meanwhile, urge all examination of where to cut back, or reduce operating costs that make sense (as in, traffic is "x" but I have "xy+"). Without knowing your operational requirements/expenses can't help there, but will say, if the sites are online only because you have credit cards, you're not making any money!

Ran into a similar a few years back with a client. They started off at the top (dedicated rack, etc) and an initial line of credit but burned through that within 10 months. When the "help me!" wail was finally uttered my recommendation was to drop the expensive iron and go with a top rated shared UNTIL THEY COULD ACTUALLY BUILD THEIR BUSINESS.

The users never saw a difference (perhaps a bit millisecond slower in some cases), the biz survived, and two years later JUSTIFIABLY moved back to "big iron". :)

[csdude55]

In my case, I need to gross a minimum of $4,500 /month (USD) to cover all of the expenses. That includes rent, utilities, payroll, etc. I've cut staff to a skeleton crew and I'm under contract for the office space, so there's not a whole lot more I can cut :-(

I've been barely covering my nut since early 2017, honestly. Before that I would usually hit $10,000 in a month, but everything just dropped at around that time. Not to be political, but I've noticed that the drop on my end was right around the 2016 election. Traffic is higher than ever, more users than ever, it's just the RPM that died. I've depleted most of my savings trying to cover everything during some of the bad months.

And I'm 100% serious, for the last few months I've been missing the minimum mark by over $1,000, so I've had to start dipping in to credit cards to cover expenses.

It's a scary position, and I'm really just hoping that ad revenue goes back up in the very near future. We had 14 years of solid profitability followed by 2 1/2 years of barely scraping by, and now a few months of being in the red. So I'm trying hard not to come up with a permanent solution (eg, going out of business) to what may be a temporary problem.

[tangor]

Given the reports of others at WW ... if your income is based on third party ads, this might be the new normal. the "small folks" are seeing drops all over the place, while the brands/big fish continue. How much of this is a change in advertiser spends, bids, etc or a consolidation of the market by vendors, or an increasingly adversarial relationship with users who are finally taking note of THEIR expenses related to advertising and responding with blockers, it not yet known. Might be a combo of all the above and other factors which are yet to be discovered.

Not to rain on anyone's parade, but the golden days are gone and will not come back. To keep liquid one needs to change as well. Brick and mortar locations aren't that important any longer, "tele-computing" is the new normal, and direct ad sales is the answer to dwindling agency ads ... but that is a different kettle of fish.

Eggs in one basket seems to be the problem and, other than the "no blueprint offered" suggestions above, I don't have any hard and fast answers!

[Dimitri]

Sorry for your misfortune csdude55.

If you have still good volume of traffic , you should consider other ways to monetize it. Like selling ad space yourself.

I don't know in which kind of domain you are, but, may be you can also provide services to other professionals. Over the years, you certainly acquire a good level of knowledge about Internet stuff. So you might try to capitalize on these knowledge to sell services to other companies. Like for example, you might propose to develop web sites, to host them, things like that. Assists others in their online development. This might create extra stream of incomes.

[csdude55]

I had a nice, long reply typed up, then my browser froze up and I had to reboot the computer... lost the whole message :'-(

@tangor, I think the problem (at least for me) is a blend of ad blockers and social media.

I try to track how many users I get with ad blockers, and it's near 50%! The concern, though, is that the wide majority of those users have no idea they even have one installed. They got a virus or spyware and took their computer to be repaired, then someone installed an ad blocker to stop the symptom instead of removing the malware. So when they get a message asking them to whitelist my site, they have no clue what I'm talking about.

But that doesn't affect the RPM from Adsense, just the total revenue.

Comparing 8/10/2019 to 8/10/2012 is darn near heartbreaking. I have about 5% more users now than then, but about 14% fewer pageviews (which makes sense, mobile users view fewer pages per session). But the revenue was 44% lower! That implies that fewer people are bidding on the same ads.

And that makes sense, too, in large part thanks to social media. In 2012, everyone put their marketing budget on Google. But now, they split it between Google, Facebook, Twitter, etc.

Worse, where the internet used to be the great equalizer between big business and small business, it's becoming next to impossible for small businesses to compete. How can a free information site compete with a company that started out with $350 million from investors, and gets another $350 million in tax subsidies?

[subsidytracker.goodjobsfirst.org...]

And while I'm having to pay my annual taxes with a credit card, they get away with paying virtually nothing:

[itep.org...]

This means that information companies like mine are going out of business, which means that they're not advertising. Fewer bidders means that the remaining big companies pay less for the same ad.

How can a small retail company compete with Amazon that gets $2.4 billion in tax subsidies, plus discounted shipping rates?

[subsidytracker.goodjobsfirst.org...]

And again, while the small retailer pays about 30% of their income on taxes, Amazon pays nothing:

[usatoday.com...]

It's just impossible to compete against a company that could not sell a single ad or product, and would still make a multi-million dollar profit. So again, these small retailers are going out of business, which means that they're not advertising, which results in lower bids on the ads.

It's a problem that grows exponentially. Free information sites die away, leaving fewer and fewer options for the existing retailers to use to market their wares. So the retailers die away, leaving fewer bidders on the ads for the remaining free information sites.

As long as the government continues to spend tax money to reward big business we're going to continue to circle the drain, until there's nothing left on the internet but Facebook, Twitter, Amazon, and Ebay.

Sorry for the rant :-(


@Dimitri, my business actually began with me doing web design! At the time, the average contract was around $2,000 :-) But I officially stopped doing it a few years ago when, thanks in large part to Wordpress, I was competing with developers that could build an attractive site for about $200. Sure, they were losing out on the marketing advice and flexibility that custom programming would give them, but it's hard to argue against a $200 option that, to the naive eye, looks good!

We still offer hosting, but it's tough to compete against GoDaddy's $6 /month plan with a free domain.

So in both cases (web design and hosting) I need to get a high number of clients to make it worthwhile. It would take 1,000 hosting clients to pay the bills... I have 24 right now, so that's a long way to go! LOL

I started another thread a few weeks ago about direct ad sales. They do help a bit, but they've never really been huge for me. That's a whole 'nother story, though.

[tangor]

While big biz does enjoy some advantages (as in workplace/infrastructure/and JOBS) for some local and state entities, moaning about that difference is generally a waste of time and emotion. Apples and Oranges and all that.

One CAN, however, change the way business is conducted in "your" niche of the web. All of those options are not "cut and paste" or "set it and forget it" solutions. Real work, and TIME are required to make things happen. Sadly, building the pipeline of customers/advertisers for the future is more difficult now than it ever has been on the web.

About 2 billion others are already working on those concepts! (slight exaggeration, but only very slight!)

As for your hosting options ... the number of boutique hosts is large, and profitable, if one goes after BUSINESS, not the $6/month "noise". If your code/dev chops are still there, that's where some significant change can be made. WP is fine for some, but real business needs something more than that. However, that is real nose to the grindstone labor ... but you can charge accordingly. If you are running on your own iron, paying the costs/operations, etc, 12 clients at FULL SERVICE per year can cover your missing nut, and then some. Something to consider.

[Steven29]

God, grant me the serenity to accept the things I cannot change,
Courage to change the things I can,
And wisdom to know the difference.

You cannot change the ad blockers or the competution. Dont let it get you stuck. You did something right for 10 years... maybe its time you showed them whay cssdude can really do!