Rivet Software » Analytics

The future start tag for all integrated reports?

Stewart McKie - Executive Advisor — Sat, 31 Dec 2011 12:32:13 +0000

In my roundup of the XBRL news of 2011 (coming soon), I mention the Integrated Reporting (IR) movement, which has been discussed in this blog over the years. The IR’s discussion document Towards Integrated Reporting: Communicating Value in the 21^st Century is well worth a read and also some thought in response to the questions it helpfully poses within the text. But what interests me more is their new logo , which looks to me like an XBRL tag – something that we should not regard as a coincidence.

This tag could be the tag that starts every corporate report in the future. This is because, like it or not, the integrated report of the future will simply be a container for a set of more or less XBRL feeds encompassed by the start and end tags. These XBRL feeds will comprise tomorrow’s integrated report ‘components’ used to assemble the report content. Today these components can be financial GAAP/IFRS and non-financial GRI sustainability XBRL; tomorrow when a great deal of other data is easily available as an XBRL feed, a much wider range of XBRL components could also be assembled within the tags.

In effect, integrated reporting will become a mashup, powered by XBRL.

There’s nothing really new about this. Remember row and column sets in FRx? Same deal. It’s just that that the content of these IR components consists of standardised XBRL data rather than some legacy proprietary format.

Get the Most Out of Your XBRL Software

Charlie Hymer, CPA - Regional Sales Director — Mon, 19 Sep 2011 20:21:58 +0000

Charlie Hymer, CPA – Regional Sales Director, reminisces about the old days of financial reporting and how XBRL will forever change its future.

The financial reporting industry has changed drastically over the last 20 years. Not to date myself, but oh well, here goes. I remember when I first started out at Arthur Andersen we didn’t have computers. We used columnar paper and had to continually tape more paper on the end of the spreadsheets to add more columns if the paper wasn’t long enough. To make a change was completely manual—erasing, re-adding, and hoping that everything flowed through properly. In 1989 we got computers to go out in the field. They looked like huge suitcases and we newbies always had to carry them because they were so heavy. We used Lotus 123 and thought that it couldn’t get any better than that. Then a little company called Microsoft gave us Excel and Word. What? No more hand created spreadsheets? Where’s the fun in that? It was accounting nirvana. Although this was a drastic change in the late 1980’s, there haven’t been any revolutions since. Until now! We’ve moved from a paper-based system to an electronic system. Now we are moving to Internet systems, based on XBRL. This will further revolutionize the way financial reports are processed.

I know, XBRL has been a burden to the public companies filing with the SEC. What is forgotten is that XBRL is a way of tagging data that frees you from your spreadsheets. Let me explain in a little more detail because this way of thinking is completely new for us spreadsheet focused accountants. How many times have we looked at our summary Excel or Word document and said, “Hmmm, where did this number come from?” You hover over the cell and learn that it came from a previous spreadsheet in your workbook. You go to that spreadsheet, click on the cell and learn that it came from another spreadsheet. Eventually you learn that the original source is located on a coworker’s desktop. A couple of alarms go off in my head.

The audit trail – Tracing data back to the source can take a significant amount of time to move through all the links. At Arthur Andersen, my senior would make me print out the cell references to help with the audit trail.
Excel linking – What happens if your coworker’s hard drive goes down? Or somewhere in the spreadsheets someone accidentally breaks the link causing the dreaded #REF#!?
Changes – How do you make a last minute change, ensure it flows through properly, and that the final document is accurate without visually watching this change happen?

Now let’s think about this from the XBRL perspective. All XBRL does is attach information about the number to the number. Or for us spreadsheet focused accountants, the column and row information is attached to the number wherever the number goes. Let’s use a simple example. Your PP&E at September 30, 2011 for Subsidiary 1 is $1,200,000. XBRL tags the 1,200,000 with the dollar sign so we know it’s US dollars, the date so we know it’s for 9/30/11 and that it’s PP&E for Sub 1. Now wherever this number is used you will be able to see this information. No more searching back through documents for the audit trail to confirm the number. I am over simplifying, but hopefully you are starting to see the benefit.

Because of the XBRL tag, there is no need for Excel linking. Merely refer to the element name and date and the number will be updated in every instance correctly. No more Find & Replace! No more Excel #REF#!

This decreases the amount of time for both internal and external reporting departments. Accounting departments are not a revenue generating department. We should be a value-add department though. XBRL allows for quick and accurate changes. This changes the accounting world. Now when management asks for a report, we will be able to create and deliver it quickly and efficiently. Information will become timely thereby increasing our value to our companies. And we won’t be doing administrative tasks to ensure data integrity and accuracy.

With XBRL, we no longer have to worry about taking weeks or months to process our financial reports. XBRL gives us the ability to get reports completed at the speed of light (well almost) and the list of benefits stemming from XBRL grows daily. Many companies are already incorporating XBRL into their accounting departments to take advantage of this technology. Here are some features to look for when you consider implementation:

Ability to review online while not requiring users to be online if they so desire – I’ve never been able to convince upper level management to make changes online and selecting a package that doesn’t force them to do so will make your life a little more pleasant
Multi-User Collaboration – allows multiple users to review documents and data at the same time
Analyze – analysis features allow users to compare their financial performance vs. peers and competitors
Document Management – Look for solutions that include version control and change tracking all fully integrated with Microsoft Excel and Word
Change Once, Change Everywhere – updates to tagged data automatically change in all instances within the document
Data Aggregation – The solution must use XBRL to aggregate data. Seek out a solution that is built on XBRL. Many solutions are adding XBRL as a module or as a requirement to get the beloved yes on the RFI.

No matter what vendor you chose, make sure that you have confidence in your reports. XBRL makes data accurate, reliable, and reusable. We accountants are sometimes overworked, underappreciated, and undervalued. XBRL just might change all that!

Associative Search

Stewart McKie - Executive Advisor — Mon, 04 Jul 2011 16:26:04 +0000

We’re having a heat wave here in the UK so as I lay out under my lawn sprinkler (suitably primed with a mixture of spring water and a rather nice chardonnay – me that is, not the lawn sprinkler) I began leafing through one of my favorite journals - Biological Cybernetics. Yes, I know you all have a prized copy hidden away somewhere – under the bed or in the garden shed – but there, back in the 1981 volume, I came across Associative Search Network: A Reinforcement Learning Associative Memory introduced by the authors as:

In this paper we describe an associative memory structure which is not told by some outside process (e.g., a “teacher”) what pattern it is to associate with each key. Instead, for each key, the network must search for that pattern which maximizes an external payoff or reinforcement signal. As this kind of learning proceeds, each key causes the retrieval of better choices for the pattern to be associated with that key.

and I wondered what relevance this might have to XBRL.

I wouldn’t recommend trying to read this paper. I dozed off pretty quickly. That generally happens whenever I see something like this:

p(t) = ~ wPi(t)wi(t ).

Which sounds a lot like a sparrow choking on a large worm…

Anyway, Norbert Weiner’s Cybernetics deals with the positive outcome that can result from negative feedback. Whereas associative search deals with the positive outcome that can result from positive choices i.e. rather than saying ‘not that, not that’ to get to a query result (negative feedback) you say ‘that looks right, that looks right’ (positive feedback) to get to the same place.

Since 1981, associative search has emerged in business intelligence software as a kind of contextual drilldown. So in an XBRL context it could start by you typing in a term that might be part of an XBRL tag to enable the search engine to find other tags that are contextually linked to that tag. Then you pick one of the ‘found’ linked tags to repeat the process, until you end up with just one tag (or set of the same tags if you were doing a peer group comparison).

The issue is the contextual linking. XBRL is great at ‘vertical’ drilldown e.g. from high level summary number to transaction detail (via XBRL GL) but seems less well-equipped for doing ‘horizontal’ searches within a given context space. Or am I wrong?

Benefits of XBRL – Analyzing data even when a company uses a unique extended element

Kevin Berens - CPO — Mon, 20 Jun 2011 15:11:37 +0000

We know that it’s pretty easy to analyze XBRL data when companies are using the same base taxonomy element.

What about when companies have used different base elements?
What about companies that did not tag the totals but instead tagged all of the details?
What about companies that tagged using an extended element?

It’s still a simple process to compare multiple companies even if some use base elements and others use extended elements.

Because the XBRL data is captured at the element level rather than the page level, you can simply combine two elements onto a single row. Now each time you refresh the data, the extended element is picked up on the row as if they had used a base element.

The company may have used an extend element for tagging, but it does not mean you have to put these in different rows for analyzing the data.

Take this example for looking at total operating expenses:

Acorda Therapeutics and ONYX Pharmaceuticals used the element – Operating Expenses, Total
Amgen and Forest Laboratories used the element – Costs and Expenses, Total
Lilly Eli and Schering Plough used extended elements for total costs and expenses
Allergan did not use any of the total expense taxonomy elements, they tagged the individual expense items but not the total.
In addition, Forest Labs Fiscal year end is March 31.

However, on a single row, we can show the data for September. This ties to each of the companys’ financial statements.

To get this data for December or March, simply change the date and the values are automatically updated.

Compare these to Yahoo or MSN. The advantage is that there was no work to get March data. You simply refreshed the data and it appeared.

Benefits of XBRL – Peer Comparison Reports using XBRL and Rivet’s Crossfire Reporting Software

Kevin Berens - CPO — Fri, 17 Jun 2011 17:11:52 +0000

Thanks to companies leveraging XBRL to report their financials, I can now compare companies side-by-side – and I can do it more quickly than ever before.

I created a report to compare 24 different companies. In a single report, I gathered data for the 3 previous months and created 16 different ratios for these companies. This report was created in just a few hours. Now, when any of these companies file, this report is updated automatically with no additional work involved. I am even alerted when these ratios fall outside the norms that I defined.

Because these financials were tagged using XBRL, I know the answers to questions within minutes that in the not too distant past would have taken me weeks to figure out.

These are the 24 pharmaceutical companies I analyzed. I chose them at random.

Which two companies have a current ratio of greater than 15? If you used Rivet’s Crossfire, you could have set alerts to be notified anytime a company had a current ratio greater than 15 and you would know immediately.

Do you know which of these companies who had the largest Accounts receivable turnover ratio?

What are the top 5 companies based upon return on sales? How about the bottom 5?

Which company had a return on sales greater than 50% but a return on equity of less than -20%? How much money did they spend on R&D?

Which companies have ROS that has increased for each of the last 2 quarters? There was only 1. What was that particular company’s R&D expense ratio as compared to the other 23 companies? It was less than all but 2 of these companies.

The data provided in the XBRL standard allows you to answer these questions quickly and timely.

I can already hear the naysayers…

“You cherry picked the companies.” – I picked 24 pharmaceutical companies because it is hard to cherry pick 24 companies.
“You are only analyzing elements that every company used.” - I gathered numerous elements for both the balance sheet and income statement.
“All of the companies used the same calendar year end.” – The companies were chosen randomly so the calendar year ends are all over the board. Some have a September year end, some have a March year end. Some quarters end on 4/3.

In this example, some of the companies used the element representing total operating expenses, some used total costs and expenses, some extended this element, others did not tag the total but just tagged the children. Some companies had a March year end, some of the quarter ends were not 3/31. Some of these are Fortune 500 companies, others are not. Some are Rivet tagging clients, others are not.

None of these issues caused significant report creation issues. XBRL, a single repository for the data, and a new breed of applications enables this type of timely and powerful reporting. Because the data was submitted in an XBRL format, we were able to easily compare the 24 companies side by side. And the best part, we did not have to wait for the data for weeks or rebuild the report, so we were able to spend more time analyzing the data rather than gathering the data.

Benefits of XBRL – XBRL provides timely data

Kevin Berens - CPO — Wed, 15 Jun 2011 19:05:51 +0000

Imagine a world where a company files today with the SEC and you are automatically alerted if their Return on Sales has dropped below 20% and their Quick Ratio has dropped below 2.0. Well that world is today, thanks to XBRL.

Prior to XBRL, this was not possible, unless you paid to have this information delivered to you. Companies would produce their SEC filings in HTML. To get at this data, you would go the SEC site, copy and paste the data into Excel, and begin the process of “fixing” the data (modifying all of the negative numbers, etc) because the HTML tables did not come over seamlessly. Now imagine doing this for 20 of your peers. It might take you weeks to manually gather and prepare the data to perform a side-by-side comparison. You would spend all your time gathering the data and have no time to analyze the data.

One alternative is to wait until the data is captured and displayed on MSN or Yahoo. The problem with this is it takes days for the larger companies to appear on these sites. For the smaller companies, some of that data is not available for 30 days. Once the data is available, you only get the ratios they want you to have. You are not able to get the ratios you want.

For example:

Quanex filed June 14th – This data is not available on either MSN or Yahoo (as of June 15^th), but Crossfire by Rivet has accessed this data and generated customized KPI reports, as soon as it was filed.
Hormel filed June 10th– This data is not available on Yahoo (as of June 15^th), but Crossfire by Rivet has accessed this data and generated customized KPI reports, as soon as it was filed.
Buckle filed June 9th – This data is not available on Yahoo (as of June 15^th) , but Crossfire by Rivet has accessed this data and generated customized KPI reports, as soon as it was filed.

If the data for these large companies is not available after almost week, imagine how long the delay is for the smaller companies. Some of this data may not be available for weeks. Who has time to wait?

Below is a report displaying 6 companies who have filed in the last few days. None of this data is available on Yahoo as of June 15^th.

Tomorrow, I will cover Peer comparison reports where we compare 24 companies side-by-side.

Recognizing the Benefits Made Possible by XBRL

Kevin Berens - CPO — Tue, 14 Jun 2011 21:04:37 +0000

For the past year, I have seen multiple blogs where people have questioned the benefits of XBRL. I believe the time has come to start recognizing the benefits of XBRL. Whether you want to analyze how your company is performing as compared to your 5 closest competitors, analyzing the financials of companies who you want to invest in as well as their peers, or just wanting to look at the contingencies and commitments for selected companies, XBRL is a tool that will allow you do all of these plus more.

Over the following week, I will have a series of blogs that will showcase the benefits of XBRL.These benefits can be broken down into three categories:

Timeliness of the data – getting the data anywhere from days to even weeks faster
Analyzing the data – XBRL will help you spend less time gathering data and more time analyzing the data
Systematically getting at selected financial information buried in the notes to the financial statements – XBRL allows you to get at this data for multiple peer companies for the first time

Prior to XBRL, analysts at large companies said they would rarely compare themselves to their peers because it took so long to get at the data. With XBRL, the paradigm is now shifted. Instead of spending 95% of your time gathering the data and 5% of the time analyzing the data, you can now spend 95% of your time analyzing the data to make intelligent business decisions.

Tomorrow, I will cover examples of how XBRL data can be used for timely reporting.

XAID

Stewart McKie - Executive Advisor — Tue, 24 May 2011 10:12:14 +0000

If you’re from New York you might well think that the Tunis Consensus is some kind of cool new sandwich.

Hey! Luigi…gimmee a Tunis Consensus on rye and hold the paprika.

But sorry, it’s not.

In fact it’s a document from 2010 about realizing Africa’s own vision for development and part of that vision includes greater transparency in terms of how aid funding is decided and spent:

Participants agreed that better information flows were essential to improving accountability. Nothing less than full disclosure of development expenditure is necessary. Fiscal transparency includes publishing budget plans and reports and ensuring that the public has ready access to information on the state of public finances and on the structure, functions and financing of public institutions. [p.10 - The Tunis Consensus]

The portal aideffectiveness.org has a page on mutual accountability. However the site has very little to say about data standards and nothing about XBRL. But yet again this is a problem with a clear solution – an XBRL taxonomy for reporting aid data (XAID).

There is absolutely no reason why this aid taxonomy needs to be that complicated. It would help to resolve a global issue by creating a global standard that Governments and NGOs can use to track, report and visualize aid flows by means of a common data standard.

But more importantly, XAID will help donor taxpayers – especially in generous countries like the USA - to fully understand where their aid money is going, whether promises are being kept and what the aid ROI actually is. All of which are at the heart of ‘mutual accountability’.

Something else for Congressman Issa to consider?

Land of the Rising Taxonomy

Stewart McKie - Executive Advisor — Tue, 17 May 2011 10:06:47 +0000

Fumiko Satoh of IBM Research in Tokyo has recently published a paper that advocates an XBRL Taxonomy for Estimating the Effects of Greenhouse Gas Emissions on Corporate Financial Positions. As the abstract suggests:

…disclosing the emissions data by using XBRL will be very beneficial for the analysis of the financial positions and emissions results of these companies. Interested third parties will want to combine the XBRL reports of the financial data and the emissions data. This will allow them to easily evaluate the companies from both the financial and environmental perspectives.

Luckily I don’t have to worry about the carbon impact of my blog postings. Though it might be a different matter if the focus of emissions disclosure was methane.

Global Brains or Dare to Share (9)

Stewart McKie - Executive Advisor — Mon, 16 May 2011 10:30:36 +0000

I enjoyed skimming Tim O’Reilly’s presentation from TED 2011 Towards a Global Brain. It has some good dare to share quotes in it like this one from Peter Norvig, Chief Scientist at Google:

We don’t have better algorithms. We just have more data.

As XBRL is just one of many bricks in the wall of data sharing, this is encouraging.

One of the concepts O’Reilly also refers to is ‘selection by association’, which is something that has an XBRL connection in relation to the kind of compliance analytics Rivet supports e.g. the decision to extend or not extend a tag or to select or not select the same tag for a fact as your peer group. Associating your tag selection choices with those of your peer group could help to avoid ‘red flags’ when your S.E.C. filings are reviewed by regulator or investor analysis tools.