Data quality technology continues to see increased demand from technology buyers and vendors in related markets. As a result, the market for data quality tools is growing at a solid pace, and new vendors are entering this market. Large vendors, such as Business Objects, IBM and Pitney Bowes, have entered via acquisitions, while numerous small vendors have entered via the organic development of new data quality technology. The entry of larger vendors signals a change in the competitive landscape and also raises questions as to the long-term outlook for vendors of pure-play data quality tools. In addition, various aspects of data quality functionality, beginning with profiling, are appearing in related technology markets, such as the data integration tools market, and in the enterprise application vendors' platforms.

When evaluating offerings in this market, organizations must consider the breadth of functional capabilities (for example, data profiling, matching, cleansing and enrichment) relative to their requirements. Other key criteria include the degree of integration of these capabilities into a single architecture and product, as well as the size, viability and partnerships of the vendors.

Throughout its history, the discipline of data quality has focused generally on specific domains, such as customer information files and campaign management. This domain specificity has often resulted in highly regionalized or narrow solutions that ignore the breadth and global nature of today's business environment. Many contemporary data quality initiatives still focus on a single domain, but there has been a clear move toward multifaceted efforts that address multiple domains. Companies are discovering that data quality has a significant impact on their most strategic business initiatives, and, as such, are raising the visibility of data quality activity within their enterprise information management programs. Compliance and transparency are at the top of the list of concerns, along with ensuring the capture of expected benefits from investments in business intelligence and master data management. A common thread underpins successful initiatives in each of these areas: the prerequisite for high-quality data to add accuracy and value to the relevant business processes.

Data quality is often understood as "data cleansing" only. However, this is only one area of the overall data quality continuum. A more complete view includes areas such as profiling, standardization, matching and enrichment. In addition to offering a range of functional capabilities, data quality tools support multiple modes of deployment. Historically, the tools have been deployed in batch-mode silos — as one step in an overall flow or as a stand-alone, offline process. Increasingly, the tools are being deployed in a real-time manner, integrated into operational systems with the goal of standardizing and cleansing data at the point of capture and maintenance. Well-documented application programming interfaces (both proprietary and standards-based) to the tools are facilitating their integration with other packaged software — from business applications, such as ERP and CRM, to infrastructure components, such as data integration tools. In addition, data quality is playing an important role in master data management (MDM) environments; in particular, matching, deduplication and standardization technologies are often used to define master records. Most tool use has been for improving the quality of customer data, but organizations are increasingly looking to deploy the tools against other data subject areas — in particular, product data (often within the context of a product information management [PIM] initiative) and financial data (as it relates to compliance directives in financial reporting). An additional emerging trend is the application of data quality tools in the domain of content management, to understand and improve the quality of less-structured data.

The vendors in this market offer a broad range of data quality functionality. However, within their portfolios, many still retain the various data quality components as separate product offerings, with some degree of integration between them. Data quality technology must further evolve toward a model where a common set of business rules drives all the various data quality components (from profiling, matching and cleansing, to validation, standardization and enrichment). In addition, these data quality capabilities must be able to behave in a service-oriented manner, existing as an enterprise service that is consumed by all applications, processes, infrastructure tools and users that require data quality operations (see "The Emerging Vision for Data Services: Use Data Quality, Profiling and Mining to Achieve Policy-Based Business Rules").

The market for data quality tools is modest in size but is growing at a rapid pace (see "Forecast: Data Quality Tools, Worldwide, 2006-2011") compared with many other software markets. Interestingly, much of the innovation is coming from outside the United States. As a result, the veteran data quality tool vendors are being challenged by entrants that have an international focus and propensity toward designing and deploying domain-agnostic data quality services (stand-alone or embedded in applications), based on a centrally managed set of business rules. However, with the increasing trend toward embedding data quality capabilities in business applications, data integration tools and other software offerings from larger vendors, these small competitors will face significant challenges in survival and growth. Several of the "mega vendors" have started to build rudimentary data quality capabilities into their data management products, but for the most part they are leveraging third-party applications to provide deeper data quality. Within the next 24 months, the mega vendors, as well as business intelligence and data integration vendors that are still lacking data quality capabilities, will look to expand their portfolios. This will lead to more acquisitions and consolidation in the data quality tools market.

The data quality tool market comprises vendors that offer stand-alone software products for addressing various aspects of the data quality problem:

• Parsing and standardization: decomposition of text fields into component parts and formatting of values into consistent layouts based on industry standards, local standards (for example, postal authority standards for address data), user-defined business rules, and knowledge bases of values and patterns.
• Generalized "cleansing": modification of data values to meet domain restrictions, integrity constraints or other business rules that define sufficient data quality for the organization.
• Matching: identification, linking or merging related entries within or across sets of data.
• Profiling: analysis of data to capture statistics (metadata) that provide insight into the quality of the data and aid in the identification of data quality issues.
• Monitoring: deployment of controls to ensure ongoing conformance of data to business rules that define data quality for the organization.
• Enrichment: enhancing the value of internally held data by appending related attributes from external sources (for example, consumer demographic attributes or geographic descriptors).

The tools provided by vendors in this market are generally consumed by technology users for internal deployment in their IT infrastructure, although hosted data quality solutions are continuing to emerge and grow in popularity.

For vendors to be included in the Magic Quadrant, they must meet the following criteria:

• Offer stand-alone (not only embedded in, or dependent on, other products and services) packaged tools that are positioned, marketed and sold specifically for data quality applications.
• Deliver functionality that addresses, at a minimum, profiling, parsing, standardization, cleansing and matching. Vendors offering only narrow functionality (for example, only address cleansing and validation, or only matching) are excluded because they do not provide complete data quality tool suites.
• Support this functionality for data in more than one language, and specific to more than one country (in the case of address standardization).
• Maintain an installed base of at least 50 production customers for their data quality products.
• Demonstrate, via customer references, the use of the tools at an enterprise (cross-departmental or multiapplication) level.

A vendor that does not meet the above criteria may be considered for inclusion if it is a new entrant to the market that is demonstrably different from established vendors and represents a future direction for data quality tools.

It should be noted that a large number of data quality tools vendors currently exist, but most do not meet the above criteria and are, therefore, not included in the Magic Quadrant. Many vendors provide products to address one very specific data quality problem, such as address cleansing and validation, but cannot support other types of applications, or lack the full breadth of functionality expected in today's modern data quality solutions. Others provide a range of functionality, but operate only in a single country or support only narrow, departmental implementations. Others may meet all the functional, deployment and geographic requirements but are at a very early stage in their life span and, therefore, have few, if any, production customers. The following vendors are occasionally considered by Gartner clients alongside those appearing in the Magic Quadrant when deployment needs are aligned with their specific capabilities, or are newer market entrants beginning to gain visibility in the market but lacking a significant customer base:

• AddressDoctor, Maxdorf, Germany, www.addressdoctor.com
• BackOffice Associates, Harwich, Massachusetts, www.boaweb.com
• Business Data Quality, London, U.K., www.businessdataquality.com
• Certica Solutions, Wakefield, Massachusetts, www.certicasolutions.com
• ChoiceMaker, New York, New York, www.choicemaker.com
• DQ Global, Fareham, U.K., www.dqglobal.com
• Identity Systems (a division of Nokia), Old Greenwich, Connecticut, www.identitysystems.com
• Infogix, Naperville, Illinois, www.infogix.com
• InQuera, Tefen, Israel, www.inquera.com
• Omikron, Pforzheim, Germany, www.omikron.net
• Melissa Data, Rancho Santa Margarita, California, www.melissadata.com
• QAS (a subsidiary of Experian), London, U.K., www.qas.com
• Silver Creek Systems, Louisville, Colorado, www.silvercreeksystems.com
• SRC, Orange, California, www.extendthereach.com
• Stalworth, San Mateo, California, www.stalworth.com
• Udex, Newport, U.K., www.udex.com
• Utopia, Mundelein, Illlinois, www.utopiainc.com
• Zoomix, Jerusalem, Israel, www.zoomix.com

Gartner will continue to monitor the status of these vendors for consideration as new entrants to the data quality tool Magic Quadrant during subsequent updates.

Datactics, Belfast, Northern Ireland, www.datactics.com

DataMentors, Wesley Chapel, Florida, www.datamentors.com

Datanomic, Cambridge, U.K., www.datanomic.com

Fuzzy! Informatik, Ludwigsburg, Germany, www.fazi.de

Netrics, Princeton, New Jersey, www.netrics.com

Uniserv, Pforzheim, Germany, www.uniserv.com

No vendors were dropped from the previous version of the Magic Quadrant.

Vendors' ability to execute in the data quality tools market is determined by the following criteria:

• Product/Service: how well the vendor supports the range of data quality functionality required by the market, the manner (architecture) in which this functionality is delivered, and the overall usability of the tools. Product capabilities are critical to the success of data quality tool deployments, and, therefore, receive a high weighting.
• Overall Viability: the magnitude of the vendor's financial resources and the strength of its people and organizational structure.
• Sales Execution/Pricing: the effectiveness of the vendor's pricing model, and the effectiveness of its direct and indirect sales channels.
• Market Responsiveness and Track Record: the degree to which the vendor has demonstrated the ability to successfully respond to market demand for data quality capabilities over time.
• Marketing Execution: the overall effectiveness of the vendor's marketing efforts, and degree of "mind share," market share and account penetration that the vendor has achieved as a result. Marketing execution is a significant driver of sales, growth and brand awareness, and, therefore, receives a high weighting.
• Customer Experience: the quality of the vendor's general customer service, implementation service and technical support, and customers' perception of overall value.

Vendors' completeness of vision for the data quality tools market is determined by the following criteria:

• Market Understanding: the degree to which the vendor leads the market in new directions (technology, product, services or otherwise) and its ability to adapt to significant market changes and disruptions. Given the dynamic nature of this market, this item receives a high weighting.
• Marketing Strategy: the degree to which the vendor's marketing approach aligns with and/or leverages emerging trends and the overall direction of the market.
• Sales Strategy: the alignment of the vendor's sales model to how customers' preferred buying approaches will evolve over time.
• Offering (Product) Strategy: the degree to which the vendor's product road map reflects demand trends in the market and fills current gaps or weaknesses. Also, the strength of the vendor's strategy regarding alliances of different types is considered.
• Business Model: the overall approach the vendor takes to execute on its strategy for the data quality market. With a reasonably high degree of similarity across the vendors in this market, this item receives a low weighting.
• Vertical/Industry Strategy: the level of emphasis the vendor places on vertical solutions, and the vendor's depth of vertical expertise. Given the broad cross-industry nature of the data quality discipline, vertical strategies are less critical and, therefore, this item receives a low weighting.
• Innovation: the degree to which the vendor has demonstrated a willingness to make new investments to support the strategy and enhance product capabilities, the level of investment in R&D directed toward development of the tools, and the extent to which the vendor demonstrates creative energy. With rapidly evolving technology requirements — in the face of trends such as service-oriented architecture (SOA) — and increased competition in the market from large vendors, this item receives a high weighting.
• Geographic Strategy: the global presence of the vendor and the manner in which it is achieved (direct local presence, resellers, distributors and so on), in light of emerging demand for data quality capabilities in regions such as Asia, and the desire of multinational enterprises to leverage common tools worldwide.

Leaders in the market demonstrate strength across a complete range of data quality functionality, including profiling, parsing, standardization, matching, validation and enrichment. They exhibit a clear understanding and vision for where the market is headed, including recognition of noncustomer data quality issues and the delivery of enterprise-level data quality implementations. Leaders have an established market presence, significant size and a multinational presence (directly or as a result of a parent company).

Challengers in the market provide strong product capabilities but may not have the same breadth of offering as Leaders. For example, they may lack several functional capabilities of a complete data quality solution. Challengers have an established presence, credibility and viability, but may demonstrate strength only in a specific domain (for example, only customer name and address cleansing) and/or may not demonstrate thought leadership and innovation to a significant degree.

Visionaries in the market demonstrate a strong understanding of current and future market trends and directions, such as the importance of ongoing monitoring of data quality, engagement of business subject matter experts and delivery of data quality services. They exhibit capabilities aligned with these trends, but may lack the market presence, brand recognition, customer base and resources of larger vendors.

Niche Players often have limited breadth of functional capabilities and may lack strength in rapidly evolving functional areas such as data profiling and international support. In addition, they may tend to focus solely on a specific market segment (such as midsize businesses), limited geographies or a single domain (such as customer data), as opposed to positioning toward broader use. Niche Players may also have good functional breadth, but may have an early-stage presence in the market with a small customer base and limited resources. Niche Players that specialize in a particular geography or data domain may have very strong offerings for their chosen area of focus and deliver substantial value for their customers in that segment.

• Business Objects entered the data quality tools market through its acquisition of Firstlogic in 1Q06. Business Object's significant global presence, combined with Firstlogic's brand recognition in North America, large installed base (over 2,500 customers) and market share leadership, creates significant leverage and cross-sell opportunities for the vendor to grow its data quality tools business.
• Business Objects provides good breadth of functional data quality capabilities, including data profiling (via Data Insight XI) and the common data cleansing operations (via Data Quality XI). The core data quality functionality in Data Quality XI, based on the new IQ8 architecture of the Firstlogic technology, enables the delivery of data quality services in an SOA context. The vendor continues to increase the depth of integration across the product portfolio with each major release.
• With the addition of the Firstlogic's heritage, technology, customer base, people and experience, Business Objects' strength remains very much in applications of customer data quality, specifically in matching/linking, de-duplication, and name and address standardization and validation.

• Because Firstlogic's technology was heavily biased toward customer data quality, Business Objects is lagging behind many of its competitors with regard to product capabilities and experience in addressing other data domains. It will attempt to close this gap with the delivery of more generalized parsing and cleansing operations, planned for 3Q07.
• Since acquiring Firstlogic, Business Objects has appeared in competitive situations less frequently than Firstlogic did prior to 2006. This is likely due to the removal of the Firstlogic branding and Business Objects' limited recognition as a data quality tools vendor.
• Business Objects is making many value-added infrastructural enhancements to its "EIM" line of products (such as scalability improvements via grid computing and caching, and productivity improvements via enhanced reporting and automated tuning). This will initially benefit the vendor's data integration tools (primarily, Data Integrator), and will not impact the data quality tools until slightly later in the product road map.

• Datactics is a small data quality vendor based in Northern Ireland. Its software is used in a range of subject areas, not limited to typical name/address verification scenarios. Datactics operates primarily in Europe, but there are a number of value-added resellers in the Americas and Asia.
• The flagship product, DataTrawler, is fully 64-bit and Unicode enabled, supports most of the European languages, runs on many platforms and supplies broad capabilities in profiling, matching/merging, cleansing and monitoring. Data quality scorecards can be constructed to monitor quality-related metrics.
• Datactics has partnerships with consultancies and system integrators that have used the DataTrawler product in some strategic data quality programs, which is quick to implement at a reasonable cost, mostly in the government sector. Datactics has also built an alliance with ETI, a data integration tools vendor, and other software companies that include DataTrawler services.

• With only four sales employees, limited marketing budgets and relatively low-profile partnerships, Datactics is "flying underneath the radar" for most organizations looking for a provider of data quality tools. Datactics needs to find stronger, more visible partners in the software arena, such as in business intelligence.
• The current system integrator channel does not provide sufficient inroads and introductions to potential new accounts, as they are only working as intermediaries. A strong reseller agreement is required to build stronger bonds between Datactics and its partners.
• Although the DataTrawler product is service-enabled, references indicate that the main use of the technology is in batch environments only, with a focus on matching. Ease of use is repeatedly cited as a challenge.

• DataFlux has been moving out of the large shadow of its parent company, SAS. It is demonstrating high growth and has begun to be seen as a major competitor, most recently in Europe. The company is investing heavily in R&D, and is establishing key relationships with large system integrators.
• The vendor has expanded its functionality beyond the core competency areas of data cleansing and matching, in particular in data profiling, rooftop geocoding, team development, internationalization and service-oriented architecture (SOA) support. It has also added data quality monitoring, including some basic dashboarding capabilities.
• The DataFlux platform includes profiling, cleansing and monitoring capabilities in a single architecture, thereby reducing the efforts to integrate discrete products and increasing usability through a consistent user experience.

• DataFlux has been seen mostly as a stand-alone data quality technology provider. However, through the new customer data integration (CDI) solution, and a broader data integration positioning with its Integration Server product, DataFlux will increasingly try to address needs beyond its core market and into the world of MDM. But this may put DataFlux at odds with its parent company and could generate some confusion in the market, as SAS is also targeting data integration opportunities.
• To become a data quality name brand independent of SAS, DataFlux will need to strike partnerships and build a channel with other vendors in the business intelligence and data integration space, possibly companies that compete directly with SAS.
• Its attempt to become a more globally visible data quality player is still in its early stages. Unicode support is very new and has shown little traction.

• DataLever provides integrated data profiling and data cleansing functionality in a single product. All operations can be readily deployed in both batch and real-time modes.
• DataLever takes a domain-agnostic view of data quality issues, enabling its technology to be applied toward various data domains, including customer, product and others. Much of the current installed base applies DataLever's technology toward customer data quality issues, with some users working in a combination of other areas.
• Customers cite overall ease of use, relatively short implementation times and lower cost in comparison to alternative offerings as the main selling points of DataLever's products. In addition, the lower complexity of the product enables its use by business subject matter experts in addition to IT personnel.

• Still a relative newcomer in this market, DataLever supports a small customer base of approximately 150, with virtually no presence outside North America.
• DataLever currently has very limited runtime platform support (Windows and Linux only), but expects to deliver Unix support in 4Q07. Its lack of unicode capabilities and lack of support for languages and address data outside the U.S. will continue to hamper its traction with large multinational prospects. Initial steps toward Unicode compatibility, as well as partnerships to support international address cleansing, are also planned by YE07.
• DataLever's customer base is oriented toward departmental projects or midsize businesses. However, the vendor is beginning to demonstrate success in larger-scale, enterprise-class implementations.

• DataMentors specializes in customer data quality applications, providing matching, linking, standardization and cleansing operations via its DataFuse product, and data profiling capabilities via ValiData. Its partnership with smartFocus enables the vendor to offer campaign management, analytics and mapping capabilities (branded as DataMentors PinPoint). The vendor's roots are in database marketing, with the management team having been involved in large-scale applications of this type for over 20 years.
• Customer references cite accuracy of matching, ease of use, and attractive pricing relative to some of the more prominent vendors in the market as key strengths and reasons for their selection of DataMentors' technology.
• DataMentors demonstrates strong knowledge and experience in the financial services industry, and the majority of its customer base is in this industry.

• DataMentors' focus on customer data quality issues places the vendor at a competitive disadvantage when prospects have broader requirements, including quality issues in non-customer data domains.
• With a small installed base (approximately 60 customers) and limited resources for marketing, DataMentors will be challenged to gain mind share in a market increasingly populated with much larger providers.
• From a product functionality perspective, DataMentors has weaknesses in runtime platform support (Windows is the only deployment option) and in international capabilities due to lack of unicode support and absence of prepackaged functionality for non-English language and non-U.S. address data. Unicode capabilities are planned for delivery by YE07.

• Datanomic is a relatively new entrant in the data quality tools market. Since it was founded in 2001, the vendor has grown to about 60 customers, most of which are in the U.K., some in mainland Europe, and a handful in the U.S. and Asia. As a new player, Datanomic has been able to build its Director platform on modern technology and with an attractive user interface, without any major legacy baggage.
• With the dn:Director and dn:Dashboard products, Datanomic has brought the separate components — such as dn:Clean, dn:Match or dn:Audit — into a comprehensive platform that is Unicode-enabled and runs on many platforms, both in batch and interactive modes.
• Datanomic has a strong focus on the financial industry, with a few clients also in the telecommunications sector. Datanomic products are domain-agnostic and not specifically targeted at customer data, though most implementations are found in a CRM context.

• In addition to its direct sales channel, Datanomic boasts partnerships with Tier 1 system integrators, such as Accenture and PricewaterhouseCoopers. These partners are introducing Datanomic to their customers. They are also using the product in client engagements for initial evaluations of clients' data, though they aren't actively selling the dn:Director product. Despite the international reach of the system integrators, Datanomic has been unable to benefit from it, and has virtually no visibility outside its home market in the United Kingdom.
• While the dn:Director product is built on an SOA, its connectivity is somewhat limited. While Java Database Connectivity (JDBC), Java Messaging Service (JMS) and Web services are supported, adapters for native database access are missing.
• Although the vendor maintains an alliance with Oracle, it does not participate in a sales and marketing "ecosystem" with a number of data integration or business intelligence platform companies, thereby missing out on channel sales opportunities.

• Fuzzy! Informatik is an established data quality vendor in the "D/A/CH" region (made up of the German speaking countries of Germany, Austria and Switzerland). It was founded in 1994, and has its roots in Daimler Benz, which go as far back as 1980.
• With about 350 customers, virtually all of which are in Europe, Fuzzy! Informatik focuses mainly on CRM implementations with name and address profiling and correction, through certified integration with SAP and Siebel. The validation of address, banking and telephone information, as well as blacklist matching, are some of the core capabilities of the various toolsets.
• Fuzzy! Informatik maintains various partnerships with universities in Germany and the U.S., and raises data quality awareness through board membership at the German Data Quality Association (DGIQ).

• The merging and deduplication capabilities are somewhat incomplete, which puts Fuzzy! Informatik at a disadvantage when compared with other vendors. Although both of those functions are supported, the final merging and deduplication through reduction of records is not executed. Potential record duplicates and candidates for merging are merely indicated, and it is left to the end user to act appropriately.
• Address data is the only subject area that is available for profiling, cleansing and enrichment. To become a more all-purpose data quality vendor, Fuzzy! Informatik must expand into areas such as product data or financial data.
• Fuzzy! Informatik currently maintains no major channel partnerships. In order to grow visibility, in terms of both mind share and market share, particularly outside Germany, the vendor has to engage, train and build relationships with system integrators and potential OEM partners that can then be introduced into new accounts.

• Pitney Bowes Group 1 Software continues to focus on its traditional positioning of "Customer Data Quality." The vendor specializes in global name and address standardization and validation, matching-related capabilities (including linking and deduplication), and geocoding. It has significant strength in each of these areas, and although the underlying technology can be considered domain-agnostic, customer data quality applications are Group 1's sole focus.
• Group1 retains a large installed base (over 2,400 customers), making it one of the market share leaders for data quality tools. While the majority of the customer base is in North America, the vendor has established a foothold in Asia/Pacific, where it now has several hundred customers.
• With the significant financial resources of Pitney Bowes, Group 1 continues to expand its capabilities through acquisitions — such as the 2007 addition of MapInfo, which brings further geospatial and mapping services to the portfolio. Also, the vendor continues to fund organic development of its core data quality technology, as evidenced by the recent addition of service-oriented capabilities, SaaS delivery models, and increased depth of integration with the Group 1 Data Flow data integration tools.

• Group 1's focus on customer data places the vendor at a competitive disadvantage when prospects have broader requirements. While it offers data profiling capabilities via its professional services engagements, the vendor does not directly market and sell a data profiling product.
• Lack of clear communication regarding the status of the Group 1 data quality products relative to the existing Pitney Bowes name and address cleansing products continues to create confusion and concern among customers. The vendor needs to publish a clear and consistent product road map showing the plans and milestones regarding integration, support or decommissioning of the various products.
• While Group 1 offers a range of pricing models and options, mainframe-based customers (which represent the core of its customer base) occasionally report challenges in negotiating the cost of upgrades and ongoing support/maintenance.

• Human Inference has been providing data quality solutions to large customers mostly in the European financial services industry (banking and insurance), where the vendor has some long-standing relationships with approximately 220 clients.
• The HIquality components include technology for inspection and profiling, name and address cleansing, matching, merging and enrichment. One of the key differentiators for Human Inference is its maintenance of reference datasets, which serve as a knowledge base for names, addresses and other specific meanings from a variety of contexts. The SOA-based data quality platform also integrates well in heterogeneous environments through relevant technology adapters for Java, .NET, SOAP and vendor-specific adapters for MQ, Oracle and Microsoft databases, among others.
• As one of the larger European data quality vendors, Human Inference has high mind share in the Netherlands, and is increasingly active in other European countries, mostly driven by successful marketing programs. SAP NetWeaver certification indicates the vendor's strategy to invest more in integration with application vendors such as SAP.

• Human Inference focuses almost exclusively on customers in Western Europe through its offices in the Netherlands, Belgium, Germany and the United Kingdom, in addition to a number of distributors in France, Italy and Switzerland. While the direct sales force is successful, the vendor's channel strategy is in its infancy, which will hamper growth.
• Human Inference needs to build a channel and establish partnerships with data integration, process integration and application vendors to extend its presence. That way, it can better compete with the product offerings from large international infrastructure companies that, because of their holistic approach, regularly win deals over small providers.
• The vendor has the technical capability to provide data quality in an SaaS fashion, and other independent software vendors (ISVs) are leveraging the software in a services model. But Human Inference does not provide those data quality services itself through a formal hosted offering or specific branding.

• IBM sees data quality functionality as a key component within its integration portfolio, and has started to deliver data quality products under the Information Server packaging. As one of the best-known brands with worldwide consulting, service, and support functions, IBM is well equipped to position its vision of data quality in organizations worldwide.
• As part of its vision for information-on-demand, IBM built software components named Information Analyzer (the successor of ProfileStage, for discovery, profiling and analysis) and updated QualityStage (parsing, standardization and sophisticated matching). The rearchitected products that were delivered with IBM's release of Information Server underwent significant development to harmonize the user interfaces and administration functionality, and increase ease of use and developer productivity, which were known challenges for previous versions.
• IBM has a broad data management portfolio including extraction, transformation and loading (ETL) tools, federation, replication and metadata, which creates an opportunity and advantage for IBM in cross-selling data quality products, as data quality is increasingly being recognized as critical to data integration activities of all types. Information Analyzer and QualityStage can be deployed as individually selectable modules on top of the metadata-driven Information Server infrastructure, thereby increasing the consistency and interoperability among the various data quality and data integration components.

• Ascential had better brand recognition in terms of data integration and data quality. However, when it was acquired by IBM its strong focus on data profiling and cleansing was somewhat diluted, as it fell under the large umbrella brand of WebSphere and was subsequently repositioned within the new Information Server. Although IBM has the highest overall software brand recognition of all vendors evaluated in this market, it does not exercise a strong positioning as a data quality tools provider.
• The Information Server versions of the data quality products are seeing slow adoption in the market, with virtually no customer references available that use these versions of the products in a production environment. This is mostly because the new product version is still very new and customers with prior versions are early in their planning and analysis stages of the upgrade effort.
• Although the data quality products are service-enabled and IBM has extensive hosting capabilities, it has not embarked on a go-to-market strategy with a service offering of hosted profiling, matching, cleansing or enrichment services, leaving this field open for competitors.

• Informatica stepped into the data quality market with its acquisition of Similarity Systems in January 2006, which itself had acquired the assets of Evoke Software in 3Q05. At that time, the established partnerships between Informatica and Trillium and Firstlogic ended. Through the acquisitions, Informatica now has all the key pieces of profiling, standardization, matching and cleansing. The company has since moved all data-quality-related development into a separate business unit.
• The data quality arena represents a big market opportunity for Informatica, and the vendor's strategy of cross-selling the data quality products into its large installed base appears to be working. In addition, the data quality capabilities further enhance Informatica's strategy for developing a comprehensive data integration and data quality offering.
• The tools have been used in several significant data quality implementations inside large global companies, supporting initiatives such as MDM in addition to traditional customer data quality applications. The Informatica Data Quality product is well positioned for domain-agnostic cleansing, and the Data Explorer is one of the most mature profiling engines in the market.

• Informatica maintains two separate profiling solutions — the PowerCenter profiling option and Informatica Data Explorer. While the company positions these offerings for specific and different scenarios, the overlap in functionality can sometimes create confusion for customers. Informatica plans to unify these into a single offering by 2008, thereby removing any redundancy.
• Organizational integration following the acquisitions has been completed and progress has been made with product integration. However, architectural rationalization of the various data-quality-relevant components into one concise product family is still under way.
• Informatica has traditionally sold to IT buyers. But because data quality improvement requires heavy business buy-in and involvement, in order to raise the data quality products' currently low market share, the Informatica sales force must be trained to sell to buyers on the business side.

• Innovative Systems has the most longevity of the vendors in this market, with a history spanning nearly 35 years. Innovative's i/Lytics platform, released two years ago and now on version 2.1, provides proven capabilities based on its deep experience in customer data matching and cleansing applications. i/Lytics provides strong support for both mainframe and distributed platforms, and enables data quality functionality to be exposed via service interfaces.
• Innovative's customer base (approximately 160 customers) reflects the vendor's strong experience in the banking and insurance industries — these verticals include about two-thirds of the vendor's customers. While most of these customers are in North America, Innovative also supports customers in Europe and is experiencing growth in Latin America (a region in which it has significant experience).
• Complementary to its financial services depth, Innovative continues to focus on its compliance watchlist screening offerings, as there remains strong demand for these in the market. In addition, the vendor increasingly emphasizes the quality of its professional services and the ability to deliver complete solutions rather than just implementations of its tools.

• With a strong emphasis on customer data quality issues, Innovative will be challenged to win new business or expand its presence in existing accounts when multidomain data quality improvement initiatives are required. While the vendor supports data quality activities in other domains, such applications are a small portion of the installed base.
• Given its long history in the market, Innovative's relatively small installed base indicates limited growth in recent years. It has been generally successful in retaining its longtime customers, but will need to increase the pace of new customer acquisition in order to remain competitive.
• Innovative has expanded its product portfolio with the addition of a data profiling product, but it has limited capabilities and a bias toward customer data, rendering it immature compared with the market leaders. In addition, while Innovative's technology can support multilingual data, the lack of full unicode capabilities limits Innovative's ability to compete on a global basis.

• Netrics, a relatively new entrant to the data quality tools market, provides a range of capabilities with a specific focus on matching. The vendor utilizes a machine learning approach to implementing matching and standardization, based on the customer working through a sample set of data in order to train the technology.
• Customer references claim better accuracy in highly complex matching problems as compared to more traditional matching approaches, with a shorter time to implementation because comparatively less "programming" is needed.
• Netrics' technology is essentially an embeddable data quality and matching engine, enabling the deployment of data-quality-related services inside any type of application. This is a significant differentiation from most other vendors in the market, and enables Netrics to focus primarily on an indirect channel strategy with OEM and system integration (SI) partners.

• Netrics' strong emphasis on matching comes at the expense of other data quality operations, such as profiling and address validation, where it has limited capabilities. In addition, the lack of a user interface means the vendor does not provide out-of-the-box functionality for exposing profiling results, matching results or runtime statistics.
• With a small installed base (approximately 100 customers) and limited resources for marketing, Netrics will be challenged to gain mind share in a market increasingly populated by much larger providers.

• Harte-Hanks Trillium Software provides a broad data quality tool suite, including data profiling (TS Discovery), core data quality components (TS Quality) and a new data quality dashboard offering (TS Insight). Its data enrichment capabilities are focused on customer data (addresses, geocoding and watch-list compliance).
• Trillium continues to enjoy strong brand recognition and remains a market share leader with a large installed base of nearly 1,600 customers. The large size of its parent company, Harte-Hanks, provides a solid financial basis relative to many other vendors in this market.
• Trillium continues to expand its focus on alternative delivery models for its data quality capabilities. Its Diamond Data IS offering provides the TS Quality functionality in a hosted model for customers of software as a service (SaaS) application providers such as salesforce.com. Trillium also markets and sells directly its TS On-Demand solution, giving customers a choice of on-premise or hosted deployment of TS Quality.

• Trillium's functionality, marketing and product road map are largely geared toward data quality issues in customer data, and only a small fraction of the customer base is applying TS Quality in other data domains. While customer data quality will remain a mainstay of market demand, Trillium will be increasingly challenged by newer domain-agnostic competitors and pushed by its customers for greater extensibility to other data types. Trillium will expand upon its non-customer data quality support with the introduction of Universal Data Libraries (pre-built functionality for common data attributes including units of measure, currencies and package types) in the forthcoming v11 release, planned for 3Q07.
• Trillium's large customer base is heavily skewed toward North America. To ensure long-term market leadership, the vendor will need to generate significant growth in other regions in response to competition from larger and more globally visible vendors.
• In 3Q06, Harte-Hanks acquired U.K.-based address cleansing specialist Global Address, creating some overlap of functionality with TS Quality. Harte-Hanks has not provided a clear road map indicating how, or indeed if, these redundant capabilities will be rationalized.

• Uniserv is a longtime provider of data quality solutions with a solid track record. Based in Germany, the vendor focuses almost exclusively on customer data, name and address verification, and geocoding. Most of Uniserv's customers are in German-speaking countries, though the vendor also has a large customer base in France, and has also branched out into other European countries and North America.
• Uniserv has more than 60 employees in technical roles, such as development, services and support. This allows its customers to easily and quickly see the benefits of its mailBatch, mailRetrieval and postal applications. A relatively low average software license per implementation also makes it easy for customers to deploy Uniserv software.
• Partnerships with SAP and Oracle, both of which have limited data quality capabilities, give Uniserv a large potential to integrate its software within business applications.

• As many organizations start to view data quality as a domain-agnostic issue, the strong customer data focus will put Uniserv at a disadvantage compared with other providers that market themselves with a broader data quality view toward, for example, product data or financial data.
• Although Uniserv is an established brand for matching, merging, cleansing, or address and bank data verification technologies, increasingly popular areas such as data profiling or quality monitoring are currently limited or unsupported functions. This will make it more difficult for the vendor to compete with other functionally rich data quality platforms.
• Uniserv software is regularly implemented by partners. However, the strong concentration on its direct sales force, and the lack of large international alliances with system integrators and software companies that leverage Uniserv technology as an OEM, puts the vendor under increasing pressure from highly networked and larger international competitors.