Top 3 Use Cases of Solving B2B Database Update and Maintenance Challenges

A Californian B2B data aggregator was looking to update its 50 million-record database on an ongoing basis to remain competitive.

New CXOs, location change, M&As etc. made business data change at an incredible speed. Keeping the database comprehensive, accurate, up to date and sellable was a challenge for the aggregator. Outdated or incomplete data could potentially derail their marketing strategies, undermine customer experience and cripple growth opportunities.

A holistic data update and maintenance program entailed capturing, qualifying and inputting thousands of business records from listed, unlisted, traditional and nontraditional sources. All this translated to more than 150K checked records entered every month! The data sales company partnered with Hitech BPO to support their database building operations.

Data specialists at Hitech BPO developed a robust workflow to seamlessly manage the ongoing, voluminous and omni channel data capture and validation activities. This included an optimal mix of manual interventions and custom tools and macros. Enriching existing records with additional information was an integral part of the solution.

Steps to consistently update customer database included:

• Harvesting company data

  Custom bots and scheduled crawlers were deployed to capture information from public and private sources-both structured and unstructured – across time zones and geographies. The records included basic information such as the size, revenue, or industry of the company as well as value-add company insights related to information on founder name, acquisition amount, IPO launch, revenue, employee increment etc. Manual web research addressed sources which the crawlers could not penetrate.

• Authenticating captured data

  Multi-layered validation checks were used to authenticate captured data against reliable online sources like business directories, news sites, blogs, forums, social media sites, etc. These included manual as well as business rules-driven checks.

• Appending missing data

  The captured data was scrutinized manually as well as run through specific business rules to check for missing and incomplete firmographic, geographic, demographic and behavioral information. Custom bots were used to fetch missing information and what could not be automatically fetched was attended to by a team of web researchers.

The B2B aggregator today owns a cleansed, enriched, verified, and updated database that meets industry accuracy benchmarks. The all-inclusive data on businesses with information on funding and investments, founding members, leaders, mergers and acquisitions, news, industry trends, etc. empowered their business to meet customer expectations and growth targets.

A French data aggregation and market intelligence company offered its global clientele access to gigantic 15-million record company database spanning over 50 attributes and 60 countries. However, keeping this database current, updated and accurate was a massive challenge. Especially since the data sales business and customer experience depended heavily on pumping in two million fresh records every year.

Managing operations with an in house data team was proving to be a heavy drain on financial and human resources. The huge struggle to cope with the volume, time and quality requirements took away focus from core sales and marketing activities.

The data sales company partnered with Hitech BPO to manage their entire range of data acquisition, update and maintenance activities. Data comprehensiveness, data hygiene and data currency were defined as key metrics. The team of data professionals at Hitech BPO designed a hybrid workflow to take care of data collection, validation, enrichment and update. The blend of manual and automated approaches pushed authenticated records into the system.

4 step data authentication process

• Validate and verify existing data

  The 15 million existing records were first screened for clarity and specificity. Fuzzy data was set aside for manual validation. The rest of the data was pushed through a rules-driven funnel for validation against pre-defined sources. The manually validated data was later merged with this auto-validated data pool.

• Addition of new data

  2 million+ new records were captured and pushed into the system using a mix of scheduled crawlers and manual web research techniques. Multi-layered validation and verification checks were administered before the records entered the database. This included calling up relevant contacts to ensure authenticity of data.

• Data Segmentation

  The records were labelled and tagged by assigning industry codes based on client defined logic and hospitality industry standards. This improved search and enhanced client profiling.

• Audit and review

  Programmable macros and scripts were deployed to conduct ongoing authenticity checks and dirty data was flushed out. This kept data decay at bay.

The offshore model and long term partnership enabled the data aggregator to leverage technology gains and apply data hygiene best practices to unlock huge efficiencies.

A US-based B2B data provider wanted to create a comprehensive data repository of existing and deceased attorney members registered with California Bar Council. The challenge was to identify, validate and update 300K plus attorney profiles using information available across multiple sources. They also needed a system to gather national statistics on attorney licensing.

A partnership with Hitech BPO was forged to leverage the blend of automated and manual processes to take the project ahead.

5 step data management process

• Source identification

  A detailed list of sources which were to be used for capturing data was prepared. Authenticity of sources was identified through a mix of manual and rule-based verification processes.

• Data capture

  Custom bots and scripts were developed for parsing data for pre-defined fields. The source structure and complexity were factored in when tailoring the bots. Highly secure sources which could not be accessed by bots, were separated for manual data collection.

• Authentication

  The data fetched through automated and manual processes were merged and pushed through a validation check against pre-defined business rules. A random manual audit further authenticated the data.

• Data enrichment

  Records with incomplete or outdated data were enriched manually by a manual web research and B2B data append process.

• Classification

  Manual tagging and labelling was done and the records were categorized into two segments – licensing and discipline.

The 300,000+ records database of attorneys practicing in California was aggregated in a record time of 45 days. And with an accuracy of over 99%.