Integrity is the practice of being honest and showing a consistent and uncompromising adherence to strong moral and ethical principles and values. In ethics, integrity is regarded as the honesty and truthfulness or accuracy of one's actions. Wikipedia
Revised: July 17, 2017
Price Integrity Rank Quality Scale
How to interpret the Price Integrity Rank
How good is the sale?
00-20 Not So Good
41-60 Not Too Bad
61-80 Pretty Good
It Should Be So Simple
It should be so simple to calculate a discount. Just take the MSRP ( the Manufactures Suggested Retail Price ) for an item, subtract from that the Sale Price, and then divide the difference by the MSRP and multiply by 100. That would be the Discount (%).
With the rapid fire automated price adjustments used by so many online stores the concept of a Sale, a Discount, or Savings has become largely meaningless. Our prime directive is saving you some money today. In order for our website to function properly we must have a method for determining the best deals, the best prices - so we have invented a method for calculating a Price Integrity Ranking Factor without using those now meaningless terms. We have developed a proprietary algorithm for calculating a Price Integrity Ranking Factor which allows us to determine the best deals regardless the pricing shenanigans of the merchants.
It could be so simple, but with a bazillion products being promoted online and with many of them having a dozen or more different sizes in any number of colors or styles it's difficult, if not impossible, to keep track of all of those numbers. So even for those merchants trying to do it right, it's easy to mess it up. And then, of course, some merchants just make up numbers that make it look like you're getting a good discount when actually you are not.
How To Calculate A Ranking Factor
In order to calculate the actual bona-fide sale price, as opposed to a mistaken or made up one, we keep track of the prices actually quoted for an item that is available for sale. Then as we build up a price history for the product, over time, we look to the prices actually quoted on the merchants web site to see what they are really charging for an item. From this price history we calculate the merchants established price for the product, even if they change that price several times a day.
A merchant may offer an item for, say $100 without any discount, so the discount to start off with is Zero. Then later they may put that item On Sale and change the Sale Price to say $80. That would be a Discount of $20, which is 20% off the original Listed Price. At least that how it's supposed to work in that ideal world that doesn't really exist. It's a bit more complicated than that. Continue reading to find out just how complicated it really is.
Why Do Our Numbers Differ From The Merchants Numbers?
After a month or two, or when the seasons change, the merchant may put an item that was listed for $100 on the Clearance Rack and whoever is punching the numbers into the merchants database might put it in as say $40 with an MSRP of $40 TOO!! That certainly isn't the actual MSRP, but the kid (or the computer) punching in the numbers didn't know. It looks as if the $40 sale price is the established price and there is no discount at all. Since we have maintained a price history we know that there is actually a substantial savings on that item. It's the real discount that we use to rank that item in our reports, even though the online store may say something different. That is one reason our numbers may be different. There are more.
In other cases a merchant may put an item listed for $100 on sale for $80 and then changed the list price to $120, as if that list price was the established price. That`s a bit dishonest, and could actually get the merchant in trouble. We will only report to you a ranking factor based on the actual established price, not a made up one. That is another reason our numbers may be different from the merchants numbers, but there are even more.
An increasing number of websites employ AI (artificial intelligence) which will move prices up and down on an almost continuous basis. The merchants website, if they have tracking information that was collected from your browser, might quote a different price to you than they would quote to your next door neighbor. This is called dynamic pricing. In some cases two or more merchants employing AI will engage in what is called a 'race to the bottom' where they check each others prices and try to be the lowest one.
With all these rapid fire ups and downs we had to create a new method for estimating the merchants established price and we do that with a price averaging system based on the RMS (Root - Mean - Square) value of the price history for an item. It is the RMS price that we use to calculate the Ranking Factor for an item.
What Is A Good Discount?
Most of the time a discount is listed as a percentage off a previous price. In other cases the merchant declares that "you save" a certain amount. All products are not the same, however, and a 5 percent discount on one category of products may be a outstanding deal, while a 5 percent discount on some other category of products may be a lousy discount.
The value of a discount has to take into account several factors. They are the labor component, the material component, and the processing component that determine the manufacturing cost of an item.
Apparel for example has a relatively low labor, material, and processing components. For apparel end of season discounts can easily be in the 50 to 80 percent range.
For items such as cameras, appliances, or computers the labor, material and processing components are much higher than products such as apparel. In many cases the merchant would hardly be able to offer a 10 percent discount before loosing money on the item. Of course, sometimes a merchant will deliberately loose money on an item just to get you in the store, but those items are of limited quantity.
In order to level the playing field our Price Integrity Rank algorithm takes into account the category of the items so as to rank them as 'the best in category'. That way a high priced camera doesn't have to compete with a inexpensive bikini in order to get a good Price Integrity Rank.
We Keep A List For You
Our Price Stalker robots are checking the prices on all our cataloged items all day long, every day, and making notes of the price changes. We keep a record of these price changes and provide a variety of lists for you to view.
You can see that list here :
We also maintain our HOTLIST, which sorts the items by the Ranking Factor, showing you the best prices available on our cataloged items.
For those who want to shop for a specific brand, our SHOP BY BRAND page is made for you.
When you are shopping for a number of items, from different stores, the multiple shipping costs can totally destroy the savings.
For those who want to shop from just one store, our SHOP BY DOMAIN page is what you need.
As you scroll down through the many items listed ( there are thousands of them that we keep track of for you ) you may see the link Show Price History. This link will show you the price history for that specific item in graphical form. You won't see this on every item, just those where the price has been reduced. For items that have a stable price that has not changed ( since we began recording it ) we do not provide the link since the price history graph is just a straight line, which isn't very interesting. For prices that have only gone up we do not show the price history because the Ranking Factor is Zero.
How Timely Is Our Data?
Our robots can't check every price every second. That would kill our bandwidth and the merchants bandwidth too. So instead we check each item every 8 hours or so which is about 3 times a day. We don't check them as a batch but spread them out over the day so as not to overload the servers.
We calculate the Ranking Factors and update our price histories each hour. So there may be a delay of several hours before we learn of a price change, but throughout the day we'll catch up and show the correct prices.
If you need data faster than that you will have to monitor the merchant yourself, but since we monitor all the items while you are sleeping, working, driving and eating, you're likely going to find that we have found the best prices faster than you could have.
How Long Is A Sale A Sale?
We strive to provide data that is useful to the serious shopper. If you're shopping for an item on Labor Day does it really matter what the price was on Memorial Day? That`s so last season! The value of a sale diminishes over time and after a while that sale price is just the regular price. How long is a while? For our purposes a while is 90 days. That is about how long a season lasts and consumer prices are typically seasonal.
If you're shopping for Easter clothing does it really matter what the price of that outfit was last Christmas? If your shopping for a Halloween costume does it really matter what the costume cost on the 4th of July? For serious shoppers a sale is only a sale if the price was reduced recently. The price 3 months ago is ancient history. Our Pricing Factors are calculated based on the price history over the last 90 days, so they may differ significantly from what the merchant claims is their discount based on a previous price that is weeks or months old.
It takes about 10 days to gather a price history sufficient to calculate a pricing factor. For a recently added item we will start reporting a Price Integrity Rank after about 10 days. At 90 days we start deleting the data older than 90 days so that our data window becomes a 90 day sliding window. Price history older than 90 days is not considered when calculating the pricing factor.
The Games People Play
One would think that the price of an item would change from time to time, such as a seasonal sale, a clearance sale, etc. Prices typically should start out near the MSRP and then go down over time, until the item is sold out.
What should be, though, ain't always so. Prices go up and down sporadically, erratically, and for no apparent reason. Sometimes the merchant will play around with prices to see if they can stimulate a sale. Most often these days it is AI ( Artificial Intelligence ) software that tries to predict the optimum price to maximize sales. AI may change the price several times a day. In some cases a merchant may offer you a different price than your neighbor down the street because they have purchased internet tracking data and think they know how much you, and your neighbor, would be willing to pay. When the merchant or AI software changed prices often, the price history data looks really crazy, because for the most part, it's just crazy. The only true intelligence was the salesman that sold the merchant that ridiculous AI software package. Read our report MSRP: Crime or Cluster Fxxx for more information about this subject.
We post our price history data for you to see, and you might ask why would we post some dumb looking crazy plots. We do it because if you look carefully there is some interesting and useful information in them. Here are some examples:
Now The Complicated Stuff
If you hate math, stop reading right now!
Ok, you're still here. We stated earlier that our discount calculations will be different than the merchants. In fact, they're really - really different in many cases and here is why.
Looking at the wiggly jiggly price fluctuations it's impossible to figure out what a merchant really is charging for an item. With all the ups and downs should we be looking at the highs and lows? Should we use a median value, or the average value? What best represents a reasonable assesment of what a merchants established price for an item actually is?
Many States and Federal governments have setup rules and guidelines about what is a bona-fide sale price, and what is deceptive. One of the terms they use is "reasonably substantial period of time". They say that a merchant calculating a discount based on their "regular price", that "regular" price must have been offered, with the intent to sell items at that price, for a "reasonably substantial period of time". They neglect to say, however, just what a "reasonably substantial period of time" actually is.
In a number of court cases the courts attempted to define the terms "reasonable" and "substantial" but failed to do so in a meaningful way. Opinions varied from State to State and region to region. Some States mandated that the "regular" price must have been the "regular" price for at least 55% of the time during a 90 day period. Other regulators indicate that substantial sales must have been made at that "regular" price, although the US FTC says that no sales at the "regular" price does not necessarily make it deceptive.
Who are we going believe? Which opinions are the valid ones? With these flibbity jibbity terms being tossed around without any hard definitive indicators as to what it all means, how do we know? These opinions are no more valuable than the wiggly jiggly data in our charts. When prices go up and down minute by minute, or change depending on who's looking at them (so called dynamic pricing) the legal terms such as "reasonably" and "substantial" lose all meaning. There has to be a better way.
The Has To Be A Better Way
In the absence of anything definitive (they leave it all up to subjective analysis) we have come to the conclusion that the best way to determine the "regular" price is to calculate the RMS value of the fluctuating prices over time. RMS means that we take the square root (R) of the average value (M) of the squares (S) of all the individual data points. the "RMS value" is very close to the "average value" but we like it better because in our analysis it better represents a "regular" price when there is a substantial number of high or low values for a substantial period of time. (there are those words again)
Given all this, therefore, our calculation of a "Price Integrity Rank" will be based on the current advertised price of an item compared to the RMS value of the prices listed for the last 90 days, or less if we don't have 90 days worth of data. We do need at least 10 days of collected price data to calculate RMS so items that have been in our catalog for less than 10 days won't have any Price Integrity Rank calculated.
To avoid any confusion we no longer refer to the values as a "Sale Price" or a "Discount" or list an amount that "You Save". Instead we report our Ranking Factor (Price Integrity Rank) and use that to determine which items have the best prices, right now, not months ago.
A Sale Is Best When Served Fresh
As we discussed in a previous paragraph a sale isn't a sale forever. After 3 months or so it becomes irrelevant what the price used to be. The regulators and the courts seem to agree in that they put limits on how long a sale can be considered a sale before it becomes just the regular price.
Our price algorithm handles this in a special, and we think a very elegant way. If an item with a long term established price is put on sale, say at a 20% discount, our algorithm will calculate a Price Integrity Ranking factor of 20, for awhile when the discount is fresh. As time moves forward the value of the sale diminishes, it fades away. The sale value, the displayed Ranking Factor, will slowly fade away over a 90 day period. We think this is more realistic than saying that you suddenly have a 20% discount one day, and then day after day for 90 days you still have a 20% discount, and then boom, the discount goes to zero. Life is more graceful than that, and so is our algorithm.
Due to this algorithmic aging process if there are two items at the same price that were reduced the same amount, the freshest one, the most recently changed one, will rank higher than the older one.
If It's Broke, Vix It
Our term "wiggly jiggly" to describe the price fluctuations shown in our charts is descriptive, and cute, but we need a better way to determine just how "wiggly jiggly" our data sets really are. There is another group out there with wiggly data, called the stock market. The stock market uses a method of calculating the volatility of a data set called the "Vix" (the volatility index). We too, have a Vix, although our equation is much simpler, and if we do say so our self, more elegant. Our Vix ranges in a manner similar to the stock market Vix and is a measure of just how wiggly jiggly our data really is.
Why do we want to quantify the jiggliness? It's a subjective analysis, but it's also a measure of the integrity of the merchant. The really jiggly stuff makes you wonder just what the merchant is up to. Do they really want to make a sale at a reasonable price or are they playing games hoping to wring every last penny out of your purse whether they deserve it or not. Wiggly data doesn't necessarily indicate a dishonest merchant, but it's a warning indicator to be on your guard when purchasing from them.
Is A High Vix Always A Bad Thing?
A High Vix isn't always a bad thing. Price volatility can also reveal a pattern that a shopper can take advantage of. Looking at the chart of the price fluctuation may indeed reveal a good time to buy. If the item you are interested in has many ups and downs, and you can predict when the next down might be, then you are in a good position to purchase that item at a rock bottom price.
Why Do We Do This?
We do this as a public service and because we find keeping track of such things is an interesting project.
We promote on our pages the most popular items at the best prices that we can find. In some cases we have no relationship with the merchant. In some other cases we may have an affiliate marketing relationship with the merchant and we will get a small ( very small ) commission if you purchase a product after having clicked through to the merchant from one of our pages. Our algorithm calculates everything the same way regardless if we have a relationship with the merchant or not.
In all cases we are looking for the best bargains on the best products out there in Cyber Space.
If there is a brand or a specific product that you would like us to track let us know. Leave us a comment below.