Introduction

RIXML is

  • a standards suite for describing investment research, capturing inter-firm interactions, and streamlining updates to coverage lists and research team rosters.
  • a consortium of firms working together to design and enhance the standards
  • an organization available to help you learn how the RIXML Standards Suite can help your firm

 

Introduction

RIXML.org was formed when a number of buy- and sell-side firms came together to discuss a common problem: the amount of investment research that was being distributed had outgrown the tools available to organize it. The consumers of investment research needed better ways to access the precise research they needed, and the creators of investment research needed a better way to ensure that their research was seen by those who would benefit from it.

Since the Research Standard was released, the organization has created other standards to address the needs of the industry, including an analyst roster change standard and a coverage standard. The newest standard, the RIXML Interactions Standard, enables firms to track interactions as required to comply with the EU’s MiFID II regulations.

More on the RIXML.org consortium

More on the RIXML Standards Suite

 

 

Background

The primary objective of the RIXML Research Standard is to provide extensive capabilities for enhancing any piece of financial research content, in any form or media, with sufficient tagging (also called metadata) to allow research users to search, sort, commingle, parse, and filter the published research in order to deliver highly relevant information to decision-makers.

When the RIXML consortium began, all of the participating firms provided a set of real-life scenarios — use cases — in which a standard like RIXML would help them.

  • Buy-side participants said that they would like to be able to search, sort, and filter information published by the research providers, to provide relevant research to their decision-makers in an easy-to-use format, and to create tools to help them commingle research from different sources.
  • Sell-side firms stated that they would like to be able to focus on developing and producing content and to dedicate fewer resources to the technical issues around delivering it.
  • Vendors and others who create products used by these end users

The resulting list of requirements included a listing of the various types of research, such as:

  • Single-company reports
  • Morning notes
  • IPO reports
  • Macroeconomic research
  • Sector, industry, and country overviews
  • Compilations of several company reports
  • Weekly, monthly, and quarterly publications

It also included the various tasks that need to be accomplished, such as:

  • Searching
  • Browsing
  • Alerting
  • Distribution of all research to research distribution vendors
  • Distribution of subsets (such as models) to individual buy-side firms
  • Email distribution to targeted recipients

Based on these use cases, the group determined the initial list of tags that would be needed. From there, a technical committee was given the task of translating this working draft into the RIXML schema.

The technical committee started by modeling the business requirements into an object model. It also re-organized the draft document to better address the requirements of and expectations for the RIXML schema. A parallel effort helped translate this object model into the actual RIXML schema. Some members of the technical committee took the schema back to their organizations and presented it to the potential end-users of the standard. The comments and concerns they elicited were either incorporated into the Research Standard or were added to the issues log to be further reviewed and addressed in subsequent releases of the standard.

The primary objective of the RIXML Research Standard is to provide extensive capabilities for enhancing any piece of financial research content, in any form or media, with sufficient tagging (also called metadata) to allow research users to search, sort, commingle, parse, and filter the published research in order to deliver highly relevant information to decision-makers.

 

Implementing RIXML Research

What does it mean to implement the RIXML Research Standard?

The manner in which your firm implements the RIXML Research Standard will depend on where your firm is in the investment research chain:

Sell-side firms interested in implementing the RIXML Research Standard will need to:

  • Update their publishing tools so that they produce RIXML tagging data as reports are published
  • Ensure that their tagging policies include the RIXML "required" and "highly recommended" tags, minimizing use of the "PublisherDefined" option
  • Validate their content against the standard
  • Apply the RIXML Research Standard to their distribution models

Vendors will need to:

  • Update their content ingestion protocols
  • Update their applications to take advantage of RIXML tagging and enumeration lists
  • standardize terms used to describe certain aspects of research, such as security type, region, etc. in accordance with RIXML defined terms
  • Require content providers to adhere to the standard

Buy-side firms will need to:

  • Work with technology vendors and/or internal systems developers to adopt the standard
  • Require content providers to adhere to the standard
  • Integrate the RIXML Research Standard into their investment process

Implementation Options

The RIXML Research Standard offers the ability enhance investment research with hundreds of tags, allowing the producers of investment research to describe each piece of investment research in great detail, and even to describe parts of their research reports in great detail. However, the RIXML organization recognizes that this can seem like an overwhelming task, and so has defined two levels of implementation:

Level One

The basic implementation of RIXML Research includes the tags that were identified by RIXML-member firms as the ones most critical to research. Level One is a subset of the full schema that represents a common starting point for introducing RIXML into a research platform. It includes the most important and most widely implemented tags. At a high level, the tags include:

  • Time and date of publication
  • Title, subtitle, abstract, synopsis
  • Name of the author(s)
  • Name of the publishing firm
  • Publication status (released, revised, deleted)
  • Product type (report, comment, model)
  • Language
  • Subject
  • Reference to the research report (PDF, audio file, etc.)
  • Issuers discussed, with cash flow, revenue
  • Securities discussed, with rating, target price, estimates
  • Sector/industry classifications
  • Country or Region
  • Intended audience

Supporting RIXML Research Level One involves providing tagging for the subset of tags identified in the Level One standard. This ensures that documents have the minimum amount of information identified by RIXML-member firms as essential. Not all tags will be used for all publications; for example, the RIXML file for an economic overview with no reference to any specific company will not include a SecurityID tag, even though SecurityID is a Level One tag – it just doesn't apply to that publication. However, in order to be RIXML Level One compliant, the RIXML file must include all level one tags that do apply for that publication. The absence of a Level One tag from a particular RIXML file will mean that although the publisher supports the tag, it determined that the most accurate, meaningful action for that particular RIXML file was to omit it.

Vendor products will be required to support all Level One tags whenever populated by a publisher. Level One tags will be part of the data model behind each vendor platform, and each tag will be available to end users in a meaningful way via the end user interface. Substantive participation will typically mean the ability of platform users to include the tag in searching and/or filtering criteria. Some tags may not be useful or meaningful in search actions, but are used by back-end systems to manage entitlements instead.

Note that any valid Level One RIXML instance document is a fully valid RIXML document. Level One includes all required tags, along with the optional tags that RIXML.org member firms defined as tags that are critical in describing investment research.

Full implementation

While Level One provides the ability to tag the information about what the report is about, the full implementation enables a larger degree of componentization – that is, the ability to tag discrete bits of information. When a research document is created, it is just a "brick" of information. Some publishers will find it helpful to componentize the information in the research report, pulling out information such as:

  • Rating action
  • Event details (event date, venue, etc.)
  • Author’s contact information

There are hundreds of tags in the RIXML Research standard. The more tags a publisher adds, the more sophisticated a vendor can get, and the more targeted a user’s search results can be.

More details about implementing RIXML Research can be found in the RIXML Research User Guide.

 

 

 

Guiding Principles

From the time the RIXML organization was originally formed, and throughout the development and revision of all our standards, there have been three guiding principles that the organization has kept in mind:

Guiding Principle 1: The RIXML Standards Suite must provide an improvement for all participants in the process

Because our standards have been built by a consortium of buy-side firms, sell-side firms, and the vendors that create products for them, they have been specifically designed to meet the needs of our industry. They ensure compliance with all relevant industry regulations and guidelines, and they address the content creation, distribution, consumption, reporting and management needs of all types of firms in the industry.

Whether you are a provider of investment interactions or research content, a consumer of them, or a vendor working to create solutions for the industry, the RIXML Standards Suite offers the detailed, specific tagging structure you need.

Guiding Principle 2: The standards must be able to expand and evolve to adapt to technology and business changes

The RIXML Standards Suite is designed to keep up with changes in the marketplace and is device-, product-, and distribution channel-independent. From the very start of our organization, a key goal was to ensure that our standards can change and evolve as our industry does. The first version of our Research Standard covered only equities, yet we intentionally overbuilt certain tag sets so that different asset classes, with their different tagging requirements, could easily be added in future releases. When firms needed a way to comply with the European Union’s MiFID II regulations, we were able to leverage many of our existing tag sets and add additional ones to address the specific content needs of interactions.

Guiding Principle 3: End users must be able to benefit from RIXML without even needing to know it exists

RIXML is a suite of solutions addressing a number of business needs, including the need to better manage and organize the huge amount of investment research in the marketplace today and the need to capture interaction information accurately and consistently in order to comply with the EU’s MiFID II directives. However, these solutions are implemented behind the scenes. End users will notice that their search results are getting better, that content producers are delivering content formatted for optimal viewing on mobile devices, that vendors are creating products that deliver better integration between investment research and other investment data, or that they are able to easily compile the information they need about their firm’s MiFID-related interactions, but they don’t need to know about RIXML to do so. 

RIXML Research Scenarios

We have compiled a number of scenarios that highlight the benefits of a RIXML-enabled research report creation, distribution, and consumption workflow over the traditional workflow.

Co-mingled internal & external content

Scenario: An asset management firm would like to create a page on their intranet integrating internal information with external information from various sources.
Problem: Integrating multiple internal data sources with multiple external data f requires building and maintaining multiple maps, data storage solutions, and APIs to accommodate different vendors’ search protocols, etc.
RIXML Solution: As an industry standard, RIXML provides tremendous efficiencies to asset management firms and others who need to commingle research from disparate internal and external sources. Whether building a custom solution in-house, or partnering with external vendors, asset management firms can greatly streamline their research creation and consumption process.

Multi-language publication

Scenario: A document is published on Monday, then French, German, and Japanese translations of that document are published on the following Wednesday.
Problem: These documents are published separately and there is no way to indicate that the translations are related to the original document.
RIXML Solution: Each product captures an entire research concept regardless of language or format. Thus, multiple resources can comprise a product and each resource can be tagged with its format and language. In this scenario, the translated versions of the original document will each have their own unique resourceID, but will all have the same productID.

Analyst & team publishing

Scenario: A sell-side analyst publishes research as an individual and also as part of a team. Some consumers want to see all research by this person, some want to see just the team research, and some want to see all except the team research.
Problem: While existing non-RIXML systems may make it possible to code research by a team with the names of the individual team members, it is not possible to indicate whether the team or the individual is the primary author.
RIXML Solution: RIXML provides the ability include the name of a team in the authorship information; it also provides the ability to include the names of the individuals on that team to be associated with the team. Finally, there is a way to indicate which author – either an individual or a team – is the primary author, while also including the names of the additional contributors.
  While at first glance, it might seem as though using a system with this amount of flexibility might make the process of creating or consuming research to be unbearable complex, the beauty of RIXML is that the tagging is invisible to the end user. In this example, the tagging is hidden both at the research creation end of the process and the research consumption end:
  Research creation: A research creation tool might input the author’s own name, phone number, email, etc. into the research report, but also give them the ability to add the team name, its members, and to select which team member was the primary author
  Research consumption: A search screen might have an author lookup box, and under that the user would check or uncheck an “include team research” option and a “primary author only” option.

Integrating financial data

Scenario: An asset management firm would like to filter research to display reports matching specific forecasted financial results criteria to display in a particular Alerts module of their intranet.
Problem: Historically, there was no way to connect SEC filings, sell-side models using the data in the SEC filings, and the research reports that provide in-depth analysis of the filings and explanation of the resulting models.
RIXML Solution: RIXML has integrated XBRL into the RIXML specification for describing financial data, allowing for greater integration between investment research and financial data, such as SEC filings and sell-side analysis.

Region & country search

Scenario: An economist wants to find macroeconomic research on individual Latin American countries, not an overview on the region or a sector report that discusses Argentina. An analyst at the same firm wants to find a sector overview of the oil industry in Latin America, not company reports on oil companies in Brazil and Chile. Meanwhile, a portfolio manager is looking for a report that discusses how elections in one South American country will affect neighboring countries.
Problem: What one person means when they say they want a report “about Latin America” can be very different than what another person means.
RIXML Solution: The RIXML specification has the a number of concepts that when used together allow the end users to locate exactly what they need:
  • The concept of ProductFocus provides the ability to indicate the primary focus of the content. It is used to indicate that one report is primarily an industry report, while another is primarily about a region, while another is a multi-country overview. This gives a publisher the ability to code one report as being primarily an industry report, but also indicate that it focuses on that industry within specific countries/regions. It can describe a different report as being a macroeconomic report about a specific region, but also indicate the specific countries, industries, and companies that are mentioned in the report. A research consumer can narrow or broaden a search to include or exclude content as needed to retrieve results in line with what they need.
  • The specification provides clear guidance about when to code a report a region report vs. when to code it a country report.
  • RIXML provides a standardized region classification list.