Like math? You better learn to. - Page 3 - WickedFire - Affiliate Marketing Forum

bigsherm · 06-22-2009, 01:05 PM

Assignment 3
You are in discussions with an advertiser for a special deal on affiliating their product. The advertiser pays you NET30 (30 days after the commission was registered). This means if you sent leads worth of $ 1000 in commissions one day, you are credited that amount 30 days later. Currently you send leads worth $ 40 000 in commissions every month.

You offer this deal to the advertiser:
"You can decrease my commissions by 1 % if you pay me NET10 from now on."
Under what condition does he accept your offer?

My Answer:

I would assume the advertiser wants the potential to make more money than they're giving up. The 1% is really insignificant here, it's the 20 day difference in payments that affects the outcome. With 40k a month, that comes out to 1,333.33 a day (30 day month). With that daily income, we multiply it by 20 days = 26,667 and take out the 1% commission decrease to get $26,400 dollars you make in 20 days at the current, non-reinvested rate. Beat this and the advertiser should consider.

Taking that 20 day total, you need to show the advertiser that you can reinvest your payments every 10 days to offset the risk they take by fronting the cash to you while they wait for their payment on the backend from customers, transaction cycle times, and processing which could be instant or have a "14 day trial" where they get the value back later than the NET10 you want.

With that, we need a little bit more information about how each side is making the money, how the advertiser is getting paid and on what terms, and what kind of metrics you're getting on your investment as the affiliate, namely the ROI and how the speedy payments can grow past that 26,400 mark in 20 days.

Assuming you get an ROI of say 35%, you're in full swing getting commissions every 10 days on the new terms and you reinvest 100% instead of renting a Lambo for the weekend...

10 x 1,333.33 = 13,333,33 *1.35 = $18,000 (Note: you have two 10 day periods in the difference of payments to grow your return in hopes of surpassing the 30 day value that is already there, this does not count towards it as this is your starting value without the new NET10. In other words, you make this currently and begin to take advantage of NET10 the following week.)

18,000 x 1.35 = $24,300 (day 10 of 30days with NET10...the idea behind this is that you have added some value to the daily investments that, when compared with the previous 30 day total, will be of more value to the advertiser. That and it provides a certain % cushion against the risk the advertiser faces of non-payment and thus in fronting you the money; based on the assumptions here).

24,300 * 1.35 = $32,805 at day 30 vs the $26,400. With these calculations and assumptions, the benefit to the advertiser would be that when you reinvest your earnings, you create an additional 6,405 in revenues on top of the $26,400 or 24.5% return. If the advertiser has a greater risk than 24.5% of providing a faster payment (ie chargebacks and cancellations or just non-payments), it may not be a good idea based on the numbers. however, if their risk is that bad already they probably have a poor product.

I hope that made sense.

bigsherm · 06-22-2009, 01:25 PM

Assignment 4
You have a website that is budgeted to generate $ 27 000 in revenues pr year the next 4 years.
Webhost costs are $ 1 000 pr year. Advertising costs are $ 2 000 pr year.
You outsource the article writing for the website. You will put on 40 articles pr month, 25 of which are 600 words (per article). The rest are 1000 words per article.
You will pay the article writer $4/100 words.

It is expected that you can sell it for $ 25 000 in 4 years.
Someone is offering you $ 50 000 for the website now.
The rate of return is 10 %.
What do you do - sell it now or keep it running and then sell it after 4 years?

Answer:
Fun little problem, good thing I have my financial calculator from college handy...I may be rusty so this could be off, but here it goes.

The key to this problem is to get the NPV of your website when taking into consideration the cash flows of each year and the cash flow in year 4 + sale price. If it's more than 50,000 you hold it and sell. If it isn't, take the money and run. Assumptions here are that there are no other investments that the 50,000 now can be put into and that we just use the data here.

This can definitely be done in Excel, but I had my calculator handy so bear with me.

$27,000 RevenueYr
- (3,000) Upkeep
- (14,400) Content -(25x600)+(15x1000)= 30,000 Words/100 x $4 =$1200/month
______________
$9,600 profit a year.

note: Multiplying $9,600 by 4 and adding the $25,000 doesn't give you the real value. The rate of return is going to be used to calculate the time value of money and take into account inflation, buying power, and the actual return.

I: 10%
CF1: 9600
CF2: 9600
CF3: 9600
CF4: 9600 + 25000
NPV: $57,106.04

So with my calculations, hoping they're right, you're better off holding it. Personally, I'd take the $50,000 and find a better investment like PPC, however the numbers as they are setup say holding is more profitable. If this is right, I guess I can almost justify spending over 100k for college

bigsherm · 06-22-2009, 01:33 PM

Assignment 5
You are an email marketer. You have observed the following:
Total emails sent pr week for 10 weeks: 49872, 52034, 47322, 56999, 48932, 43456, 53104, 50888, 45032, 49999
Total costs pr week for 10 weeks (the same weeks as above): 1500, 1743, 1388, 1844, 1599, 1333, 1900, 1011, 1803, 1800
You want to estimate a function for your total costs for a week as a function of sent emails that week. What is it?

Answer:

This can be done simply by finding the average of each week's cost per email and then average the 10 weeks for an overall average.

I used Excel to get an average cost/email of $0.0320.

# of Emails * $.0320 = Cost/Week

knukk · 06-22-2009, 02:43 PM

bigsherm:

Assignment 3: God, you make it complicated rofl

The 1 % _is_ very important
The $ 40 000 are irrelevant
The advertiser has to find the effective interest rate. Find it!
The advertiser compares this rate to ... and chooses to ... if ...

Assignment 4: VERY good! Everything is correct except for the NPV (and thus the end-answer). Maybe you pressed something wrong on your financial calculator.

In pure mathematics: NPV = 9600/1.10 + 9600/1.10^2 + 9600/1.10^3 + (9600 + 25000)/1.10^4 ~= 47 506

NPV < 50 000 --> thus, sell now!

Despite of that, very good and explaining answer.

Assignment 5:
Good

Alternatively, you can use regression and you will find:
C = 262,79 + 0,02671 * x
Where C = total costs, x = no. of sent emails.

bigsherm · 06-22-2009, 05:05 PM

Yeah, haven't used this calculator for a while now, looks like the interest was compounded 12 times a year as opposed to the 1 in the calculator. Oops.

As for #3, I'm going to have to look at that one, the answer didn't come right to mind like the others. Obviously.

Regression, I remember that from my stats classes...teacher was batshit crazy, maybe that's why I don't like it.

Thanks for the input.

keeper33 · 06-23-2009, 02:19 AM

That regression is funky, to say the least.

The regression coefficient (r) is .3699606367, so we have a weak correlation.

The coefficient of determination (r^2) is .1368708727, so 13.687% of the variation in y is accounted for by the regression of x on y, y being costs per week, and x being emails sent per week.

Needless to say, stats is only 10% running the numbers, the other 90% lies just below the surface.

knukk · 06-23-2009, 04:34 AM

I didn't check the r^2, but if those numbers are true, everything you say is correct.

iloveshi · 06-27-2009, 05:47 PM

Great stuff. Will make use of the spreadsheet and calculator when analyzing offers. That's one less mistake (out of one thousand) that I'll make on my noob journey.

+rep

James P · 06-30-2009, 01:07 PM

Hi,

Well, I just did what Henry Ford used to do. Rather than waste time working it ok, I just got someone more intelligent than me to tell me the answer....

bigsherm · 06-30-2009, 10:57 PM

Knukk,

Can you give the breakdown of question 3? I'm curious as to what you're searching for here as I have no ideas other than calculating interest over extra 20 days in holding vs. increase in revenue with faster payment turnaround....if that is the case, I'll give it hell again.

WFBenC · 07-05-2009, 07:17 AM

Option C has a claimed CTR of 0.1%

WFBenC · 07-05-2009, 07:19 AM

Yeah, haven't used this calculator for a while now, looks like the interest was compounded 12 times a year as opposed to the 1 in the calculator. Oops.

As for #3, I'm going to have to look at that one, the answer didn't come right to mind like the others. Obviously.

Regression, I remember that from my stats classes...teacher was batshit crazy, maybe that's why I don't like it.

Thanks for the input.

Jokerr · 07-05-2009, 11:08 AM

Just thought I'd share a very simple script. You set your estimated cpc, budget, payout and then the script calculates how many conversions you need to break even. You need to have PHP installed on your server, if you do, just upload it and the rest is pretty self explanatory.

PHP Code:


			
<?php

if ($_POST['cpc'] && $_POST['budget'] && $_POST['payout']) {

       $cpc = $_POST['cpc'];
       $budget = $_POST['budget'];
       $payout = $_POST['payout'];

       $clicks = $budget/$cpc;
       $conversionsTBE = $budget/$payout;
       $conversionRateTBE = $conversionsTBE/$clicks;
       $conversionRateTBE = $conversionRateTBE*100;

       $data = "You can afford apr. ".floor($clicks)." clicks. You need apr. ".floor($conversionsTBE)." converions to break even, a conversion rate of <b>".substr($conversionRateTBE, 0,3)."%</b><br />";

}

?>

<form method="post">
Esitmated CPC <input type="text" name="cpc" />
Your Budget <input type="text" name="budget" />
Offer Payout <input type="text" name="payout" />
<input type="submit" name="submit" value="Calculate!" />
</form>

<?php if ($data) { echo '<div style="border: 1px solid #000; background: green; padding: 5px; color: #fff;">'.$data.'</div>'; } 
?>

knukk · 07-05-2009, 12:18 PM

Quote:

Originally Posted by bigsherm

Knukk,

Can you give the breakdown of question 3? I'm curious as to what you're searching for here as I have no ideas other than calculating interest over extra 20 days in holding vs. increase in revenue with faster payment turnaround....if that is the case, I'll give it hell again.

Here is my solution to it:
The idea is that by you offering the advertiser a 1 % decrease on their payouts to you by paying you in 10 days instead of 30 days, they can measure an effective interest rate.

The formula for the effective interest rate is: (1 + per cent factor)^(days pr year / (previous payment time - new payment time)) - 1

Effective interest rate: (1+0,01)^(365 / (30 - 10)) - 1 = 0,1991 = 19,91 %. If the advertiser can finance himself at a cheaper rate, he will accept it.

Mkting Samurai · 08-09-2009, 02:15 AM

Thanks for the both the spread sheet and the script ...

The breakdown on question 3 makes more sense now ... I think

Thanks again

brucep · 08-10-2009, 04:05 PM

Oh, before I go, here's the spreadsheet I have open 24/7 for quick calculations. If you're not using something like this, take mine, or build your own. If you don't understand it at first, just look at the calculations I'm using. The colored cells are the only ones you need to change.

Attachment 675[/QUOTE]

Thanks for the spreadsheet it will be a great help!!! Also good to find a link where I do not get DICK ROLLED.

okuku · 09-01-2009, 10:25 AM

The spreadsheet looks cool. Though I don't know what the hell I'm gonna use it for yet.

I believe I will soon know wit you guy's help.

PhillipMarlow · 09-12-2009, 03:27 AM

Fuck yeah! Was about to whip one of these together...this saved me hours. Thanks +rep

imouttehre · 09-14-2009, 09:58 PM

Again another thing that makes me feel like a total dumbass..

Thanks for sharing the info and the questions guys. No PPC course or gurus teach this kind of practical/complicated/much needed stuff.

Mike, thanks for the spread sheet.

Can we get some more scenarios?? The cost valuation question for the website was really cool. With a small biz background, i would have jumped on the $50k offer.

Thanks again

orange · 09-23-2009, 11:42 AM

wooh good thread! tnxx so much, had a good time while reading -)

stevo7 · 09-24-2009, 09:25 PM

Much thx for the spreadsheet, Mike! Mucho useful.

theemerchant · 09-24-2009, 09:30 PM

I can't wait for the answer. I'll go for B if it was my decision.

imouttehre · 09-29-2009, 01:31 PM

Quote:

Originally Posted by theemerchant

I can't wait for the answer. I'll go for B if it was my decision.

the answers are already posted. Read through the thread.

balright · 09-30-2009, 10:41 AM

Damn shame my math major doesn't help me write content. Who wants to market on a math site? Anyone? Oh.

Carpaccio · 10-02-2009, 03:36 PM

Newb and math retard over here. Thank you, this is absolutely awesome.

whatyaknow · 10-15-2009, 04:30 PM

Awesome thread Mike and really appreciate the xls.

Anyone else have worksheets they use for calculations?

dhamma · 10-22-2009, 02:51 PM

Excellent examples - and xls.

Thanks

Harsh · 10-27-2009, 08:13 AM

i justed click on A but i saw the meaning of A after voting

bloodsucker · 10-27-2009, 06:46 PM

thank s God, i love math. I m more in peace..

Mike · 11-16-2009, 02:37 AM

Since I posted that original XLS spreadsheet, I've made a slight change to the CPM column by adding "Country".

"Country" on my spreadsheet is the percentage of traffic that comes from countries that I can theoretically convert. So, if I've got an offer that is US only, then I need to know what percent of their traffic is US. Obviously, if you are targeting US only then the percent is 100% (duh), but a lot of sites you'll find can't geotarget traffic for you, so then you'll need to enter the percent of traffic you can actually convert.

Also, I added "Off Page CTR" and changed the conversion rate so that it's based on the clicks off page, and not total clicks.

Just play around with it, you'll figure it out.

cpc-calculator.zip

knukk · 11-25-2009, 10:55 AM

Once exams are over I will see to that I post a new batch of assignments here.

metoo · 11-26-2009, 05:42 PM

I'll answer from just reading the 1st post, cos I know the rest of the thread will be filled with the answers...

You make 80 cents per click, so:

A: $1000
B: $2000
C: Take it if the CTR for the banner is > then 0.25

I'll leave this here with no edits. If I screwed it up, I figure who gives a shit. At least I had a go. My poor calculator, hehe

invinceable · 11-26-2009, 09:03 PM

Thank you very much for the spread sheet, looks very nice!

knukk · 12-10-2009, 05:40 AM

Okai folks! It has been a long time now since there has been some work here; time for a new batch of assignments. I will add more soon, hopefully

If you haven't done the previous assignments I posted, you should go back and practice on them. Solution proposals have been posted in this thread.

As for the new assignments, assignment 6 is very easy while assignment 7 requires some more work.

Assignment 6
You buy a computer for $1500 and do not use it for a year. Assume that you can and do sell it for $1500 after that year.

Question 1: What did the computer cost you?

Now, forget the computer.
Question 2: Generally, in your decision making, should you care about absolute or relative prices?

Assignment 7
You are interested in a new niche you are wondering about Internet marketing a product in. Assume that the product's a priori probability for high demand (H) is 0.4. There is only high (H) and low (L) demand. You want to test the niche by PPC. The testing will give you an indication 'g' if the signal is good, and 'b' if the signal is bad. From your previous experience in testing new products in similar niches, the probability of a good indication is 60% given high demand, while the probability of a bad signal given low demand is 90%. Assume these measures are representable here too. If you decide to market the product you will have a 35 000 profits if there is high demand, and 5 000 profits if there is low demand. If you do not proceed in marketing the product your profits are 15 000 from other marketing activites, independent of the demand state for the product in question. Assume that you are risk neutral, i.e. you only care about expected profits. How much should you spend on testing the product in the niche?

Expert · 12-11-2009, 02:31 AM

I really use these when dealing with my business models based on recurring income. You have to incorporate attrtion rate into the equation though.

knukk · 12-21-2009, 11:33 AM

Meh, does anyone want to give the new assignments a try?

IcanWrite · 12-23-2009, 05:01 PM

Nice job on these. The spreadsheets seem really useful. I love math though haha
hell yeah! i got it right!!

*niko* · 12-26-2009, 07:07 PM

Quote:

Originally Posted by knukk

Assignment 6
You buy a computer for $1500 and do not use it for a year. Assume that you can and do sell it for $1500 after that year.

Question 1: What did the computer cost you?

Inflation Rate, so if inflation is 3% I would lose $45

Quote:

Question 2: Generally, in your decision making, should you care about absolute or relative prices?

Totally depends on the lifespan of your project and the item. Also if you are thinking about selling said item after you use it or not. In general electronics have issues with resale value...

I tend to go with absolute price. And if it matters I weigh it by projected ROI % - 30 (because I always tend to overestimate)

Quote:

Assignment 7
You are interested in a new niche you are wondering about Internet marketing a product in. Assume that the product's a priori probability for high demand (H) is 0.4. There is only high (H) and low (L) demand. You want to test the niche by PPC. The testing will give you an indication 'g' if the signal is good, and 'b' if the signal is bad. From your previous experience in testing new products in similar niches, the probability of a good indication is 60% given high demand, while the probability of a bad signal given low demand is 90%. Assume these measures are representable here too. If you decide to market the product you will have a 35 000 profits if there is high demand, and 5 000 profits if there is low demand. If you do not proceed in marketing the product your profits are 15 000 from other marketing activites, independent of the demand state for the product in question. Assume that you are risk neutral, i.e. you only care about expected profits. How much should you spend on testing the product in the niche?

Working on this one...

knukk · 12-27-2009, 05:49 PM

Quote:

Originally Posted by *niko*

Inflation Rate, so if inflation is 3% I would lose $45

Good. I understand it so that you by inflation refer to a general price increase. I don't define inflation as such, but that's what you meant, so it's good. But such a general price increase is only part of the actual costs during that year. The general idea is to consider relevant alternative costs, one of which you mentioned.

Quote:

Totally depends on the lifespan of your project and the item. Also if you are thinking about selling said item after you use it or not. In general electronics have issues with resale value...

I tend to go with absolute price. And if it matters I weigh it by projected ROI % - 30 (because I always tend to overestimate)

Price - Wikipedia, the free encyclopedia

Basically what matters are the relative prices. It is not interesting in itself if a computer costs $1000 or $2000, but what the price ratios to prices of other commodities are. A cell phone for $10 000 may sound pretty expensive, but that is assuming that you are facing prices of other goods similar to those today. If the prices of all other goods were given a large shift upwards, then those $10 000 would start to look like a better offer.

Quote:

Working on this one...

papajohn56 · 01-07-2010, 07:21 PM

There's a lot more advanced math that you can apply to affiliate marketing, too. If you're good with statistical analysis and calculus, there's methods for it. I'll see what I can come up with to post here for more info.

I love math

knukk · 01-09-2010, 05:52 AM

Solution proposal assignment 7:
A quick intro to probability notation:
Pr(x|i) means the probability of x given i.
Pr(x & i) means the probability of x and i occuring together.
Pr(x) means the unconditional probability of x occuring.

Now let's solve the problem at hand:

Apriori:
Pr(g|H) = 0.6
Pr(b|H) = 0.4
Pr(g|L) = 0.1
Pr(b|L) = 0.9

Needed for later:
Pr(H & g) = Pr(H) * Pr(g|H) = 0.4 * 0.6 = 0.24
Pr(H & b) = Pr(H) * Pr(b|H) = 0.4 * 0.4 = 0.16
Pr(L & g) = Pr(L) * Pr(g|L) = 0.6 * 0.1 = 0.06
Pr(L & b) = Pr(L) * Pr(b|L) = 0.6 * 0.9 = 0.54
Pr(g) = Pr(H & g) + Pr(L & g) = 0.30
Pr(b) = Pr(H & b) + Pr(L & b) = 0.70

Aposteriori probabilities:
Pr(H|g) = Pr(H & g)/Pr(g) = 0.24/0.30 = 0.8
Pr(H|b) = Pr(H & b)/Pr(b) = 0.16/0.70 ~= 0.229
Pr(L|g) = Pr(L & g)/Pr(g) = 0.06/0.30 = 0.2
Pr(L|b) = Pr(L & b)/Pr(b) = 0.54/0.70 ~= 0.771

Optimal choices based on expected profits:
Apriori:
Marketing the product yields expected profits: 0.4 * 35 000 + 0.6 * 5 000 = 17 000
Not marketing yields: 15 000
Thus you choose to market apriori.

Aposteriori:
Expected profits if indication is g:
If marketing: Pr(H|g) * 35 000 + Pr(L|g) * 10 000 = 29 000
If not marketing: 15 000
You choose marketing.

Expected profits if indication is b:
If marketing: Pr(H|b) * 35 000 + Pr(L|b) * 15 000 = 11 870
If not marketing: 15 000
You choose not marketing.

Value of testing = Pr(g) * 29 000 + Pr(b) * 15 000 - 17 000 = 2 200

imouttehre · 01-11-2010, 06:04 PM

a request for more cpm problems?

Mike · 01-11-2010, 06:13 PM

Quote:

Originally Posted by knukk

Solution proposal assignment 7:
A quick intro to probability notation:
Pr(x|i) means the probability of x given i.
Pr(x & i) means the probability of x and i occuring together.
Pr(x) means the unconditional probability of x occuring.

Now let's solve the problem at hand:

Apriori:
Pr(g|H) = 0.6
Pr(b|H) = 0.4
Pr(g|L) = 0.1
Pr(b|L) = 0.9

Needed for later:
Pr(H & g) = Pr(H) * Pr(g|H) = 0.4 * 0.6 = 0.24
Pr(H & b) = Pr(H) * Pr(b|H) = 0.4 * 0.4 = 0.16
Pr(L & g) = Pr(L) * Pr(g|L) = 0.6 * 0.1 = 0.06
Pr(L & b) = Pr(L) * Pr(b|L) = 0.6 * 0.9 = 0.54
Pr(g) = Pr(H & g) + Pr(L & g) = 0.30
Pr(b) = Pr(H & b) + Pr(L & b) = 0.70

Aposteriori probabilities:
Pr(H|g) = Pr(H & g)/Pr(g) = 0.24/0.30 = 0.8
Pr(H|b) = Pr(H & b)/Pr(b) = 0.16/0.70 ~= 0.229
Pr(L|g) = Pr(L & g)/Pr(g) = 0.06/0.30 = 0.2
Pr(L|b) = Pr(L & b)/Pr(b) = 0.54/0.70 ~= 0.771

Optimal choices based on expected profits:
Apriori:
Marketing the product yields expected profits: 0.4 * 35 000 + 0.6 * 5 000 = 17 000
Not marketing yields: 15 000
Thus you choose to market apriori.

Aposteriori:
Expected profits if indication is g:
If marketing: Pr(H|g) * 35 000 + Pr(L|g) * 10 000 = 29 000
If not marketing: 15 000
You choose marketing.

Expected profits if indication is b:
If marketing: Pr(H|b) * 35 000 + Pr(L|b) * 15 000 = 11 870
If not marketing: 15 000
You choose not marketing.

Value of testing = Pr(g) * 29 000 + Pr(b) * 15 000 - 17 000 = 2 200

Ummmm...yeah. Sounds about right

knukk · 01-12-2010, 11:06 AM

Quote:

Originally Posted by SEO_Mike

Ummmm...yeah. Sounds about right

I added som explanatory notes:

Solution proposal assignment 7:

A quick intro to probability notation:
Pr(x|i) means the probability of x given i.
Pr(x & i) means the probability of x and i occuring together.
Pr(x) means the unconditional probability of x occuring.

Now let's solve the problem at hand:

Apriori:
A priori probabilities are the probabilities we have at hand before we conduct the PPC testing (from which we get a indication/response/signal/message):
Pr(g|H) = 0.6
Pr(b|H) = 0.4
Pr(g|L) = 0.1
Pr(b|L) = 0.9

So, e.g. Pr(g|H) is the probability of a good signal, given that the demand is high.

Needed for later:
Pr(H & g) = Pr(H) * Pr(g|H) = 0.4 * 0.6 = 0.24
Pr(H & b) = Pr(H) * Pr(b|H) = 0.4 * 0.4 = 0.16
Pr(L & g) = Pr(L) * Pr(g|L) = 0.6 * 0.1 = 0.06
Pr(L & b) = Pr(L) * Pr(b|L) = 0.6 * 0.9 = 0.54
Pr(g) = Pr(H & g) + Pr(L & g) = 0.30
Pr(b) = Pr(H & b) + Pr(L & b) = 0.70

Aposteriori probabilities:
These are the probabilities for a state of demand, given a signal response from the PPC testing. E.g., Pr(H|g) is the probability of high demand, given a good signal (g).
Pr(H|g) = Pr(H & g)/Pr(g) = 0.24/0.30 = 0.8
Pr(H|b) = Pr(H & b)/Pr(b) = 0.16/0.70 ~= 0.229
Pr(L|g) = Pr(L & g)/Pr(g) = 0.06/0.30 = 0.2
Pr(L|b) = Pr(L & b)/Pr(b) = 0.54/0.70 ~= 0.771

Optimal choices based on expected profits:
In this stage we figure out what our optimal choices are (to market or not to market), under our different possible situations. First we find out what our optimal responses are before we have a signal from testing, i.e. a priori.
Apriori:
Marketing the product yields expected profits: 0.4 * 35 000 + 0.6 * 5 000 = 17 000
Not marketing yields: 15 000
Thus you choose to market apriori, because it has higher expected profits.

Aposteriori:
This is where we check what actions are the most profitable if we conduct PPC testing and receive the different signals (good or bad indication).
Expected profits if indication is g:
If we choose marketing after a good signal from testing: Pr(H|g) * 35 000 + Pr(L|g) * 10 000 = 29 000
If not marketing: 15 000
You choose marketing.

Expected profits if indication is b:
If marketing after a bad indication: Pr(H|b) * 35 000 + Pr(L|b) * 15 000 = 11 870
If not marketing: 15 000
You choose not marketing.

Value of testing = What you are willing to spend to get the result = Difference in expected profits a posteriori and a priori = Probability of a good signal * Expected profits from our optimal choice given that signal + Probability of a bad signal * Expected profits from our optimal choice given that signal - expected profits from optimal choice a priori = Pr(g) * 29 000 + Pr(b) * 15 000 - 17 000 = 2 200

writeordie · 01-18-2010, 12:13 AM

Wow, I just starting in AM and I never really thought about all the factors I need to consider. This site is awesome thanks a lot!

SEOMERELIN · 02-04-2010, 06:22 AM

i will go for option b,its the best according to mathematics

darenet · 02-09-2010, 06:15 PM

Just get Marketing Experiments Fundamentals of Online Testing I learned all the math about this stuff from there...

And no it's not the same type of math you learn in school.

jbartinla · 02-28-2010, 01:06 PM

#7 is pretty slick, but you seem to be pulling a couple numbers our of the air in your equations like PPC spend, revenue and profitability.

knukk · 03-04-2010, 02:29 PM

Quote:

Originally Posted by jbartinla

#7 is pretty slick, but you seem to be pulling a couple numbers our of the air in your equations like PPC spend, revenue and profitability.

Whether I pull them out of the air or not is not important. In fact it doesn't change anything. The whole point of such an assignment is to understand the tools you have your disposal to solve for a solution suggestion. The idea is to provide a set of information and assumptions that can be applied to solving the problem at hand. In real life investment decisions projects are also evaluated beforehand and predictions are made to calculate the projects' values. You can think of #7 as what you do after you have predicted what the relevant variables' values will be.

knukk · 03-04-2010, 02:49 PM

Assignment 8

You are about to launch a project that will give $ 100 000 of revenues for sure.
To be able to generate revenues you need to invest in a technology.

Technology A costs $ 30 000 in present value over the years with a capital cost of 10 %. Technology B costs you $ 20 000 (present value) in total over the years with a capital cost of 15 %.

The project lasts 5 years. There are no other costs to consider. There are no taxes.

What are the equivalent annual cost for each of the two technologies? Which technology do you choose for your project and why?

View Poll Results: Which option?
Option A	190	23.51%
Option B	344	42.57%
Option C	107	13.24%
Just tell me, I'm too stupid to figure it out.	167	20.67%
Voters: 808. You may not vote on this poll

06-22-2009, 01:05 PM	#101 (permalink)
bigsherm Junior Member Join Date: Feb 2009 Posts: 17 iTrader: 0 / 0%	Assignment 3 You are in discussions with an advertiser for a special deal on affiliating their product. The advertiser pays you NET30 (30 days after the commission was registered). This means if you sent leads worth of $ 1000 in commissions one day, you are credited that amount 30 days later. Currently you send leads worth $ 40 000 in commissions every month. You offer this deal to the advertiser: "You can decrease my commissions by 1 % if you pay me NET10 from now on." Under what condition does he accept your offer? My Answer: I would assume the advertiser wants the potential to make more money than they're giving up. The 1% is really insignificant here, it's the 20 day difference in payments that affects the outcome. With 40k a month, that comes out to 1,333.33 a day (30 day month). With that daily income, we multiply it by 20 days = 26,667 and take out the 1% commission decrease to get $26,400 dollars you make in 20 days at the current, non-reinvested rate. Beat this and the advertiser should consider. Taking that 20 day total, you need to show the advertiser that you can reinvest your payments every 10 days to offset the risk they take by fronting the cash to you while they wait for their payment on the backend from customers, transaction cycle times, and processing which could be instant or have a "14 day trial" where they get the value back later than the NET10 you want. With that, we need a little bit more information about how each side is making the money, how the advertiser is getting paid and on what terms, and what kind of metrics you're getting on your investment as the affiliate, namely the ROI and how the speedy payments can grow past that 26,400 mark in 20 days. Assuming you get an ROI of say 35%, you're in full swing getting commissions every 10 days on the new terms and you reinvest 100% instead of renting a Lambo for the weekend... 10 x 1,333.33 = 13,333,33 1.35 = $18,000 (Note: you have two 10 day periods in the difference of payments to grow your return in hopes of surpassing the 30 day value that is already there, this does not count towards it as this is your starting value without the new NET10. In other words, you make this currently and begin to take advantage of NET10 the following week.) 18,000 x 1.35 = $24,300 (day 10 of 30days with NET10...the idea behind this is that you have added some value to the daily investments that, when compared with the previous 30 day total, will be of more value to the advertiser. That and it provides a certain % cushion against the risk the advertiser faces of non-payment and thus in fronting you the money; based on the assumptions here). 24,300 1.35 = $32,805 at day 30 vs the $26,400. With these calculations and assumptions, the benefit to the advertiser would be that when you reinvest your earnings, you create an additional 6,405 in revenues on top of the $26,400 or 24.5% return. If the advertiser has a greater risk than 24.5% of providing a faster payment (ie chargebacks and cancellations or just non-payments), it may not be a good idea based on the numbers. however, if their risk is that bad already they probably have a poor product. I hope that made sense. Share Share this post on Digg Del.icio.us Technorati Twitter

06-22-2009, 01:25 PM	#102 (permalink)
bigsherm Junior Member Join Date: Feb 2009 Posts: 17 iTrader: 0 / 0%	Assignment 4 You have a website that is budgeted to generate $ 27 000 in revenues pr year the next 4 years. Webhost costs are $ 1 000 pr year. Advertising costs are $ 2 000 pr year. You outsource the article writing for the website. You will put on 40 articles pr month, 25 of which are 600 words (per article). The rest are 1000 words per article. You will pay the article writer $4/100 words. It is expected that you can sell it for $ 25 000 in 4 years. Someone is offering you $ 50 000 for the website now. The rate of return is 10 %. What do you do - sell it now or keep it running and then sell it after 4 years? Answer: Fun little problem, good thing I have my financial calculator from college handy...I may be rusty so this could be off, but here it goes. The key to this problem is to get the NPV of your website when taking into consideration the cash flows of each year and the cash flow in year 4 + sale price. If it's more than 50,000 you hold it and sell. If it isn't, take the money and run. Assumptions here are that there are no other investments that the 50,000 now can be put into and that we just use the data here. This can definitely be done in Excel, but I had my calculator handy so bear with me. $27,000 RevenueYr - (3,000) Upkeep - (14,400) Content -(25x600)+(15x1000)= 30,000 Words/100 x $4 =$1200/month ______________ $9,600 profit a year. note: Multiplying $9,600 by 4 and adding the $25,000 doesn't give you the real value. The rate of return is going to be used to calculate the time value of money and take into account inflation, buying power, and the actual return. I: 10% CF1: 9600 CF2: 9600 CF3: 9600 CF4: 9600 + 25000 NPV: $57,106.04 So with my calculations, hoping they're right, you're better off holding it. Personally, I'd take the $50,000 and find a better investment like PPC, however the numbers as they are setup say holding is more profitable. If this is right, I guess I can almost justify spending over 100k for college Share Share this post on Digg Del.icio.us Technorati Twitter

06-22-2009, 01:33 PM	#103 (permalink)
bigsherm Junior Member Join Date: Feb 2009 Posts: 17 iTrader: 0 / 0%	Assignment 5 You are an email marketer. You have observed the following: Total emails sent pr week for 10 weeks: 49872, 52034, 47322, 56999, 48932, 43456, 53104, 50888, 45032, 49999 Total costs pr week for 10 weeks (the same weeks as above): 1500, 1743, 1388, 1844, 1599, 1333, 1900, 1011, 1803, 1800 You want to estimate a function for your total costs for a week as a function of sent emails that week. What is it? Answer: This can be done simply by finding the average of each week's cost per email and then average the 10 weeks for an overall average. I used Excel to get an average cost/email of $0.0320. # of Emails * $.0320 = Cost/Week Share Share this post on Digg Del.icio.us Technorati Twitter

06-22-2009, 05:05 PM	#105 (permalink)
bigsherm Junior Member Join Date: Feb 2009 Posts: 17 iTrader: 0 / 0%	Yeah, haven't used this calculator for a while now, looks like the interest was compounded 12 times a year as opposed to the 1 in the calculator. Oops. As for #3, I'm going to have to look at that one, the answer didn't come right to mind like the others. Obviously. Regression, I remember that from my stats classes...teacher was batshit crazy, maybe that's why I don't like it. Thanks for the input. Share Share this post on Digg Del.icio.us Technorati Twitter

06-23-2009, 02:19 AM	#106 (permalink)
keeper33 Senior Member Join Date: Jul 2008 Posts: 291 iTrader: 18 / 100%	That regression is funky, to say the least. The regression coefficient (r) is .3699606367, so we have a weak correlation. The coefficient of determination (r^2) is .1368708727, so 13.687% of the variation in y is accounted for by the regression of x on y, y being costs per week, and x being emails sent per week. Needless to say, stats is only 10% running the numbers, the other 90% lies just below the surface. Share Share this post on Digg Del.icio.us Technorati Twitter

06-27-2009, 05:47 PM	#108 (permalink)
iloveshi Member Join Date: Jun 2009 Posts: 32 iTrader: 1 / 100%	Great stuff. Will make use of the spreadsheet and calculator when analyzing offers. That's one less mistake (out of one thousand) that I'll make on my noob journey. +rep Share Share this post on Digg Del.icio.us Technorati Twitter

06-30-2009, 01:07 PM	#109 (permalink)
James P Junior Member Join Date: Jun 2009 Posts: 4 iTrader: 0 / 0%	Hi, Well, I just did what Henry Ford used to do. Rather than waste time working it ok, I just got someone more intelligent than me to tell me the answer.... Share Share this post on Digg Del.icio.us Technorati Twitter

06-30-2009, 10:57 PM	#110 (permalink)
bigsherm Junior Member Join Date: Feb 2009 Posts: 17 iTrader: 0 / 0%	Knukk, Can you give the breakdown of question 3? I'm curious as to what you're searching for here as I have no ideas other than calculating interest over extra 20 days in holding vs. increase in revenue with faster payment turnaround....if that is the case, I'll give it hell again. Share Share this post on Digg Del.icio.us Technorati Twitter

07-05-2009, 07:17 AM	#111 (permalink)
WFBenC Junior Member Join Date: Jul 2009 Posts: 12 iTrader: 0 / 0%	Option C has a claimed CTR of 0.1% Share Share this post on Digg Del.icio.us Technorati Twitter

07-05-2009, 07:19 AM	#112 (permalink)
WFBenC Junior Member Join Date: Jul 2009 Posts: 12 iTrader: 0 / 0%	Yeah, haven't used this calculator for a while now, looks like the interest was compounded 12 times a year as opposed to the 1 in the calculator. Oops. As for #3, I'm going to have to look at that one, the answer didn't come right to mind like the others. Obviously. Regression, I remember that from my stats classes...teacher was batshit crazy, maybe that's why I don't like it. Thanks for the input. Share Share this post on Digg Del.icio.us Technorati Twitter

07-05-2009, 11:08 AM	#113 (permalink)
Jokerr Mucho Macho Muchacho Join Date: Mar 2007 Posts: 363 iTrader: 12 / 100%	Just thought I'd share a very simple script. You set your estimated cpc, budget, payout and then the script calculates how many conversions you need to break even. You need to have PHP installed on your server, if you do, just upload it and the rest is pretty self explanatory. PHP Code: <?php if ($_POST['cpc'] && $_POST['budget'] && $_POST['payout']) { $cpc = $_POST['cpc']; $budget = $_POST['budget']; $payout = $_POST['payout']; $clicks = $budget/$cpc; $conversionsTBE = $budget/$payout; $conversionRateTBE = $conversionsTBE/$clicks; $conversionRateTBE = $conversionRateTBE*100; $data = "You can afford apr. ".floor($clicks)." clicks. You need apr. ".floor($conversionsTBE)." converions to break even, a conversion rate of <b>".substr($conversionRateTBE, 0,3)."%</b><br />"; } ?> <form method="post"> Esitmated CPC <input type="text" name="cpc" /> Your Budget <input type="text" name="budget" /> Offer Payout <input type="text" name="payout" /> <input type="submit" name="submit" value="Calculate!" /> </form> <?php if ($data) { echo '<div style="border: 1px solid #000; background: green; padding: 5px; color: #fff;">'.$data.'</div>'; } ?> Share Share this post on Digg Del.icio.us Technorati Twitter

08-09-2009, 02:15 AM	#115 (permalink)
Mkting Samurai Napoleon had my complex Join Date: Aug 2009 Location: Nor Cal Posts: 59 iTrader: 0 / 0%	Thanks for the both the spread sheet and the script ... The breakdown on question 3 makes more sense now ... I think Thanks again Share Share this post on Digg Del.icio.us Technorati Twitter

08-10-2009, 04:05 PM	#116 (permalink)
brucep Extreme Newbie Join Date: Aug 2009 Location: Planet Earth Posts: 4 iTrader: 0 / 0%	Oh, before I go, here's the spreadsheet I have open 24/7 for quick calculations. If you're not using something like this, take mine, or build your own. If you don't understand it at first, just look at the calculations I'm using. The colored cells are the only ones you need to change. Attachment 675[/QUOTE] Thanks for the spreadsheet it will be a great help!!! Also good to find a link where I do not get DICK ROLLED. Share Share this post on Digg Del.icio.us Technorati Twitter

09-01-2009, 10:25 AM	#117 (permalink)
okuku Junior Member Join Date: Aug 2009 Location: the other world Posts: 9 iTrader: 0 / 0%	The spreadsheet looks cool. Though I don't know what the hell I'm gonna use it for yet. I believe I will soon know wit you guy's help. Share Share this post on Digg Del.icio.us Technorati Twitter

09-12-2009, 03:27 AM	#118 (permalink)
PhillipMarlow Pheasant Heavy Breathing Join Date: Mar 2008 Posts: 1,771 iTrader: 7 / 100%	Fuck yeah! Was about to whip one of these together...this saved me hours. Thanks +rep Share Share this post on Digg Del.icio.us Technorati Twitter __________________ Best Article I've ever read online

09-14-2009, 09:58 PM	#119 (permalink)
imouttehre Senior Member Join Date: Sep 2009 Posts: 118 iTrader: 1 / 100%	Again another thing that makes me feel like a total dumbass.. Thanks for sharing the info and the questions guys. No PPC course or gurus teach this kind of practical/complicated/much needed stuff. Mike, thanks for the spread sheet. Can we get some more scenarios?? The cost valuation question for the website was really cool. With a small biz background, i would have jumped on the $50k offer. Thanks again Share Share this post on Digg Del.icio.us Technorati Twitter

09-23-2009, 11:42 AM	#120 (permalink)
orange Junior Member Join Date: Jul 2009 Posts: 9 iTrader: 0 / 0%	wooh good thread! tnxx so much, had a good time while reading -) Share Share this post on Digg Del.icio.us Technorati Twitter

09-24-2009, 09:25 PM	#121 (permalink)
stevo7 Check Six Join Date: Sep 2009 Posts: 21 iTrader: 0 / 0%	Much thx for the spreadsheet, Mike! Mucho useful. Share Share this post on Digg Del.icio.us Technorati Twitter

09-24-2009, 09:30 PM	#122 (permalink)
theemerchant Junior Member Join Date: Sep 2009 Posts: 18 iTrader: 0 / 0%	Cant Wait I can't wait for the answer. I'll go for B if it was my decision. Share Share this post on Digg Del.icio.us Technorati Twitter

09-30-2009, 10:41 AM	#124 (permalink)
balright Junior Member Join Date: Sep 2009 Posts: 7 iTrader: 0 / 0%	Damn shame my math major doesn't help me write content. Who wants to market on a math site? Anyone? Oh. Share Share this post on Digg Del.icio.us Technorati Twitter

10-02-2009, 03:36 PM	#125 (permalink)
Carpaccio Member Join Date: Oct 2009 Posts: 42 iTrader: 0 / 0%	Newb and math retard over here. Thank you, this is absolutely awesome. Share Share this post on Digg Del.icio.us Technorati Twitter

10-15-2009, 04:30 PM	#126 (permalink)
whatyaknow Junior Member Join Date: Apr 2009 Posts: 19 iTrader: 0 / 0%	Awesome thread Mike and really appreciate the xls. Anyone else have worksheets they use for calculations? Share Share this post on Digg Del.icio.us Technorati Twitter

10-22-2009, 02:51 PM	#127 (permalink)
dhamma Senior Member Join Date: Dec 2008 Posts: 148 iTrader: 0 / 0%	Excellent examples - and xls. Thanks Share Share this post on Digg Del.icio.us Technorati Twitter

10-27-2009, 08:13 AM	#128 (permalink)
Harsh Junior Member Join Date: Oct 2009 Posts: 27 iTrader: 0 / 0%	i justed click on A but i saw the meaning of A after voting Share Share this post on Digg Del.icio.us Technorati Twitter

10-27-2009, 06:46 PM	#129 (permalink)
bloodsucker Junior Member Join Date: Oct 2009 Posts: 1 iTrader: 0 / 0%	thank s God, i love math. I m more in peace.. Share Share this post on Digg Del.icio.us Technorati Twitter

11-16-2009, 02:37 AM	#130 (permalink)
Mike Join Date: Jun 2006 Location: marvelously gleeful Posts: 6,024 iTrader: 30 / 100%	Since I posted that original XLS spreadsheet, I've made a slight change to the CPM column by adding "Country". "Country" on my spreadsheet is the percentage of traffic that comes from countries that I can theoretically convert. So, if I've got an offer that is US only, then I need to know what percent of their traffic is US. Obviously, if you are targeting US only then the percent is 100% (duh), but a lot of sites you'll find can't geotarget traffic for you, so then you'll need to enter the percent of traffic you can actually convert. Also, I added "Off Page CTR" and changed the conversion rate so that it's based on the clicks off page, and not total clicks. Just play around with it, you'll figure it out. cpc-calculator.zip Share Share this post on Digg Del.icio.us Technorati Twitter __________________ ΜΟΛΩΝ ΛΑΒE

11-26-2009, 05:42 PM	#132 (permalink)
metoo Senior Member Join Date: Mar 2008 Posts: 228 iTrader: 2 / 100%	I'll answer from just reading the 1st post, cos I know the rest of the thread will be filled with the answers... You make 80 cents per click, so: A: $1000 B: $2000 C: Take it if the CTR for the banner is > then 0.25 I'll leave this here with no edits. If I screwed it up, I figure who gives a shit. At least I had a go. My poor calculator, hehe Share Share this post on Digg Del.icio.us Technorati Twitter

11-26-2009, 09:03 PM	#133 (permalink)
invinceable Member Join Date: Nov 2008 Posts: 73 iTrader: 0 / 0%	Thank you very much for the spread sheet, looks very nice! Share Share this post on Digg Del.icio.us Technorati Twitter

12-11-2009, 02:31 AM	#135 (permalink)
Expert Senior Member Join Date: Mar 2009 Posts: 102 iTrader: 0 / 0%	I really use these when dealing with my business models based on recurring income. You have to incorporate attrtion rate into the equation though. Share Share this post on Digg Del.icio.us Technorati Twitter

12-23-2009, 05:01 PM	#137 (permalink)
IcanWrite I am Lost Join Date: Dec 2009 Posts: 67 iTrader: 8 / 100%	Nice job on these. The spreadsheets seem really useful. I love math though haha hell yeah! i got it right!! Share Share this post on Digg Del.icio.us Technorati Twitter

01-07-2010, 07:21 PM	#140 (permalink)
papajohn56 Join Date: Jun 2008 Location: SC Posts: 6,673 iTrader: 16 / 100%	There's a lot more advanced math that you can apply to affiliate marketing, too. If you're good with statistical analysis and calculus, there's methods for it. I'll see what I can come up with to post here for more info. I love math Share Share this post on Digg Del.icio.us Technorati Twitter __________________ Twitter

01-11-2010, 06:04 PM	#142 (permalink)
imouttehre Senior Member Join Date: Sep 2009 Posts: 118 iTrader: 1 / 100%	a request for more cpm problems? Share Share this post on Digg Del.icio.us Technorati Twitter

01-18-2010, 12:13 AM	#145 (permalink)
writeordie writer extraordinair Join Date: Jan 2010 Posts: 16 iTrader: 5 / 100%	Wow, I just starting in AM and I never really thought about all the factors I need to consider. This site is awesome thanks a lot! Share Share this post on Digg Del.icio.us Technorati Twitter

02-04-2010, 06:22 AM	#146 (permalink)
SEOMERELIN Senior Member Join Date: Feb 2010 Posts: 375 iTrader: 29 / 97%	i will go for option b,its the best according to mathematics Share Share this post on Digg Del.icio.us Technorati Twitter