BDC Agent Scorecard Template

The categories worth weighting, the sheet itself, and where every number on it comes from so nobody has to self report.

Most dealership BDC scorecards are activity sheets wearing a costume. Calls made, texts sent, emails out, hours on the phone. They get built because those numbers are the easiest ones to pull, and they are close to useless because not one of them tells you whether the customer had a good experience or whether the store got an appointment out of it.

The test is simple. If an agent could raise their score without setting a single additional appointment, the sheet is measuring effort rather than work. Publish dial counts and dial counts go up the following week. Appointment counts stay exactly where they were. People optimize whatever you hang on the wall, which is an argument for being careful about what you hang.

A scorecard worth running does three things. It measures outcomes the agent genuinely controls, it measures the quality of the contact and not only the fact of it, and every figure on it can be pulled from the system without anyone typing their own number into a spreadsheet. The moment an agent reports their own performance, the review turns into a negotiation about the number instead of a conversation about the work.

What follows is a template you can put in front of your team this week, the weights we would start with, where each number comes from, and an honest account of what the whole exercise cannot tell you.

Five is the right number. Three is too coarse to coach from and ten turns into a compliance exercise nobody reads.

Speed to first contact. Minutes between the lead arriving and a human attempt going out, measured on the attempt rather than on a task being marked complete. This one belongs first because it is the single input the agent controls most directly and it decays fast.

Contact rate. The share of assigned leads that reach a live two way conversation, by phone, text or email. Not attempts. A conversation. An agent with a low contact rate is either working the wrong hours, using one channel, or giving up after two tries, and each of those has a different fix.

Quality of the conversation. Did the agent ask what the customer actually wanted, confirm the vehicle, and ask for a specific time rather than a vague visit. This is the category most stores skip because it feels unmeasurable. It is not, and the section below explains how to score it without listening to every call.

Appointments set, shown and sold. Three separate numbers, because an agent who sets many and shows few has a confirmation problem rather than a setting problem.

Follow up persistence. How many touches a non responsive lead receives before the agent stops, and whether the cadence spans channels and days. Most stores quit far earlier than the honest answer requires.

Weights are a starting point, not a rule. If your store lives on phone ups, shift weight toward contact rate. If you buy a lot of internet leads, speed matters more.

Score each category out of one hundred, multiply by the weight, and total. Publish the total and the six components together. A single number without the components invites argument, and the components without a total invite cherry picking.

This is the part that decides whether the scorecard survives past month two. If pulling it takes a manager three hours, it will be run twice and then quietly abandoned.

Speed to first contact comes from comparing the lead inbox arrival timestamp against the first outbound call or message on that record. Both live in the same system, so the subtraction is not a project. Contact rate comes from the call logs and the SMS threads: a call with meaningful duration or a thread with an inbound reply counts, a voicemail and an unanswered text do not.

Appointment counts come from the appointment records themselves, and show rate needs the status updated honestly, which is a management habit rather than a software feature. The deal visit log helps here because it records that the customer actually arrived.

Follow up persistence comes out of the activity reporting. There are three reporting layers in the platform, an activity report, a company report and a management roll up, and for this exercise the activity view is the one you want because it is per user and per lead rather than aggregated.

If you want the standing version of this rather than a monthly pull, the BDC performance dashboard page covers keeping these numbers visible daily instead of reconstructing them at month end.

Conversation quality is where most scorecards give up, and it is the category with the most coaching value in it. The reason managers skip it is arithmetic: a BDC agent might handle sixty calls a week, and nobody has thirty hours to listen.

Call recording with transcription changes that math completely. A twelve minute call becomes something a manager reads in about thirty seconds, and reading is roughly ten times faster than listening. Pull five calls per agent per week, read the transcripts, and score against a short rubric. Five is enough to see a pattern and small enough that you will actually do it.

Keep the rubric to five yes or no questions. Did the agent confirm the specific vehicle the customer asked about. Did they ask at least one question about the customer's situation rather than only pitching. Did they ask for a specific day and time instead of inviting the customer to come by. Did they confirm the appointment in writing before hanging up. Did they set the next step if the answer was no.

Score out of five, multiply by twenty, done. The value is not the number. It is that you are now coaching from what was said instead of from what the agent remembers saying, and those two are rarely the same story. Our page on dealership call transcription covers the mechanics of getting there.

A scorecard nobody discusses is a report. The review is where it becomes management.

Weekly, fifteen minutes per agent, same day and time. Longer than that and it becomes a performance review, which people brace for. Shorter and it becomes a number read out loud. Fifteen minutes is enough to look at one category, read one transcript together, and agree one thing to change before the next session.

Pick a single category per week per agent. An agent with a speed problem and a persistence problem does not get both in one sitting. Fix the one that costs the most, confirm it stuck the following week, then move.

Publish the totals to the team but hold the transcript detail one to one. Public totals create healthy comparison. Public criticism of a specific call creates an agent who stops taking risks on the phone, which is the opposite of what you want.

Once a month, look at the sheet across the whole team rather than agent by agent. If everybody is weak in the same category, that is a process or staffing problem and coaching individuals will not touch it. Persistent low contact rates across a whole team usually mean coverage hours do not match when your customers are reachable, and that gets fixed on a schedule, not in a one to one.

Assume it will be. Not because your people are dishonest, but because every measurement system creates an incentive and people follow incentives without deciding to.

The classic is appointment inflation. Set rate climbs, show rate collapses, and it turns out agents are logging soft maybes as appointments. Scoring set and shown as separate categories mostly kills this, because inflating the first damages the second.

The second is cherry picking. An agent quietly concentrates on the easy leads and lets the harder ones age out. You catch it by looking at worked percentage of assigned leads alongside the outcome numbers, not instead of them.

The third is speed theater. A one word text sent thirty seconds after the lead lands, which scores beautifully on speed to first contact and accomplishes nothing. Reading a sample of first messages catches this in about a minute, and it is worth doing the first month you run the sheet.

The fourth is the abandoned lead that never gets marked abandoned, so persistence looks fine because the trail simply stops without a status change. Measuring median touches before the last activity on a record, rather than trusting a status field, closes that gap.

Running the same sheet on a two week hire and a four year veteran produces a demoralized hire and no useful information.

For the first thirty days, score only two categories: speed to first contact and follow up persistence. Both are habits rather than skills, both are entirely within a new agent's control, and both are the foundation everything else sits on. Conversation quality gets read and discussed but not scored, because scoring a skill somebody has not been taught yet is just a way of telling them they are bad at their job.

Days thirty to ninety, add contact rate and conversation quality at half weight. After ninety days, the full sheet.

Keep the transcripts from the first two weeks. Reading them back to an agent at day ninety alongside a current call is the single most effective coaching moment in the whole system, because progress on the phone is invisible to the person making the calls. They hear themselves every day and notice nothing. A side by side makes it obvious.

If you are still building the team rather than measuring it, BDC software for dealerships covers the tooling side and response time SLAs covers setting the speed target in the first place.

Two honest limits, because a measurement system oversold is a measurement system distrusted.

First, it cannot separate agent performance from lead flow. An agent handed leads from a channel that produces long research cycles will show a worse contact rate than one handling phone ups, and neither number is about the person. Segment the sheet by lead source before you draw conclusions about people, and never rank agents across different sources on one list.

Second, none of this predicts sold units on its own. We cannot guarantee that a scorecard raises your close rate, and any vendor who tells you otherwise is selling. What it reliably does is make the difference between agents visible, make coaching specific, and make it obvious within a week rather than a quarter when something has slipped.

Two more things worth remembering. Customer behavior is not agent failure: a shopper asking about a vehicle well above their budget is a normal lead, and what to expect from car sales leads explains why penalizing an agent for that mix is unfair. And a scorecard is only as good as the data underneath it, which is why every number on the template above has a system source rather than a person attached to it.

How often should we score BDC agents?

Pull the numbers weekly and review them weekly in a fifteen minute one to one. Monthly is too slow to correct a habit before it sets, and daily turns into surveillance that people learn to perform for rather than work under.

How many calls should we review per agent?

Five per week is the practical number. With call recording and transcription a manager reads five transcripts in a few minutes, which is why the sample actually gets pulled instead of being skipped like listening to recordings usually is.

Should activity counts be on the scorecard at all?

Keep them visible but unweighted. Dials and messages sent are useful context when you are diagnosing a low contact rate, and they are harmful as a scored category because they rise the week you start publishing them without anything else changing.

What is a reasonable speed to first contact target?

Set the target against your own current median rather than an industry number, then move it. Publishing a target the team has never once hit teaches them the sheet is theater. Halve the median, hold it for a month, then halve it again.

Can the scorecard be pulled automatically?

Most of it. Speed, contact rate, appointments and persistence come from the lead inbox timestamps, call and message logs, appointment records and the activity report. Conversation quality still needs a human reading a small sample against a rubric.

Will a scorecard improve our appointment show rate?

We cannot guarantee any outcome, because show rate also depends on your traffic, your market and your confirmation process. What scoring set and shown separately does is make a confirmation problem visible immediately instead of hiding inside a healthy looking set count.

Fifteen minutes and we will show you where speed, contact rate, transcripts and appointment status actually live in the platform. Month to month, no long term contract. Call 844-376-2274.