The Employee Directory Data Problem Is Fixable. Here's the Proof

Employee data does not improve because you launch an employee directory.

A new interface makes the information easier to browse. It does not fill in a missing manager, add a photo to a blank avatar or correct a phone number that changed two roles ago. The gaps survive the redesign.

In April 2026 we published data on how incomplete employee profiles were across 65 real Microsoft 365 and Entra ID tenants. Those figures were measured at first sync: the moment each organization connected its Microsoft 365 tenant and its employee profiles were read for the first time, before anyone had touched them.

The average profile was 30.8% complete. Only 42% of employees had a manager recorded. One organization in three met our directory-ready threshold.

That is the starting line, and it is the same starting line for every directory on the market. Sync that tenant into any employee directory, org chart tool or intranet people web part and you inherit exactly those gaps. So we went back. In August 2026 we measured the same 65 tenants a second time, after each organization had spent a period running a directory and working on its employee profile data, most of it inside Entra ID and the Microsoft 365 admin center, some of it in OneDirectory itself. The data improved. In some fields, substantially.

Key Findings

• Across 65 Microsoft 365 and Entra ID customer tenants, average employee profile completeness rose from 30.8% at first sync to 44.1% by August 2026.
• Manager coverage across the same 65 tenants rose from 42% to 79% of employees, an increase of 37 percentage points and the largest movement of any field OneDirectory measured.
• The share of organizations meeting OneDirectory's directory-ready threshold rose from 33% to 61%, close to double.
• Organizations with phone coverage above 80% rose from 20% to 39%. Organizations with photo coverage above 80% rose from 27% to 37%.
• The improvement clustered in the fields a directory surfaces and an administrator can correct in bulk. Fields that depend on individual employees, profile photos above all, moved least.
• The 30.8% baseline is not a OneDirectory number. It is what Microsoft 365 hands to any directory on day one. What changed after that is the part worth reading.

What Changed Across the Same 65 Tenants

Source: OneDirectory analysis of the same 65 Microsoft 365 and Entra ID customer tenants. Baseline is each tenant's state at first Microsoft 365 sync, before any cleanup. The second measurement is August 2026. See methodology below.

These numbers do not mean every employee profile is now perfect. They mean incomplete employee data is not a permanent condition of working in Microsoft 365. When you can see the gaps, assign them to someone and work through them, the numbers move.

A word on what this data can and cannot prove. The baseline is clean in one important sense: it is each tenant's untouched Microsoft 365 data at the moment of first sync, so the improvement is real movement rather than an artifact of who we chose to measure. What we cannot do is separate what the software contributed from what the administrators did. Every organization here deployed a directory and ran its own cleanup work over the same window, and the two are tangled together. We are not going to pretend otherwise. The pattern in the numbers is still worth your attention, and the shape of that pattern is the interesting part.

Manager Coverage Went from 42% to 79%

The manager field posted the largest absolute improvement of anything we track.

At first sync, 42% of employees had a manager recorded in Entra ID. Across the same 65 tenants in August 2026, the mean is 79%.

That difference reaches well past the employee directory. Manager relationships decide whether you can generate a reliable org chart at all. They feed reporting structures, management-level analysis, people search, and Microsoft surfaces such as Org Explorer, which shows employees no org chart whatsoever when the manager data behind it is missing.

A blank manager field leaves a structural hole in your organization's digital map. Fill in half of them and half the org chart is still guesswork. It is one of the five missing fields that quietly break an org chart, and it is the one everything else hangs off.

Almost Twice as Many Organizations Are Now Directory-Ready

We classify a tenant as directory-ready when its core profile fields, meaning name, email, job title and profile photo, are each completed for 60% or more of employees. We go deeper on which profile fields are worth holding to a standard.

At first sync, 33% of the 65 organizations cleared that bar. In August 2026, 61% do.

Moving from one organization in three to nearly two in three changes what the directory is for. Below the threshold, employees hit two or three blank profiles, decide the whole thing is stale and go back to asking in a Teams chat. Those are the failure patterns behind an abandoned directory. Above it, they use it, and the directory starts absorbing questions instead of generating them.

The remaining 39% still sit below the line. That is where the next round of work goes. If you need to justify that work, we set out the business case for closing those gaps.

The Improvement Clustered in Fields an Admin Can Bulk-Fix

Phone coverage nearly doubled. Across the 65-tenant cohort, the share of organizations with phone numbers filled in for 80% or more of employees rose from 20% at first sync to 39% in August 2026. Manager coverage rose 37 points. Photo coverage climbed from 27% to 37%, less than a third of the manager movement.

That spread is the most useful thing in this dataset, and it is the clearest signal about what actually drove the change.

Manager relationships and phone numbers get corrected centrally. The data usually already exists somewhere in HR or telephony, so one administrator working from a list of exactly which employees are missing exactly which field can fix thousands of records in an afternoon. Photos cannot work that way. Someone has to take them, or every employee has to upload one, which needs a named owner, an internal campaign or a supply of headshots from HR. There is more on why blank avatars cost more than they appear to.

Read the two rates together and the pattern is hard to miss. The fields that improved most are the fields a directory puts in front of you and an administrator can correct in bulk. The fields that improved least are the ones that need hundreds of individual people to act. Plan accordingly, and treat the photo campaign as its own project with its own owner rather than wondering why one line on the chart refuses to move.

How to Find and Fix Missing Employee Data

Most organizations do not have an employee data problem they can describe. They have a vague sense that the directory is unreliable. Turning that into work someone can actually do takes two things: a number, and a route to the record.

OneDirectory reports a profile completeness score for the tenant and coverage for every individual field, calculated across all visible profiles in the directory. That applies to fields synced from Microsoft and to fields managed in OneDirectory alike, so the picture covers the whole profile rather than one system's slice of it. Instead of "our data is patchy," you get "job title sits at 61% and mobile phone at 34%, and here are the employees behind those gaps."

Fixing them depends on where the field lives. Fields that sync from Microsoft stay managed in Microsoft: from an employee's record in OneDirectory you open that person directly in Entra ID or the Microsoft 365 admin center, make the correction there, and it syncs back into the directory. If you need the step by step, we cover populating the manager field in Entra ID. Other fields can be configured as editable inside OneDirectory itself, and which fields those are depends on how the customer wants it set up.

Either way the loop is the same. Measure coverage, find the specific employees behind a gap, correct the record at whichever source owns it, then measure again next month and see whether the number moved.

Better Does Not Mean Finished

Average employee profile completeness across the 65 tenants rose from 30.8% at first sync to 44.1% in August 2026. That is a gain of 13.3 percentage points, or roughly 43% more completed profile data than these organizations started with.

44.1% also tells you something worth hearing. Employee data quality is an operational discipline with no finish line. Here is what that discipline looks like at enterprise scale.

Organizations have dozens of available profile fields. Some are business-critical. Others are optional, role-specific or impossible to maintain at scale, and chasing 100% across all of them burns effort on data nobody reads.

Pick the fields your employees actually rely on to find people and understand the organization, then hold high coverage on those. A tenant sitting at 44% overall with 95% coverage on the fields that matter is in far better shape than one at 60% spread evenly across everything.

What Actually Moves the Numbers

Installing software does not fix employee data. Nobody in this cohort improved their manager coverage because a new directory appeared in their tenant.

What changed is that the gaps stopped being invisible. These organizations could finally do six things:

• See exactly which fields are missing, and for which employees
• Measure completeness and per-field coverage the same way every month
• Decide which fields are worth chasing and which are not
• Give a named person ownership of each one
• Go straight to the record at its source and correct it
• Check next month whether the numbers went up

A traditional employee directory displays the data it receives. An intelligent directory platform shows you what is missing and takes you to the place you can fix it.

If you want to know where your own tenant sits against these figures, OneDirectory reports the same two measures, tenant profile completeness and coverage per field, on your live Microsoft 365 data.

This dataset is also summarized, alongside verified third-party research on people discovery and workplace data quality, in our employee directory statistics reference.