7Data Diagnostics

Overview of CRM Enrichment

Oracle DataFox algorithms and APIs match your CRM account records to Oracle DataFox company profiles. They help discover data inconsistencies or duplicate account records in your CRM. Enrich your CRM with Oracle DataFox firmographic information and growth signals on your top accounts.

For details on company signals, see Company Signals Overview.

If you have received your CRM Diagnostic report, see CRM Diagnostics Overview for more information on CRM diagnostics.

Clean Databases

Oracle DataFox ingests data from 50+ different data pipelines with millions of primary and secondary sources to build a rich data set of growth companies.

Oracle ensures that the company data is clean before displaying the data to Oracle DataFox company profiles.

Here's an overview of how the Oracle DataFox data business works from the perspective of how data flows through the platform.

DataFox Data Overview

It's necessary to separate correct data from potentially incorrect data. To do so, a series of algorithms detect each type of error and flag anomalies that may or may not be problems.

After it passes all tests, the data is matched to Oracle DataFox company records and ingested. After the data is ingested, internal anomaly detection tools are applied to clean, match, and enrich your CRM.

CRM Clean and Enrich Diagnostics

This article is a detailed deep dive into the CRM Clean and Enrich process, what each of the diagnostic categories means, and how they can be resolved.

For an introductory overview of this process, see CRM Overview.

Example of a Diagnostic Chart

This screenshot shows a sample client CRM diagnosis where Oracle DataFox matched CRM records to Oracle DataFox company profiles.

CRM diagnosis

Category Definitions

Before matching and enriching anything, Oracle DataFox rigorously tests each record for errors. See Overview of CRM Enrichment to know why it’s important to detect anomalies.

There are two types of anomalous categories.

  • Irregular: Something was potentially wrong with the record.

  • Duplicate: Two or more records appear to belong to the same company.

If no anomaly is detected, it means that the company is verified. The company record passed all tests, and no anomalies or duplicates were found.

Anomalous Category: Irregular

This table lists all the anomalous category irregular issues along with descriptions, examples, and actions.

Issue Description Example Action
Irregular: Missing URL The Account submitted didn’t have a URL associated with it.
Irregular: Invalid name The Account name submitted was unintelligible. ##@@$$$%%% You can either ignore these cases (and these accounts aren’t matched), or amend the name and re-submit for matching.
Irregular: Mismatched Query Data We detect any cases where the Account name points to one company and the URL points to another. Often, these cases are just fine, but we flag them to let you decide how to handle them
  • Agricon Global Corporation bayhillcapital.com

  • Arvato Hightec bertelsmann.de

To avoid making errors, we don’t match just on Name or URL because the Name and URL may match to different companies. On one hand, many records are fine as the Company Name and URL are just very different from that particular company. On the other hand, many records aren’t fine as the Name-URL mismatch is an actual problem. Because Oracle DataFox always errs on the side of avoiding incorrect matches, we don’t just "guess" at whether the Name or URL is more reliable.

You can either ignore these cases (and these accounts aren’t matched), or spot check the results and choose whether Oracle DataFox should match to the Name or URL.

Irregular: Email Flag URLs that look like email addresses to prevent the false matching of companies like Google, Yahoo or any other major email domain provider.
  • United united@gmail.com

  • Insight Software insightsoftware@yahoo.fr

You can either ignore these cases (and these accounts aren’t matched), or amend the URL and re-submit for matching.
Irregular: Parked Domain Our matching algorithm visits every URL and detects if a URL redirects to a Parked Domain (for example, GoDaddy.com, HugeDomains.com, etc.).
  • advancedtechnologiesgroup.com >> domainmarket.com

  • altomines.com >> hugedomains.com

Identifying parked domains can also indicate if a company has gone out of business or changed its URL domain.

You can resolve parked domains by:

  • facebook.com/bobs-hardware

  • etsy.com/marys-necklaces

Irregular: Vanity URL Some organizations don’t have their own URL, but rather maintain a profile on another company directory (e.g. Facebook, Wikipedia, Etsy, AngelList, or Crunchbase). We categorize these records as 'Vanity URLs'.

Vanity URLs can indicate a company is small or has a minimal web presence.

Spot check the results. In some cases, the Vanity URL can be replaced by the actual company website URL. In other cases, it's an obscure company without its own website.
Irregular: Invalid URL Your Account URL doesn’t direct to a company website. The URL is either corrupt, formatted incorrectly, missing a character, or not a website at all.
  • http:

  • http://zazzle.cccom

  • http://actel...com

  • nourl

You can either skip the records or find correct URLs and then re-submit to Oracle DataFox for matching.
Irregular: Shortened URL Sometimes, company records contain a URL shortcut, rather than the company's actual URL.
  • bit.ly/abc

  • ow.ly/xyz

You can either ignore these records or find the corrected URL and then submit to Oracle DataFox for matching.

Anomalous Category: Duplicate

Duplicate: Name Duplicate

Description: Two or more different account records share the same or similar name.

There are 3 different cases of Name Duplicates can occur.

Case 1 The Name Duplicates are actually duplicates records for the same company.

For example,

  • Three records all named "Walmart".

Action: You can choose to link or merge the records in your CRM.

Case 2 The Name Duplicates are actually different companies, but just happen to have the same name.

For example,

  • There are multiple companies named "Insight Software".

Action: Ignore. There's nothing actually wrong.

Case 3 The Name Duplicates have fake, dummy, or placeholder names.

For example,

  • Multiple records are named "XYZ", "No Name", "NewCo", or "Test".

Action: Find the correct names or flag the records as dubious.

Duplicate: Redirected URL Duplicate

Description: Our matching algorithm visits each URL to detect if a given website redirects to (1) to a unique URL or (2) the same URL for another Account record.

Example: Two Account records with very different looking URLs redirect to the same page:

  • aamyers.com >> allanmeyers.com

  • americaninfrastructure.com >> allanmeyers.com

Action: You can either (1) update the URLs to the correct company website(s), (2) merge the duplicate records (3) link them to each other if the Accounts are related to each other (e.g. subsidiaries, divisions, or acquisitions).

Duplicate: Post-Match Duplicate

Description: Some account records are duplicates even if the Account names and URLs appear to be different. This can occur because the same company can have multiple names and URLs.

Oracle DataFox is able to detect these Post-Match Duplicates because Oracle DataFox company profiles store all names and domains owned by the company. In the below example, our system knows hotwater.com and aosmith.com are owned by AO Smith, and AO Smith Water Products Company is a division of AO Smith, so these two records are Post-Match Duplicates.

example,

Example

Action: You can choose to link or merge the records in your CRM.

Duplicate: URL Duplicate

Description: Two or more Account records have the same URL domains.

Example:

Example

Action: You can choose to link or merge the records in your CRM.

Duplicate: Name and URL Duplicate

Description: Two different account records share the same Name and URL.

Example:

Example

Action: You can merge or link the two account records together.

No Anomalies Detected: Verified

All Accounts marked as Verified have passed all of the above tests and are potential matches to Oracle DataFox company profiles.

Verified: Automatch

Description: Your records matched to an Oracle DataFox record with high confidence.

Action: No action necessary. Your record is ready to be enriched with Oracle DataFox info.

Verified: Human Matching

Description: While your record passed all of our anomaly tests, it didn’t match to a Oracle DataFox company profile with high confidence.

This means either the account record details point to more than one Oracle DataFox company profile or Oracle DataFox needs to create a company profile.

Because we take every precaution to only deliver high confidence matches, we have our human analysts assess and match each of these records. We can also provide new company profiles for missing records.

Action: No immediate action necessary. We will follow up with you after our analysts have assessed and matched these records.

Verified: Already Matched

Description: For many clients, we match and re-match Accounts multiple times over the customer lifecycle. Any Account records labeled 'Already Matched' indicate the record was matched during a prior matching process.

Action: No action necessary.