Product Feeds are essential in today’s online marketing activities and can be used for anything from powering your Pay Per Click (PPC) activities through to selling your product through 3rd parties like Amazon but in a time where getting your feeds right is so important it’s shocking just how many companies are getting it wrong.
Product feeds can have a wide range of different data set in them and although there have been a number of attempts to unify the data within product feeds so far all these attempts have been mostly unsuccessful and it is easy to see why. With so many industries all using feeds, trying to agree on a “uniform” feed is near to impossible when each industry requires different data. Perhaps a better way of addressing this would to start looking at the unique fields each industry needs and create individual industry stand data sets.
When looking at data feeds there are a lot of variables that need to be taken into account, from the format you use, which would normally be CSV, XML or API, through to what fields your need to include and in some cases even how many feeds are needed. To get this right the first thing you need to decide is who is going to use your feed and what they are going to be using the feed for. Once you know this you’ll be able to start planning your feed.
The 4 first steps in the production of data feeds:
• Planning:- Plan ahead when you are setting up your feed systems and tools, because there are very few companies out there that use the same feed as each other, creating a tool that lets allows you to create new feeds with different fields and then host it automatically on a URL is a must
• Technical:- The technical requirements of products feed are fairly simply until something goes wrong, make sure you’ve setup some basic monitoring system so when your feed goes down or doesn’t update correctly you’re not the last to know
• Data:- make sure your product details are correct and up to date, this includes not just price but also category mapping of products. You are going to need to make an investment into your product feed to really make the most from it, but the rewards are well worth the investment when you get it right
• Scalability:- Make sure you are looking to the future, producing a feed that can handle a few downloads every day is quite a simple task, but as your companies grow so will the strain on your data feeds, you may only start with a handful of companies taking your feed, but as your company grows so will the number of companies using your feed and when you move to an API the strain will increase even more
From the very beginning you’ll need to make sure you’ve setup some type of conversion tracking for your feed, there are a lot of different companies all charging on different metrics that use feeds, you are going to need to take this into account when looking at what products you make available in your feeds. The two main costing models are CPC (cost per click) and CPA (cost per acquisition).
What should you take into account with:
• CPC Feed:- Most commonly this applies to price comparison sites like Ciao, Kelkoo or Shopping.com where you are being charged each time a user clicks through to your site, bear in mind there is not a guaranteed sale from the clicks you are being charged for. You can still monetize these channels by making sure you are tracking the click to sale process and to do this correctly you are going to need a reliable solution, i.e. beyond Google analytics, with this tracking you can then start to really monitor which products are producing a poor ROI and take action to rectify this.
• CPA Feed:- This normally refers to affiliate sites like twenga and Idealo. Because you only pay out if a sale occurs, you do not need to worry as much about filtering out products. However, keeping your CPA partner informed of what is converting and including details like stock quantity will definitely help your relationship with your affiliate partners. Keeping your CPA partners informed will enable them to decide whether to promote a product that you are currently low on or out of stock or push the highest converting products to their users.
Feeds can be used for a lot more than simply displaying a product from an eCommerce store on different web sites, feeds can be used to share all types of information like analytical data from tools like Google Analytics.
Product feeds are definitely here to stay and it is about time companies accepted this and looked at ways to improve both the quality and reliability of their feeds. In a world where consumers are now so used to getting the information they want when they want it, can any company really afford old, out of date data displayed or worse yet just simply not be include in results?
A quick guide to understanding feed formats
CSV (comma separated values): Although CSV are probably the most commonly available types of feeds, there are a number of flaws with them by definition, a CSV file i.e. a comma-separated value file which in is where the issues lie. Using a comma as the separating character, when one of your values is text based e.g. like a description, it is going to cause problems as many descriptions use commas, this means that each comma in the description is going to fragment it and in turn break the importing systems mapping of the feed by adding too many columns. There are a number of ways around this, for example using a pipe character as your separator rather than a comma, as this is pretty much never used in any description text. Despite a relatively simple solution to the problem, companies seem to be missing a trick.
Other issues that you may experience, but are not quite as easy to fix, are UPC (unique product codes) in your feed that starts with zero, which quite a large number of UPC’s do, automated system will often automatically reformat the number to remove the zero which changes the unique product code.
So with so many issue being found in a CSV feed why is this such a popular format?
The advantage of a CSV really comes from the fact that it’s such an easy format for everyone to use (except perhaps the techies), a simplified way of looking at a CSV file is as a spreadsheet where each product is a row and each column is a value, in fact programmes like excel even let you import and export CSV files and this feature is what helps make them so widely used, it’s allows a person of nearly no technical ability to use a feed.
XML: XML feeds are starting to take some ground and some companies like Trade Doubler are moving completely away from CSV file in favour of XML. From a technical standpoint it’s easy to see why they would want to do this, after seeing the issues with CSV files above, however XML feeds are no walk in the park either.
Unlike a CSV file an XML file doesn’t really have any “normal in life comparison” like a spread sheet is to the CSV file, to demonstrate this here’s an example of a XML feed
<– XML feed example (supplied by w3schools.com) –>
<?xml version=”1.0″ encoding=”ISO-8859-1″?>
<shiporder orderid=”889923″
xmlns:xsi=”http://www.w3.org/2001/XMLSchema-instance”
xsi:noNamespaceSchemaLocation=”shiporder.xsd”>
<orderperson>John Smith</orderperson>
<shipto>
<name>Ola Nordmann</name>
<address>Langgt 23</address>
<city>4000 Stavanger</city>
<country>Norway</country>
</shipto>
<item>
<title>Empire Burlesque</title>
<note>Special Edition</note>
<quantity>1</quantity>
<price>10.90</price>
</item>
<item>
<title>Hide your heart</title>
<quantity>1</quantity>
<price>9.90</price>
</item>
</shiporder>
<– XML feed example (supplied by w3schools.com) –>
XML can also be written in a number of ways these “styles” of writing a XML feed are known as a Schema and whenever you are looking to setup a XML feed, it’s always best to request the schema from the company you are creating the feed for. The clear advantage of a XML feed over a CSV is that once its setup there is normally very little that goes wrong with the formatting, but you can still run into problems with updating and loads(I understand what you mean by loads but will other people do?) if you’ve not planned ahead for this.
API / WebServices: there is a lot of talk about API feeds and web services being the way to go and in our opinion it’s true (for the most part). So first of all what is an API?
At a basic level an API is a way for two applications (or websites) to talk to each other and we don’t mean they have a good chin wag and catch up, it simply allows one site or application to ask the other a question and then in turn have this question answered so probably one of the best known examples of this in action is a price comparison site like Kayak. Kayak’s flight search, known as a meta search in the industry, use API’s to find the cheapest priced flights and in this example the question it asks to all the different airlines “do you fly from XX to XX” and the airlines would reply with a yes or no and if it’s a yes the airline passes the price and URL to book that specific flight.
Ok so we’ll admit this is a very simplified way of thinking about it, but ignoring the technical side this is actually quite a good way to think of an API.
So why is this a better system? Well one of them most common problems with a standard XML or a CSV feed is that the data is at best 24 hours old, but more often than not the data is a lot older still, believe it or not but we’ve seen cases where the feed has not been updated in 6 months, just think how many prices have changed or products have been added or removed in the last 6 months? Would you really want customers seeing product prices or products you stopped selling 6 months ago?
API’s also mean things like stock availability can be taken into account as the API’s response (answer) to a request (question) could include stock information and delivery time i.e. out of stock, but expect back in stock in XX days’ time.
So to summarise the advantage of an API feed is up to date information and in a world where current up to date information is expected by consumers I’d expect to see APIs used a lot more often.
