Record a per-agent baseline before the training. At 30 days, check behavior on scored calls. At 60 days, check contact, set and show rates per agent. At 90 days, check sold rate and the dollars against the full cost. Compare rates, not counts, against the same lead mix, and write down everything else that changed.
Baseline First, or There Is Nothing to Measure
Most verdicts on training are feelings, because nobody wrote down where the room stood beforehand. Pull the prior 90 days by agent and by month before the visit, and freeze the file.
Use one set of definitions throughout, the ones on our dealership BDC benchmarks page:
- Contact rate: leads where a two-way conversation happened, over workable leads. A voicemail is not a contact.
- Appointment set rate: firm appointments with a date and time, over contacts.
- Show rate: appointments that arrived, over appointments set.
- Sold rate: units sold, over appointments shown.
Add a behavior baseline too: score five to ten calls per agent on the scorecard you will use afterward. If you do not have one, the BDC call scorecard is a starting point.
A Frame for the Check-Ups: Kirkpatrick's Four Levels
Donald Kirkpatrick's four-level model, first published in 1959, is the standard way to think about whether training did anything. The definitions below are quoted from Kirkpatrick Partners. The BDC column is our own mapping, not theirs.
| Level | Kirkpatrick Partners' definition | In a sales BDC (our mapping) | When to look |
|---|---|---|---|
| 1. Reaction | The degree to which the audience "finds an experience or initiative favorable, engaging, relevant, and supportive" | Did agents find the sessions useful? | End of the visit |
| 2. Learning | The degree to which they acquire "the intended knowledge, skills, attitude, confidence, and commitment" | Can each agent deliver the skill in role-play, unprompted? | End of the visit, again at day 30 |
| 3. Behavior | The degree to which they perform "the critical behaviors in its environment" and are "supported in and accountable for" that performance | Do scored live calls show the behavior, such as asking for the appointment with two times? | Days 30 and 60 |
| 4. Results | "The degree to which targeted organizational outcomes occur as a result" | Contact, set, show and sold rates | Days 60 and 90 |
Kirkpatrick Partners advises planning from Level 4 backward: decide which result you want, then which behaviors produce it. For a BDC that means choosing the rate you are trying to move before the training, not hunting afterward for a number that went up.
Jack Phillips's ROI Methodology adds a fifth level, return on investment, where the program's monetary benefits are compared with its fully loaded costs. The ROI Institute describes a required step to "isolate the effects of the program" from other influences, and says benefits not converted to money are reported as intangibles. Both come up below.
Most stores stop at Level 1: the team liked the trainer, so it worked. A happy room and an unchanged call are a common pair.
Day 30: Did Behavior Change?
At 30 days the lagging numbers are still noisy, so look at what agents do.
- Score the same number of calls per agent as in the baseline, picked at random, on the same scorecard.
- Read the two or three behaviors the training targeted.
- Re-run one role-play per agent, cold. An agent who can do it in a drill but not on live calls needs coaching at the desk. One who cannot do it in a drill needs the skill taught again.
- Check the manager, too. Did the weekly one-on-ones happen? Count them.
If behavior has not moved at 30 days, do not wait for day 90 to see whether results appear. They will not. This is the point to call the trainer and to read why BDC training fades.
Day 60: The Rates Closest to the Agent
By 60 days you have a full post-training month of leads that have had time to become appointments and shows. Look at contact rate, set rate and show rate per agent, against that agent's own baseline.
Mind the small numbers. As an illustration only: an agent who reaches 60 people in a month and goes from 18 appointments to 24 has gained six. One good week can produce six. So look at two months together and read them alongside the scored calls. When the behavior score and the rate move together for the same agent, you can believe it. When the rate moves and the calls sound the same, something else is going on.
Day 90: Sold Rate and the Money
Sold rate on shown appointments and units come last, because an appointment set in week six may show in week seven and deliver in week eight. At 90 days, total the added appointments shown and units against the baseline rates applied to the same lead count, then put the gross next to the full cost of the training.
"Fully loaded" is Phillips's term, and it means more than the trainer's invoice: travel, the agents' paid hours off the phones, the managers' time, and any follow-up fees. ROI is conventionally written as net benefits divided by costs, times 100. The BDC training ROI calculator does the arithmetic through the lead-to-sale chain and shows the payback period; what BDC training costs covers the cost side. For the department's return in general, see measuring BDC ROI beyond appointment count.
Controlling for Lead Volume, Season and Everything Else
This is the "isolate the effects" step, and a dealership is a hard place to do it. A new ad campaign, a pay plan change or a new CRM can move the numbers more than any class. You cannot run a clean experiment, but you can avoid the obvious mistakes:
- Compare rates, not counts. More appointments in a month with more leads proves nothing. Set rate on contacts does not care how many leads arrived.
- Hold the lead mix steady. Compare by source. If third-party leads doubled while your own website leads held flat, the blended rates will drop even if every agent improved.
- Use two comparisons. The three months before the training, and the same months a year earlier if the team and sources are comparable.
- Look for a natural comparison group. An agent who missed the training, a sister store that was not trained, or the sales floor's own lead handling.
- Keep a change log. Dated notes on anything else that changed: staffing, hours, pay, lead providers, pricing strategy, CRM settings.
- Read it per agent. A team average can rise because one weak agent left.
The layout in the weekly report an outsourced BDC should send works for an in-house team as well.
When to Conclude It Did Not Work, and What That Usually Means
Give it the full 90 days unless behavior is flat at day 30. Then read the pattern:
| What you see | What it usually means |
|---|---|
| Agents can do it in role-play, but scored calls have not changed | Nobody is coaching between visits. The fix is the manager's routine, not more class time. |
| Scored calls improved, set rate did not | The wrong behaviors were targeted, or the constraint is upstream: contact rate, lead quality, response time. |
| Set rate rose, show rate fell | Softer appointments are being booked, or there is no confirmation process. |
| Shows rose, sold rate fell or stayed flat | The handoff or the floor. That is not a BDC skill problem. |
| Nothing moved and contact rate is low for everyone | Probably not a training problem: coverage, cadence or staffing. |
Training cannot answer a lead nobody is scheduled to answer. If the last row is your pattern, train, fix or outsource is the better next read.
How to Treat Vendor Result Claims
A review of 117 studies of behavior modeling training found effects were largest on learning, smaller on job behavior and smaller still on business results (Taylor, Russ-Eft and Chan, Journal of Applied Psychology, 2005). That covered workplace training in general, not dealerships.
We have not found an independent study that measures the sales effect of BDC training. What exists is vendor case studies and vendors' own result ranges. Pinnacle's Sales BDC Training page prints ranges as well. Treat them, and everyone else's, as claims to test against your baseline, not as a forecast for your store. That page also says trainees learn the KPIs (lead response time, appointment set rate, conversion) and lists weekly or monthly performance reviews and live call audits after the initial training, which is where this kind of measurement should happen with any trainer you hire. More on choosing and using training is in the Sales BDC Training Resource Center.
Frequently Asked Questions
How soon should BDC training show up in the numbers?
Behavior on scored calls should change within the first few weeks. Set and show rates take a month or two to read reliably, because appointments lag leads and one agent's monthly counts are small. Sold units come last. If scored calls have not changed by day 30, the rates are unlikely to follow.
What is the single best measure of whether BDC training worked?
There is not one. The most useful pair is a behavior score on live calls and the appointment set rate on contacts, for the same agent, before and after. When both move together you have a believable result.
Who should do the measuring, the trainer or the store?
The store should own the baseline and the numbers, since they come from your CRM. The trainer can score calls and review the results with you. Agree the definitions and the scorecard together before the training starts so nobody argues about them later.
How do I measure training results if our lead volume changed a lot?
Use rates with fixed denominators and compare them by lead source. Then apply the before and after rates to the same lead count to see the difference in appointments and units. If the source mix changed sharply, report each source separately and avoid a single blended figure.
Summary
Measure BDC training in order. Take a per-agent baseline of scored calls and of contact, set, show and sold rates. At 30 days look for changed behavior on live calls; at 60, the rates nearest the agent; at 90, sold rate and gross against the fully loaded cost. Compare rates by source and by agent, log every other change, and read the pattern when results disappoint. Vendor ranges, ours too, are claims to test.
