Most of Your CRM Records Are Probably Half-Empty — Here's How to Measure That

An auto repair shop's service advisor tried to send a follow-up reminder to every customer due for a seasonal check, only to find that roughly a third of the CRM's contact records had no phone number on file, another chunk had an email address but no vehicle information attached, and a handful had nothing beyond a name typed in at some point during a walk-in years earlier. The records weren't duplicates. They were just incomplete, and nobody had ever measured how incomplete until a campaign depending on that data failed to reach a third of the intended list.

Completeness is a different problem than duplication

Data quality conversations tend to focus heavily on duplicates, because duplicates are visually obvious once you go looking for them. Incompleteness is quieter — a record can be perfectly unique and still be nearly useless, missing exactly the field a specific campaign or workflow needs. A CRM can have zero duplicate contacts and still be unusable for a phone-based outreach campaign if half the phone number fields are blank.

Building a simple completeness score

A completeness score doesn't require statistical sophistication — it's a straightforward count of how many of a defined set of important fields are actually filled in for each record, expressed as a percentage or a simple tier. The value of the score isn't precision, it's visibility: turning an invisible, gradual data-quality problem into a number that can be tracked and improved over time.

1. Define the fields that actually matter for how the business uses its CRM — typically name, phone, email, and one or two business-specific fields like service history or referral source
2. For each contact record, count how many of those defined fields are actually populated with a valid, non-placeholder value
3. Express that as a percentage of the total defined fields, and optionally group contacts into tiers — complete, partial, minimal
4. Run this calculation across the whole database periodically, and track the average completeness score over time as a single trackable metric
5. Segment the score by how the contact was acquired (walk-in, web form, referral, import) to identify which intake channel is producing the least complete data

• A field filled with an obvious placeholder — 'N/A', 'test', a repeated dummy number — should count as missing, not present, when scoring completeness
• Required fields at the point of intake (making phone or email mandatory on a form, rather than optional) prevent a large share of future incompleteness before it happens
• Completeness requirements should differ by contact type — a cold lead doesn't need the same field depth as an active customer, and scoring everyone against one universal standard will make early-stage leads look artificially poor

A unique record isn't the same thing as a usable one. Completeness is the quieter half of data quality, and it rarely gets measured until a campaign fails because of it.

Where incompleteness actually gets created

The most common source is a fast intake process — a phone call logged in a hurry, a trade show badge scan capturing only a name and company, a quick walk-in note — where speed is prioritized over completeness at the moment of capture, reasonably so, since forcing a lengthy form on someone during a rushed interaction has its own cost. The fix isn't necessarily demanding more fields up front; it's building a deliberate follow-up step that captures the missing fields shortly after the fact, while the contact is still warm, rather than letting an incomplete record sit untouched indefinitely.

Turning the score into a cleanup priority list

Once completeness scoring exists, it becomes a practical prioritization tool rather than an abstract metric: contacts in the top spending or engagement tier with a low completeness score are the highest-value cleanup targets, since they're customers worth the effort of a manual follow-up to fill in missing details, while a low-completeness cold lead from years ago is a lower priority, if a priority at all. This turns an otherwise overwhelming full-database cleanup into a manageable, ranked list. For businesses running this kind of ongoing data quality process inside their CRM, see how NetWebMedia structures this at the CRM overview.

← Back to all articles