PDA

View Full Version : Stumble Upon Paid Discovery



orion_joel
10-30-2011, 09:01 PM
Hi,

Just wondering if anyone else here had used StumbleUpon Paid Discovery for their website, and

1. What were your results, was it beneficial?
2. Would you/Do you continue using it?
3. How long does it generally take for approaval of a campaign?

No. 3 is probably the most important to me at the moment as I got the account set up and have been waiting at a guess over 48 hours at this point. I suppose it is currently the weekend in the US, however not sure if this would impact or not?

Thanks for any comments or information you can offer.

orion_joel
10-30-2011, 10:14 PM
It seems if I had looked a little further on their website I would have found the answer to question 3.


The review process usually takes one (1) business day or less. If you submit your campaign during normal business hours (Monday – Friday, 8AM – 5PM PST), chances are our team will review it within a few hours.

Still interested in comments or experiences for Q1 & Q2

Thanks

KristineS
10-31-2011, 12:02 PM
Wow, I hadn't heard of this. I'll be interested in hearing how it works.

vangogh
10-31-2011, 12:09 PM
I've never used it myself though I've been aware of the service for quite some time. When I look at my existing stats, StumbleUpon users don't seem to do much on my site. The typical Stumbler lands on a page and is gone in under 10 seconds. It didn't seem like a good idea to pay to get more of that kind of traffic.

That's not to say traffic from StumbleUpon is always bad. I recently read a post suggesting it was better than other social traffic. If I can find the post again I'll link to it. It's possible that my site isn't a great match for the StumbleUpon or maybe I need to tweak something on the site. For now though I don't see the value in paying for the traffic.

orion_joel
10-31-2011, 11:53 PM
I have to agree with you Vangogh, previously I have seen that visitors coming from sites like Stubleupon or Digg, are fairly quick and push up the visits a lot without really resulting in any actual stay time on the site.

However I thought it would be something interesting to try. It got approved about 15 hours ago, and in that time has reached my set budget in the past hour. I thought set it low ($2 per day) which at the normal rate works out to 10cents per stumble or 20 visitors. Now maybe this is not the biggest number to be working with but, even after day 1, I am seeing that it is something that could possibly work, but it will require work to make it the most beneficial.

First to see the actual results properly you need to combine Google Analytics, your web statistics and the Advanced reports in StumbleUpon. The claim is that Google Analytics will not show all of the results as visitors, because many StumbleUpon users use Firefox with plugins to stop Javascript. So fair enough whether true or not isnt an issue for me. I can always confirm referrels through AWstats or something. However this is going to be 24-48 hours behind as my host does not allow updates outside the fixed updates of AWstats

The part of where you need to work on this though comes from the stats you get with StumbleUpon. They cover Topic, Gender, Age, Device, State, Country and Date. Each one gets broken down into whatever breakdown and gives stats based on the groups. So for example Age breaks down into six different groups. For each group it gives you some information such as the traffic from that age group, how much it cost, engagement (if they thumb up or down the page) and time on site. This is the best information I think.

The reason that information is so good is it allows you to tweak the campaign, for example I have my site appearing to anyone and everyone in Photography and Travel (it is the site in my sig). I can see how many people are showing up based on each category, and how much time they spent, then what age group spends the most time and what country they are coming from. So it means I can tweak the campaign based on the stats, ideally to improve them. Whether this works practically or not I will have to just wait and see, but good start I think. I will tweak some of the settings a little and see how if it performs better or worse tomorrow.

orion_joel
10-31-2011, 11:55 PM
By the way if you have any other questions let me know, I will be trialing this for at least 10 days to start and may expand it depending on the outcome.

By the way with the demographics pretty general average time on the site was about 10 seconds according to StumbleUpon, which compares to an average on Google Analytics of around 11 seconds. However I only got round to putting Analytics on the site a day or two ago. So not much data.

vangogh
11-01-2011, 10:43 AM
You should definitely try it. I'm only going on casual observation of the traffic to a couple of sites, which is hardly scientific. In general social traffic tends to hit and run, though there are certainly some within that general who'll stick around for more. For example Twitter sends me a fair amount of traffic. Most of it doesn't stick around too long, but Twitter makes up a pretty high % of subscribers to my blog so some of those people are sticking long enough to decide to subscribe.

With typical SU users they're in the random click mode so it seems reasonable that to make them stay you have very quickly capture their attention visually with a match to whatever it is they might be looking for or hoping to find. Inviting images at the top of pages are often reported to work well. That's a fit for your site. You might want to test by trying the first image further up on the page, maybe even before the main page heading so an image is the first thing people see.

Being able to tweak the demographics should help. That seems to be a major benefit for all social media advertising.


average time on the site was about 10

That's what I've usually seen in Analytics. I see more from search on average. With search I get a large amount of people who do the same hit and run thing, but a larger % of people who stick around for a few minutes.

orion_joel
11-02-2011, 01:30 AM
Ok second day results are in, and it is still looking pretty similar results.

The time on site is still averaging rather low, however it depends on the way that you look at the statistics. While the average stumble upon is showing is still around 10 or 11 seconds. Google Analytics is showing an increase to around 15 seconds. This could be a result of the shorter visits not being counted on Google though due to moving on before the Analytics code has loaded. Or any number of other reasons.

However when you look at the stats state by state in Stumble Upon I am seeing some visitors are staying longer. Between 1 - 5 minutes while some are not even there 1 second.

The age demographic is really something that is interesting most of the visitors I am getting are coming from 17-25 years but they stay an average of 7 seconds, while 25-25 stay an average of 15 seconds.

I think it will become more successful as long as I continue testing and really hone in on the ideal demographics, that seem to be spending time on the site.

Another interesting thing, i did a trail between sending them to my main home page, or to a specific post. The ones that arrived at the specific post, stayed longer averageing at around 20 seconds.

I think at this stage it is something that will work, it just depends on if you can afford the budget to keep the campaigns going. Which will depend on if running the campaign can improve the income from your site. (of course which first requires a product or some form of advertising happening).

vangogh
11-02-2011, 02:14 AM
People staying longer on a specific post makes sense. I wouldn't expect anyone to stay long on a home page as it means they have to click more to find something they're looking for and it comes across like an ad or like self promotion. Most people aren't going to stumble the home page of a site. I'd drop promoting the home page entirely.

Can you segregate data in Analytics by people who spend less than 10 seconds on the site and people who spend more. The less than 10 seconders are going to pull the overall data down. Those aren't the people you're looking to reach so if you can eliminate them from the data you'll likely get a better picture of the people who you are looking to reach.