Monday, 24 September 2012

And Finally: Quarterlies. Accessing Quarterly XBRL in Excel

A bit later than planned but finally you can now select XBRL 10Q filing data in XBRL to XL and analyse it in Excel.

I have created a new template for quarterly data, which automatically gets selected for download if the 1st filing you choose is a 10Q. This has been designed to make the most of both the values for the most recent quarter and the Year To Date (YTD) numbers.

In a quarterly report, a company will in effect report 5 periods worth of data: latest & previous quarter; latest YTD & previous YTD and comparable balance sheet numbers for the start of the year to which the quarter belongs. In the 1st quarter 10Q, the YTD's won't appear.

So in an XBRL quarterly filing, there are in theory up to 5 principal periods but XBRL has two types of concept (or context to use the XBRL parlance) for any given period: one for duration elements (e.g. income statement items) and one for instant elements (e.g. balance sheet items) so that might lead you to think that a quarterly filing has 10 of these things, except that it doesn't because a quarter and a corresponding YTD share the same instant balance sheet date.

This makes lining up periods for comparison tricky. especially the latest YTD which shares the same balance sheet numbers as the quarter. In XBRL to XL, we have duplicated these numbers to create fully comparable columns of data, regardless of whether you are looking at quarter or YTD figures.

In the "Filing" sheets (above), containing the data laid out as it appears in the filed document, 5 columns of data are downloaded: latest & previous quarter; start of year balance sheet; latest & previous YTD. Be aware that generally, no comparable balance sheet figures are included for the previous quarter - you will need to include the filing for the previous quarter for that.

And in the "Standard" sheet, we have created a new control allowing you to toggle between the quarter and YTD values.

Set it to true to see the YTD numbers.

"Control" is a rather over blown description for it - it's just a cell in which you type either true or false! And the item value formulas use this to determine which columns to use in the calculation.

Friday, 7 September 2012

Comparing XBRL: What happened to Step 1?

If you read my last post, you might be wondering what happened to Step 1! Well Step 1 in comparing XBRL is about actually accessing the stuff in the first place in a way that enables you to do comparisons, which is what I've been droning on about in most of my previous posts and what XBRL to XL is all about.

So it would be a bit rubbish if you couldn't actually access all the data, like in XBRL to XL where you can't currently download quarterlies. So rubbish in fact it's time to change this. As from Monday quarterly data will be available and I will talk here about the issues of accessing quarterly data in XBRL and how we have combated them to maximise comparability and transparency.

Wednesday, 5 September 2012

Comparing XBRL - Step 2: Standardisation

As I may have said before, when a company prepares a financial report, it's principal aim is not to create comparable data but a narrative that best explains its financial position and its potential for changing it. XBRL helps to close the comparability gap but this isn't a guarantee or automatic.

So a company in it's use of XBRL may well create a coherent financial view but not necessarily a complete one. In other words if you tried to fit the data to a standard chart, there might (and probably will) be gaps.

This is how revenue is standardised in XBRL to XL. You can see that Microsoft discloses a figure in XBRL for sales revenue but Oracle doesn't. Oracle does disclose a figure for (total) revenues but Microsoft don't! In fact Oracle disclose all sorts of revenue figures...

So how do you compare revenues for both companies? Well XBRL to XL at the moment does this in a rather clumsy way - it creates it's own comparative total (Total Revenues) which it does by adding up all the things it thinks is revenue as an Excel formula. Now for this formula to be truly complete it would need to include all the following items....

In fact that's not all of them, that's not even a third of them. That's only the ones I could fit in my screen shot whilst zooming out as far as I dare.

The issue of standardisation I'm sure is an issue that will be discussed at length at the data forum in next weeks national XBRL US conference in Austin, Texas.

It even gets it's own webinar - Standardizing XBRL a week later. So it must be on somebodies mind. It certainly is mine as will become clear in future instalments.

I'll be at the conference so if you see me, please say Hi. I even have a stand - Woooh! - well more of a table.