What is fuzzy matching?
Fuzzy matching is a technique used to determine similarities between data such as, company names, contact names or address information. It uses an algorithmic process called fuzzy logic to predict the probability of non-exact matching data to help in data cleansing, de-duplication or matching of disparate data-sets.
This guide provides a comprehensive overview to the role of fuzzy matching in a CRM and customer data management, why it’s important, how it works and how you can implement fuzzy matching in your CRM strategy to improve the efficiency and investment of your CRM.
Fuzzy matching is a powerful tool for transforming messy data to achieve a single customer view, by matching similar or potentially misspelled data entries.
One of the most common examples of fuzzy matching in a CRM is matching customer names. For instance you have a customer named “John Smith” in your CRM database, but when a new customer called “Jon Smyth” enters your system, fuzzy matching techniques can be used to identify and suggest that the new customer might be the same person as “John Smith.” This can help you to avoid creating a duplicate record for the same customer, which is inefficient and leads to data inconsistencies.
Another example of fuzzy matching in a CRM is matching addresses, where variations such as abbreviated street names, typos, or different formats leads to inconsistent data. Fuzzy matching can help identify similar addresses and ensure that the data is accurate and up-to-date.
What are the benefits of using fuzzy matching in your CRM?
Fuzzy matching is a valuable tool for improving the quality and reliability of CRM data, enabling better decision-making, and enhancing the customer experience.
Some of the benefits of using it for matching CRM data include:
Avoid data inconsistencies
Customer data can be entered in different formats, spellings, or variations, leading to data inconsistencies and duplication. Fuzzy matching can help to identify and merge similar or identical records to achieve a single customer view.
Improve master data management
Customer data can be stored across multiple systems and databases, making it challenging to create a unified view. Fuzzy matching can help to identify matching records across different sources, enabling a single view of all customer data.
Better customer identification
Fuzzy matching can help to accurately identify and match customers across multiple touch points, such as in-store or online. This provides a holistic view of customer behaviour, preferences, and engagement across channels, enabling better personalisation and improved marketing ROI
Improved customer insights
Gain deeper insights into customer behaviour and preferences, enabling better decision-making and more effective engagement strategies. This leads to increased customer loyalty, retention, and revenue opportunities.
Types of CRM matching
Exact matching
Matches records that are identical in every way, including spelling, case, and punctuation. This is the simplest form of matching, but it does not account for minor variations or errors in the data.
Phonetic matching
Matches records based on their phonetic sound, rather than their spelling. This can be useful for identifying records with similar-sounding names or addresses, even if they are spelled differently.
Non-Phonetic fuzzy matching
Matches keying errors and transpositions (such as John and Jihn) and reading errors (such as James and Janes)
First name nickname matching
Identifies near-matching person names, initials and even nicknames such as Dave & David, Bob & Robert, Debora for Debra.
Company name matching
Identifies company names with different definitions. For example: DQ Global Limited = DQ Global LTD, Barclays PLC = Barclays Bank
Street address matching
Identifies matching addresses with different abbreviations. For example:10 High Street would match against 10 High Str, 25 Park Road would match against 25 Park Rd.
Fuzzy name matching
Fuzzy name matching is important because names are often recorded with variations in spelling, punctuation, or order. These variations can make it difficult to accurately match names across different data sources. For example, someone may have their name recorded as “John Smith” in one dataset, but as “Jon Smyth” in another dataset.
Without fuzzy name matching, these two names would not be recognised as a the same customer, even though they refer to the same person.
Fuzzy name matching is useful when people have multiple names or aliases. For example, when someone enters one dataset by their full name (your e-commerce system for instance) but uses a nickname in another dataset (for example, your email marketing database). Fuzzy name matching helps to identify these variations and link them together to form a single customer view.
Company record matching
A Company Record is normally constructed of a business name and a legal definition. i.e. GlaxoSmithKline PLC
Therefore one common issue with company records is where the legal entity differs when the company is entered into the CRM.
For example:
- ‘GlaxoSmithKline PLC’ = ‘GlaxoSmithKline Corp’ or ‘GlaxoSmithKline AG’
- 3M = MMM
- GM Incorporated = General Motors Inc.
- NatWest Plc. = National Westminster Inc. = Nat West
Business name elements also present several unique issues which can prevent matches being determined, including; Business name acronyms and abbreviations
It’s common for Companies to use acronyms and abbreviations of their names, and there are many examples of well know companies that do this. I.e. ‘BP’, ‘HP’, ‘P&G’, ‘HSBC, ‘BT’, ‘IBM’, etc.
Without fuzzy matching techniques, none of these businesses would be recognised as the same entity.
Fuzzy matching search and transformation software
Using fuzzy matching software as part of your customer data management strategy will help to avoid non-matching data in your CRM and maintain a single customer view for each and every one of your customers.
The right software makes the connections automatically using sophisticated proprietary matching logic, regardless of spelling errors, unstandardised data, or incomplete information. An efficient system will identify:
- Acronyms
- Name reversal
- Name variations
- Phonetic spellings
- Deliberate misspellings
- Inadvertent misspellings
- Abbreviations e.g. ‘Ltd’ instead of ‘Limited’
- Insertion/removal of punctuation, spaces, special characters
- Different spelling of names e.g. ‘Elisabeth’ or ‘Elizabeth’, ‘Jon’ instead of ‘John’
- Shortened names e.g. ‘Elizabeth’ matches with ‘Betty’, ‘Beth’, ‘Elisa’, ‘Elsa’
If you are using Dynamics CRM our fuzzy matching software DQ for Dynamics™ Search is the ultimate search solution which allows you to easily search across accounts, contacts, and leads with unmatched speed and accuracy.
Using advanced fuzzy matching and data transformation capabilities, it is 3x faster than Microsoft’s built-in search function, and requires no coding or technical expertise to use.
