Fair agent performance reviews
Jun 8, 2026
8 min read
An agent can look busy and still let good clients drift.
Another agent can look quiet because they’re handling 3 awkward files properly: a seller who needs careful price advice, a buyer who keeps changing budget, and a landlord who won’t approve access times until the last minute.
Agent performance reviews go wrong when they start and end with totals. Calls made. Viewings booked. Offers received. Instructions won. Deals exchanged. Commission written.
Those numbers matter, but they’re blunt. They show activity and outcomes. They don’t always show judgment.
For a newer manager, that’s easy to miss. A dashboard can make a low-quality week look productive if the agent logged plenty of tasks. It can also make careful work look weak if the agent slowed down a risky deal before it wasted a client’s time.
The better review question is simple: did this agent move the right work forward, in the right way, with enough evidence for the branch to trust what happened?
Start with the work the agent actually controlled
Performance tracking gets unfair when every pipeline is treated as equal.
One agent may be covering fresh valuation leads in a hot patch. Another may be nursing older listings, chain issues, price reductions, and buyers who need finance clarity. Compare them only by this week’s outcomes and you’ll reward the cleanest diary, not the best handling.
Start the review by separating 3 kinds of evidence:
| Evidence type | Useful question | Weak version to avoid |
|---|---|---|
| Activity | Did the agent do enough of the right work? | Counting calls without checking who was called or why |
| Movement | Did clients, listings, viewings, or offers move to a clearer next state? | Treating every stage change as progress |
| Judgment | Did the agent make sensible calls when the work was uncertain? | Penalising every delay as poor performance |
This is where real estate pipeline management needs context. A buyer moved from “active” to “paused” may be good work if the agent found out the buyer can’t proceed until a sale completes. A viewing cancelled at short notice may be poor handling if no one told the vendor. The same surface number can mean 2 different things.
Professional standards point in the same direction. The NAR Code of Ethics frames agent duties around client interests, competence, and honest treatment. The Property Ombudsman Codes of Practice put weight on clear records and service standards. A performance review should not become a compliance lecture, but it should notice whether the agent’s work would stand up if a client asked, “what happened here?”
Use a 5-part review with evidence samples
Give each agent 20 minutes. Pick 3 live matters and 2 recently closed or lost matters. Don’t let the agent choose only their best examples.
Then review these 5 areas.
| Review area | What to check | A fair manager question |
|---|---|---|
| Client movement | Active clients have a current need, readiness state, and next action | “Which clients changed state this week, and why?” |
| Listing handling | Vendors or landlords received useful updates, not just activity totals | “What did the client learn from the latest feedback?” |
| Viewing and offer follow-up | Interested parties were chased while the intent was still warm | “Where did timing affect the outcome?” |
| Handoffs | Covering agents could understand the file without searching private messages | “Could someone else pick this up tomorrow?” |
| Record quality | Notes explain decisions, blockers, and promises well enough to trust | “What would be unclear if you were away?” |
This keeps the review away from leaderboard logic. You’re asking whether the agent’s work created reliable movement.
Look for judgment in the messy middle
An agent might:
- Hold back from pushing a weak buyer into an offer because proof of funds is missing
- Tell a vendor that feedback points to price resistance, even though the conversation is uncomfortable
- Reclassify a lead as nurturing because the client is 6 months away, not because the agent has given up
- Escalate a complaint-shaped message before it becomes a formal complaint
- Ask another agent to cover a viewing because they are too close to another negotiation
None of these moments looks impressive on a report. Some reduce short-term activity. But they protect the branch.
Real estate performance tracking should include a small judgment sample. Pick one matter where the agent decided not to push ahead. Ask what they saw, what they recorded, who they told, and what the next check-in is.
If there’s no record, treat that as the issue. The decision may have been right, but the branch can’t learn from invisible judgment.
Review the quality of next actions
An agent can close 40 reminders and still leave the important work weak if the reminders say “chase buyer”, “call vendor”, or “check docs”. Those tasks prove effort. They don’t prove control.
For each reviewed matter, check whether the next action has 4 parts:
- A named owner
- A clear client or property link
- A due time or date
- A reason the action matters
Compare these notes:
| Weak next action | Useful next action |
|---|---|
| Chase Tom | Call Tom by 3pm to confirm whether his mortgage decision affects Saturday’s second viewing |
| Update vendor | Call Mrs Patel after the 2 evening viewings with buyer objections and whether price advice needs revisiting |
| Check offer | Confirm whether Emma’s revised offer includes proof of funds before sending seller summary |
AvaroAI is built around this distinction. Tasks and reminders can sit against the contact, listing, viewing, offer, or event they belong to, so a manager reviewing performance sees whether an agent is creating useful next actions, not just ticking through a busy list.
The product decision is deliberate: a task without its attached matter is a weak performance signal.
Treat clean pauses differently
Some paused work is sloppy. A buyer hasn’t been called. A vendor update is late. An offer has no next step. That needs action.
But some paused work is clean. The buyer is waiting for finance. The seller asked for a week before reducing. The agent has recorded the reason, set the next check, and told the right person.
Use this decision rule:
| If the matter is paused because… | Review it as… | Manager action |
|---|---|---|
| The agent does not know what happens next | Control gap | Define next action before the review ends |
| The client is waiting and no promise is recorded | Trust risk | Clarify owner, deadline, and message |
| A third party is blocking progress and the check-in is dated | Clean pause | Leave it alone unless the date is stale |
| The agent made a judgment call and explained it in the record | Coaching sample | Discuss the reasoning, not just the result |
| The record says “waiting” with no reason | Hidden risk | Ask for context or reclassify the matter |
New agents often need help separating a genuine pause from avoidance. Experienced agents often need a reminder that good judgment still has to be visible.
Keep private context private, but make branch risk visible
Commission details, sensitive seller comments, negotiation strategy, and personal circumstances may need tighter access. But branch risk still needs to be visible: who owns the matter, what was promised, what is blocked, and when the next action is due.
That is why AvaroAI separates shared operational visibility from unrestricted note access. A principal can review team activity and pipeline health without turning every file into an open notebook. The aim is practical: managers should see enough to coach, cover, and intervene, while agents still have proper boundaries around sensitive context.
Contact records matter here too. Interest level, price range, moving timeline, role, source, and custom fields tell a better story than a raw lead count. A quiet buyer with a clear timeline and accurate preferences may be healthier than 10 vague new leads with no next action attached.
End with one change the agent can make this week
End each review with one measurable adjustment:
- Add a reason to every paused active client
- Record price range and moving timeline before booking a second viewing
- Turn every vendor update into a dated next action
- Clean handoff notes on the 5 oldest live matters
- Separate nurturing contacts from ready-to-act clients before Friday
- Add proof-of-funds status before sending offer summaries to sellers
Check it the following week against real records, not memory.
Production volume has its place. RealTrends Verified shows how much the industry pays attention to it. Inside a branch, principals need a closer view: whether agents move clients honestly, keep promises visible, protect handoffs, and record enough context for the business to trust the work.
That is the review habit worth building.
Related reading
Disclaimer: This page may contain AI-assisted content. The information is provided solely as a general guide and may not be correct, complete, or current, including, but not limited to, our full or applicable service offerings. While we strive for accuracy, no guarantee is made regarding correctness or completeness, and no expectation should be made as such. Please contact us directly to confirm any details before utilizing our service.
