4 replies to this topic

[Matt B]

Interesting news. It vindicates my opinion about Jupiter's recent study that gave findings that 10% of US Net users delete the cookies on their web browser every day and 40% delete every month.

It seemed a bit high, ok, very high. Seth Godin took a lot of heat for his criticism of the story, giving his favorite example of how the figure couldn't correct:

My favorite bit of proof: One of the top 100 things searched for on Yahoo! was "Yahoo". Also on the list when I was there: "web" and "search".

Well, here it is: MarketingVox has an extensive study that found multiple problems with Jupiter's data, and found three main areas of innacuracies:
1. First party cookies v. third party cookies
2. Spyware programs that delete old cookies
3. Users reported different behavior other than what they actually DID. (no!)

Attached Thumbnails

• 

[Randy]

Yeah, I have to admit that I immediately discounted the Jupiter report when I read it too Matt. Total bs IMO because the cause didn't equal the effect.

Not only was it not a valid study, but I'm a lot more security conscious than the average Internet user, and I don't delete my cookies. Ever. I may delete the odd cookie here or there, but I never wipe them all out or I'd never be able to remember my logons to various places.

Then again, I do not allow any third party cookies in the first place. If it ain't on the site I'm trying to visit, I don't want it. In my mind that's spyware waiting to happen.

I'm glad to see that someone took it upon themselves to totally debunk it, because the Jupiter stuff gets trusted immediately. Even if the study was flawed from the get go.

[Alan Perkins]

I don't delete cookies but I also don't accept many, either. Is that the same thing?

I accept all session cookies. I accept permanent cookies from trusted sites. That's it. Most sites still work. If they don't, I either make them trusted sites or I don't use them.

It's partly a privacy thing but it's also that object to my hard drive being filled up with other people's rubbish. My own is bad enough.

[cryptblade]

I think this cookie thing still doesn't matter. Users delete cookies, want to delete cookies, or intend to delee cookies.

I'm security conscience and I'm net savvy enough to know cookies aren't all bad. I still delete them. I keep my passwords in a file on a flash drive. Used on a need2use basis.

I think so long as this is talked about - cookie deletion - even if the study is wrong, the first-movers' principle of JupiterResearch's study will get more attention and perpetrate a misconception long before the study's rebuke can take into effect.

As an example - look how quickly Jupiter's research spread and lots of writers wrote about the doom of the cookie and other stuff. Now, only now, do we start seeing some rebuke. Too late.

I think the study sounded the alarm for other layfolk users to delete cookies - not because they are zealous about security or just plain ignorant, but because they don't care.

I suspect users who read about Jupiter's study and aren't Internet savvy will think that cookies are lumped in with spyware and everything. Just the posts here says it all - all of you say you allow this cookie or that cookie but not this cookie or that cookie... so what? Who on average wants to waste their time tracking what to allow or not? I think most people will take the attitude of "kill them all let God sort them out" - screw the cookies - get rid of them all... from time to time.

I know I take that attitude... and some times, I just want my machine to be purged... like a colonic type of thing....

[Haystack]

It's interesting how many seminars at SES last month talked about how much better data you'll get if you don't automatically expire cookies after 30 days.

Looking at the stats of a few sites I support, there is a significant number of sales more than 30 days after the initial visit, that the sites I have in mind are for relatively low price point retail. I'm sure the sales cycle is considerably longer for b2b and higher price point retail situations.

Same day sales skew high in the data from people who previously deleted cookies. The tough deal there is they may have come from some form of paid ad, but most stats programs will only report them ad a typed in/bookmark type visitor.

Another challenge is people surfing from more than one computer. How do you track people who researched a purchase at work, but waited until they got home to whip out the plastic?