Is the the 40% a result of too small a sampling size? In other words, losing 4 subscribers when you only have 10 is also a 40% drop. The feed in question had a lengthy history of upward trending subscribership. Prior to running ads, the feed averaged 1,250+ subscribers for the last 30 days and never once did it dip below 1,100. So the sudden 40% drop is definitely not due to a small sample size.
Could it be coincidence that almost half of the subscribers were actually not influenced by ads showing up in the feed and just happen to feel the need to unsubscribe at the same time that Feedburner Ad Network ads were turned on? It’s possible but I also think it’s highly unlikely due to the historical upward trend of subscribers.
One possible reason (at least one that I can think of) for this sudden dip in subscribers may be that, somehow, feed ads are causing the subscription numbers to be reported differently / incorrectly. Maybe ads are wreaking havoc with the stats? I know this is a stretch but it is still a possibility – one that is more probable than having almost half of your subscribers leave on the same day as when ads were turned on. As Sherlock Holmes says, “when you have eliminated the impossible, whatever remains, however improbable, must be the truth”.
Regardless of the cause, I was disappointed in the significant drop of subscribers and with feed advertising in general. So much so that I am now not even going to consider testing other feed advertising opportunities like Feedvertising from Text Link Ads – at least not anytime in the near future. Guess that just leaves more time to work on other revenue channels.
Anyone else experience a dip in subscribers after running ads in your feed?
ian,
this is don from feedburner. i’d love to talk to you about this – i highly doubt it has naything to do with advertising. your subscribers fluctuate up and down normally depending on what you’re posting, how many people go online that day and fire up their aggregator, etc. also, two days seems like a short evaluation period – it’s pretty hard to draw conclusions.
also, here’s some information on how we calculate subscriptions…they are important, but actual views, clicks, and active reach are probably more important.
let me know if you’d like to chat…we want you to be a happy customer.
thanks.
don loeb
vp partner services
feedburner
i noticed the same thing too and had to remove feedburner ads. i noticed a 35% drop. that really sucks….i won’t give up my readership just to make a few bucks more.
Don,
Thanks for the link. There is lots of information on that page and it’s definitely a good read for anyone who wants to learn more about Feedburner stats.
I do agree with you that 2 days is a rather short test period and that actual reach and activity is much more important than subscriber base. But it was still disheartening to see all that hard work in promoting the brand and getting the feed out there kind of disintegrate over night.
Yes, active users are important, no doubt about that. But at the same time, you can only increase the number of active users if they are firstly subscribed to your list. This plus the fact that it was pretty wonderful seeing on-going positive results of promotional efforts correlated with a growing subscription base. This can get addictive. 🙂
I actually quite like the newspaper analogy (newspaper being read versus one that is rain-soaked while the subscriber was on holidays) in the article that you shared. But I think that marketers can take advantage of getting their brand out to eyeballs even if there is no active clicking on content within the feed today. Some users do not click today but they may tomorrow. Even though the newspaper is left outside today, it may be brought inside and enjoyed over breakfast tomorrow. And if the subscriber cancels the subscription, there is absolutely no chance of that happening at all.
This actually leads me to wondering if we are moving into the hot topic of “how clean & active is your list” from the email marketing world. The only difference would be that in email, there are underliverable address and in feed marketing, all subsribers are potentially reachable. Do we want a massive feed subscriber base with low activity? Or do we want a highly active but smaller base? It’s probably more of a matter of choice in deciding where the fulcrum lies.
Thanks for sharing and for the offer Don. I do appreciate it.
Mark,
Sorry to hear about your dip in subscribers. I agree – sometimes, the money is just not worth it. Remember to focus on your business first and keep doing what brings you success. The money part will come later.
Hey, wow! 40% is pretty much HALF of your subscriber =(. Anyway, have you tested the Amazon ID thing in the Optimize section?
ian – glad you found the information useful.
just to be clear, i highly doubt that advertising has anything to do with your subscriber number going up or down. what’s the size of your audience? how many people does that represent?
maybe some folks went away on vacation, maybe a web-based aggregator didn’t report their numbers correctly…it could be any number of things.
we deliver ads into thousands of feeds (obviously on behalf of publishers who ask us to do that). the number of views, clicks, and subscriptions we track has risen an inordinate amount in the last year. no one is losing subscribers due to ads. again, it could be anything and it could be temporary…
also, our ads are extremely unobtrusive and the environment is very uncluttered…especially compared to most web pages.
happy to continue to discuss…
thanks.
don
Don, it’s the 3rd day after implementing FAN and subscribers are still on the decline. The latest decrease brings total lost to just over 50% (800+ subscribers). Something is definitely happening that is not part of the normal fluctuations.
Sure enough, after digging more into the stats, I see an eye-popping trend. I would say that almost roughly 95% of the lost were Google Desktop users. I had guessed in the original article that:
I am beginning to think even more so that this decrease in subscribership has something to do with the way Feedburner stats are skewed by Google Desktop’s “drive by” feed request as first reported here.
Thoughts?
Ian –
Thanks for bringing this to our attention. If you can contact me offline ([email protected]) with the feed in question, I’d be happy to dig in on this.
As Don suggests, we have thousands of feeds in FAN and we’ve never observed a single case where a publisher experienced such a precipitous drop in subscribership after joining FAN.
If, as you suggest, Google Desktop is the user agent experiencing the biggest drop-off, I’m puzzled as to how advertising could have influenced that: Google Desktop’s RSS widget doesn’t show images, only text – so any Google Desktop subscribers wouldn’t see the ads to begin with, let alone have their feed reading affected by the presence of ads.
Let’s figure this out – we’re pretty confident that FAN isn’t negatively affecting any publisher’s overall audience numbers, but I’d love to help dig in on this particular feed’s history to better understand the dynamics and figure out what happened.
I look forward to your e-mail.
Regards,
Rick Klau
VP, Publisher Services
FeedBurner
[email protected]
AIM/Y!/Skype: RickKlau
Thanks for the reply Rick,
No fair. I was the one hoping for someone at Feedburner HQ to shed some light on the matter. As an outsider who does not have access to internal FB data, I can only hazard a guess that FAN perhaps somehow triggered a cleansing of the subcribership. Maybe it affected the reporting of all auto-subscribed users from Google Desktop and the list is now being purged so to speak? How was that? Sounds logical but still only my guess. 🙂
Email coming shortly.