Long time readers are familiar with the REDUCE function that lets you “roll up” child item (or variation) data to send in a Comparison Shopping Engine feed. For example, you can write a business rule that will produce a comma-separated list of all the available sizes for a given parent item.

Somebody here at work pointed out to me that the problem is that parent items often vary by color AND size. For example, here’s an item with 4 variations:

If I were to write a REDUCE rule to get the colors, it might look like this:
REDUCE(FUNCTION(VARS(@X,@Y),JOIN("~",@X,@Y)),$CHILDREN.COLOR,"")

and the result would look like this:
~Red~Blue~Red~Yellow

I’ve got “Red” in there twice, which looks kind of silly. What I really want is the distinct color values. So, of course, I wrote a new function.

The function is called DEDUP, and it has two parameters:

• The text to operate on
• A delimiter

First, let’s be clear on what DEDUP does NOT do. It does not remove duplicate items from your feeds. Instead, you give DEDUP some kind of delimited list (note: NOT a lookup list, but a single piece of text), and tell it what the delimiter is, and it will find the distinct items in that list, order them alphabetically, and give it back as a delimited list again. Here’s an example:
DEDUP("~Red~Blue~Red~Yellow","~")

and the output:
~Blue~Red~Yellow

Of course, we wouldn’t really put a piece of text like “~Red~Blue~Red~Yellow” right in the rule; we’d use the REDUCE rule we already wrote:
DEDUP(REDUCE(FUNCTION(VARS(@X,@Y),JOIN("~",@X,@Y)),
$CHILDREN.COLOR,""),"~")

Oh wait: I don’t want the output to start with “~”. So we need another REGEXREPLACE to rip that tilde off the beginning of the output:
REGEXREPLACE(DEDUP(REDUCE(FUNCTION(VARS(@X,@Y),
JOIN("~",@X,@Y)),$CHILDREN.COLOR,""),"~"),"^~","")

and the output:
Blue~Red~Yellow

This brings my total to only 5 posts for this whole year. Anybody want to take bets on when I’ll write another one?

Share This

ChannelAdvisor’s second annual holiday catalog is now available. Shop online and save up to 30% with with discounts from more than 40 ChannelAdvisor retailers!

www.channeladvisor.com/holidaycatalog

Share This

By Simon Besley

So, as happens every year, Q4 ‘Silly Season’ and the question of ‘The Busiest Day’ rears its head and online retailers scramble to ensure that their products are in stock, keenly-priced and made available to as many online shoppers as possible.

This year will be no different and, if we’re to believe the IMRG Capgemini e-Retail Sales Index – which showed that UK shoppers spent over £4.67 billion online in December 2008 (the equivalent of £76.67p for every person in the UK, and a 14% increase on December 2007) – the recession is driving shoppers online more than ever.

Typically, most people shop online (from the comfort of their office chairs) on a Monday, no doubt taking advantage of PC access and quicker internet connections, not to mention the attraction of a leisurely start to the working week.

Indeed, 2008 saw Monday, 8th December as the peak online shopping day – compared to Saturday, 20th December on the high-street, according to Visa – and the IMRG also showed that most business was done between 1pm and 2pm.

My uncomplicated prediction for the killer date in your diary this year?

Monday December 7th.

Obviously, there’s no hard and fast rule to this (although I’m sure the Post Office will play its usual part in swaying consumer confidence), but consumers are well aware of the dangers of ordering those Christmas presents too late in the day, and are affording themselves more of a buffer.

With that in mind, here’s a quick checklist for our ChannelAdvisor clients to optimise that holiday season traffic:

-Stock levels clearly shown
-Images clear and showing correctly
-Any promotions (shipping, discounts etc.) repeated throughout the site

Worth also noting is that Amazon has (from the 13th October until 1st January) abolished shipping costs, something that resonates strongly with consumers, even if that cost has been absorbed into the product price.

I’d be interested to know whether you have your own take on the festive season and its opportunities, and if some of you have actually – historically – seen a different consumer buying pattern in the run-up to Christmas.
–
Simon Besley is a Senior Account Manager on the UK ChannelAdvisor team where he develops eCommerce strategy and growth for several leading online retailers. He has worked in the online retail environment since 2002 and previously held positions with Dealtime/Shopping.com and Webloyalty.

Share This

Today, we’re releasing an enhanced next-generation Amazon solution that makes it easier than ever before to get exposure to, target and acquire your share of Amazon’s 95 million active buyers.

We built this to offer two categories of benefits: saving time and generating more sales.

SAVE TIME

• Quick, flexible launch - get live on Amazon within 10 minutes
• Real-time dashboards that for the first time give you a dynamic, birds-eye view of your listings’ statuses
• Product Match: proprietary technology that automatically detects duplicate product listings and gives you the information you need to list onto the best possible listing

SELL MORE

• Fulfillment by Amazon support that allows you to centralize all your product data, regardless of fulfillment center
• Out-of-catalog product creation that lets you add products that don’t exist on Amazon.com within the application
• Technology that identifies listings to be optimized so they appear higher in search results and get more potential sales

We’re excited about this new opportunity to get your business selling more on Amazon. You can find more information here, or contact scott.hurff (at) channeladvisor (dot) com.

Share This

And the only cure is a regular expression!

Today Paul in the UK tossed me this head-scratcher:

I have a set of inventory with parent/child variations, but the offername fields of the child variations are all the same. At the very start of the offerdescription is a set of parentheses containing the child variations, so the description might be “(size 6, grey color) These golf shoes are really great…”

I’d like to pull the “(size 6, grey color)” but to complicate matters their may be an other set of parentheses within this, e.g “(size 6 (for 12 year olds), grey color ) These golf shoes are really great…”

Also, some of the descriptions have brackets elsewhere in the description, e.g. “(size 6 (for 12 year olds), grey color ) These golf shoes are really great (comfortable, stylish)….” I only want what’s in the first set.

Here’s the rule we came up with:
REGEXGET($offerdescription,"\([^)]+(\([^)]+\))?[^(]+\)")

I hereby declare this the winner of the world’s most ridiculous regular expression.

Share This