Sales Gaps?
 

Can anyone point me in the direction of some scientific and/or psychological studies focusing on sales patterns, buying habits, etc? I'm trying to determine why (at times) we see 'sales gaps'. You know, traffic flowing in yet hours and hours without a sale.

So far, based on my own 'research', I think these 'sales gaps' are a result of one or more of the following factors:

1. Day of the week/month/year
2. Economy
3. Scrub
4. Visa
5. Chargeback levels

What day of the week or month seems to play a significant factor. Perhaps this is because people get paid end of the week, twice a month, etc. Time of year seems to be an issue, too (Christmas, tax season, summer slowdown). The economy, overall, has an effect.

Then we have "the manipulations". No, I do not mean this in any ominous or nefarious way, just the normal course of making 'adjustments' to a business over time. So the scrub and chargeback levels can play major factors. Even Visa can suppress things in the wake of a Target hack or some such credit catastrophe.

Perhaps this is why brick-and-morter stores have their sales on Tuesdays & Wednesdays? Weekend specials and Holiday sales, etc? Yet Adult doesn't seem to follow these 'mainstream' sales trends or tactics and I'm wondering why. Has anyone tried "Half-Price Tuesdays" or Weekend Specials to try and counteract these sales gaps? Or are these gaps 'normal' and part of the human experience? :)

Thanks for the insights!

For scientific data I would start by making sure you understand these and then try to apply them:

https://en.wikipedia.org/wiki/Standard_deviation
https://en.wikipedia.org/wiki/Margin_of_error
http://www.mathsisfun.com/data/standard-deviation.html

If you're really curious grab all the data you can and start breaking it down. Look for patterns. Find out what is really going on. Compare trends based on day of the week, day of the month, season, etc.

As far as art I remember talk here about sales coinciding with the phase of the moon. Actually I wouldn't be surprised to see some kind of correlation at all! :)

If you want to analyze your data then start by collecting emails on the pre-join.

Analyze those emails, is there a gap, do the emails look legit? That will tell you a lot more then a ccbill form hit with no sale.

Scrub, visa quirks, cb levels.. you don't have / can't get that info?

You'll see inexplicable patterns if you work brick and mortar too, but some things shouldn't be mysterious.

I don't think scrub levels move up and down like one day to the next, but happy to hear a rep come talk about it.

I think this is very difficult to determine especially that sexual arousal is triggered by literraly so many things. Even the weather must be included among the factors. And you cant just make a study based on a few niches. You will need to know when one niche does better than the others -and I dont think all the adult programs will be keen on sharing their data.

it is the billing/ proccessors' fault....

Also see this; http://www.buildinganempire.com/poisson2.html - heavy, but an excellent read.

One of the most important points in the article; "So if you normally get 3 signups ... per day, you are going to get 0 signups once every 20 days. This is not because there is anything wrong with your sponsor, it's just the way the distribution works.".

That is very interesting thank you! :)

Volume is essential, of course. But I have 34 paysites so if I only did 3 sales a day I would cut off my testicles, dip them in bronze and wear them as earrings. :D

Good read, thanks!

very nice share, thanks!

"Most people think, ok, I'll send a certain number of visitors to a sponsor and see how many signups I get. What you should be doing instead is saying ok, I'll send a certain number of signups to a sponsor to test them."

and as in any study if you don't have enough data the test is rendered worthless. Most forget about this part and fill the gaps with assumptions instead.

LOL! Agreed. As others have said, volume is crucial. The more data you have, the better the decision making.

I was just illustrating that many (bad) decisions can be made based on false assumptions.

I don't know how many times I've had traffic buyers want to buy 1000-5000 hits (~$3-$15) to test how good the conversions would be. They don't understand that's not NEARLY enough traffic to test either the source or the effectiveness of the landing page.

Now that depends on their average ratios, if they normally average 1 in 150 then 5000 hits is more than enough to see if it's worth speculating on your traffic, however if they normally average 1 in 1500 they'll need ten times as much traffic to get a fair idea :2 cents:

Your site sucks. Improve it.

Oh ... sorry to disrupt a good discussion with that but i make a valid point put in a very stupid way.

I only say ot that way cause pornnerd you make these threads all the time during a panic attack aka sale gap

Get more volume then you won't shit a brick when sales drop. I say this from experience...

NOBODY does 1:150 on traffic that sells for $3/1000....

Great link. :)

I like doing 3-5 times the expected conversion ratio. So if I expect 1:1000 then I'll decide with 3000 - 5,000 clicks. That's usually enough for determining the worth of sponsors anyway. Too much and you are just wasting traffic.

No worries mate, some people love my sites others think they're shit (but the same can be said for ANY website). I make money so that's the final opinion I care about. LOL

Oh - and this thread wasn't started "in a panic" but in an honest attempt to understand sales patterns. Some extremely helpful links have been posted here so this thread was worth it (to me).

Thanks everyone!! :)

:thumbsup

Unless I miss something he is off on one point though.

With the way he is using it though he is using "how many sales in 10,000 uniques" so that IS taking into account the number of visitors. It's just that he is doing it outside of his Poisson equation beforehand as part of his variable inputs. Still it is subject to it.

IOW his value for 'm' or 'x' is actually a signup rate and not just the number of signups. It's signups per 10,000 visitors in that specific case.

You still have to take into account the number of visitors for obvious reasons such as the possibility that tracking isn't working or the sponsor is shaving. There is also opportunity cost in having a given link up and sending traffic. If the one sponsor isn't selling you can use another. It would be folly to send 100 million visitors and to keep sending because "you are trying to get 3 signups for statistical comparison purposes". :)

Only one of you is a mathematician with a degree in physics from Yale. Challenging Colin in math is not a great plan.

Also interesting to note that stuff was written at a time when sponsors would kill accounts with traffic that didn't convert because bandwidth was so expensive. Hosting companies used to get burned all the time by kids getting galleries accepted at the Hun and running up a 4k hosting bill in an afternoon and many hosting companies blocked traffic from a lot of those big sites back in the day.

LOL well everyone makes mistakes, I'm only saying what he said there doesn't seem to make much sense.

i see a sale gap too

It would be interesting to graph current weather against signups. You can pull member location from biller or IP address, and match the time and date of the signup against a comprehensive weather almanac database... Could be fun.

Even easier would be to start logging it on the fly today :thumbsup

Actually, I did a kind of 'soft analysis' regarding the weather and other factors today. LOL!

Funny thing is, about 60% of my sales come from the US, Canada and the UK so I looked at what was happening in March/April:

NCAA tourney
Tax season
Spring Break
Beginning of baseball season
shitty, horrible weather/endless Winter

No wonder there are 'gaps' eh? :D