1. Home
  2. Blog
  3. How to Measure Sales Training Effectiveness
Sales Training· South Africa

How to Measure Sales Training Effectiveness

Learn how to measure sales training effectiveness with a baseline and four questions: liked it, practice scores, real calls and results in rand, in four steps.

··20 min read
Woman in a blazer reviewing a tiered sales training evaluation chart on a standing desk in a glass-walled corporate reception area, red marker raised mid-annotation.

AI-generated

TL;DR: Quick Answer

Here is how to measure sales training effectiveness: take a baseline before the course, then ask four questions in order: did they like it, did practice scores rise, did real calls change, did results change. Measure readiness and results, not hours completed, and compare each answer with the baseline.

  • Teach and practise the six value principles of selling first. You can only measure a method that people were actually taught.
  • In a made-up example, R45,000 of training that adds one sale a month worth R9,000 in first-year gross profit returns R108,000 a year, which is a 140% return.
  • Score the same ten-point opening before and after training. Seven out of ten is a good opening, and a tough buyer needs eight or nine.

If you sell, or you manage people who sell, the question is simple: did the training change what your people do on the phone, and did that change put more rand in the bank? This guide shows how to measure sales training effectiveness in that order, using a baseline you can take this week with no special software.

The method behind it is Mike Steenkamp's. Good sales training gives you a method you can explain in a sentence, then makes you practise it out loud, with feedback, until it sticks. Measuring sales training is the same idea turned around: check that it stuck, then check that it paid.

Key Takeaways

  • Take a baseline first: one ten-point opening score per rep, five scored real calls per rep, and four business numbers in rand. In the made-up example below, R45,000 of training needs to return only R45,000 to break even.
  • Ask four questions in order: did they like it, did practice scores rise, did real calls change, did results change. Stop trusting any measure that skips the order.
  • ShiftMate Sim Studio gives a team somewhere to practise calls out loud on AI buyers, and every practice call gets a report with one thing to fix first, which makes the second question easy to answer.

How to Measure Sales Training Effectiveness: Take a Baseline First

The first step in how to measure sales training effectiveness happens before the training starts: write down where your people are today. A baseline is a set of numbers taken before the change, so that afterwards you can say "up from" or "down from" instead of guessing.

Most teams skip this, then ask three months later whether the course worked. Nobody can answer, because nobody knows what normal looked like. Taking the baseline costs one afternoon and a few recorded calls.

Keep it small. Five items are enough, and the table below is yours to copy into a spreadsheet or onto a whiteboard.

What to record How to record it What good looks like
Opening score Each rep practises the same opening scenario once. Score it out of 10 on the opening checklist. Seven out of ten is a good opening. A tough buyer needs eight or nine.
Real-call score Listen to five recent real calls per rep and score each on one scorecard. The same five-call method used every time, so the scores can be compared.
Next step agreed Count the calls that end with information to be sent and a set time to talk again. A share that rises. You do not close on the first call.
Principle fit Note which reason to buy the rep led with, and whether it suited the buyer's role. One or two principles that fit the role, not all six in one breath.
Business numbers Meetings booked, quotes sent, sales won and the average first-year gross profit per sale, in rand. Written down for a stated period before training starts.

Two rules keep the baseline honest. Use the same scenario and the same marker every time, because a different buyer or a different marker changes the score for reasons that have nothing to do with training. And record the date, because every later number is a comparison with this one.

For the scorecard itself, you can start from the sales call scorecard template and trim it to the items your method actually teaches.

Why is hours completed the wrong way of measuring sales training?

Hours completed only tells you who sat in the room, so it is the wrong way of measuring sales training: it measures attendance, not whether anyone can sell any better.

A rep can sit through eight hours of slides, tick every box and still open a call with "Hi, I'm calling to see if you're interested in our product." The hours are real. The readiness is not. Readiness means the rep can do the thing, out loud, to a buyer who is thinking "convince me".

Good sales training metrics fall into two groups, and both beat hours completed:

  • Readiness: can the rep run the opening, show value against what the buyer uses now, agree and redirect an objection, and earn a next step?
  • Results: more meetings, more quotes, more sales, more gross profit per rep, counted in rand.
A common measure What it really tells you A better measure
Hours of training completed Who attended Opening score out of 10 on a practised call
Quiz marks on the method Who remembers the words Scorecard on five real calls per rep
"Great course" feedback forms Who enjoyed the day Share of calls that end with a next step agreed
Total calls made How busy the team was Meetings booked and sales won per rep, in rand

There is also a practical reason to measure. In a Salesforce survey of sales professionals, only 26% said they receive one-to-one coaching at least weekly. That is a survey from outside South Africa, so treat it as a prompt rather than a local benchmark, but it fits what many managers admit: feedback after training is the thin part.

If feedback is thin, the scores are your feedback. A rep who can see "4 out of 10 last month, 8 out of 10 this month" is being coached by the data, even in a busy week.

Did they like it? Question one

Ask whether they liked it, but treat the answer as the weakest evidence you will collect: it tells you about the day, not about the calls that follow.

Liking still matters a little. A rep who found the training confusing or insulting will not practise afterwards, and practice is where the skill comes from. So ask, but ask questions that test understanding as well as mood. Three are enough:

  • In one sentence, what is the method? (A rep who cannot say it in a sentence cannot use it under pressure.)
  • Which of the six value principles will you lead with on your next call, and why does it suit that buyer?
  • Which step did you find hardest, and what would help you practise it?

The six principles are Safety, Quality, Ease, Speed, Volume and Price: the reasons a business buys. A rep who can name them and say which one fits the buyer in front of them has understood the core of the training. A rep who lists all six on every call has not.

The eight steps of the method give you a second quick test. Ask each rep to put the steps in call order without notes, using the figure below as the answer key.

The sales method: eight steps, in call order

  1. Talk to the right person
  2. Earn the right to talk
  3. Remember the default answer is no
  4. Prove value with the six value principles
  5. Show it's safe
  6. Give a real reason to act now
  7. Agree and redirect
  8. Earn the next step
Source: ShiftMate sales training guide, by Mike Steenkamp.

Do not stop here. A happy room and an unchanged call are common together. The next question is the first one that measures skill.

How do you measure sales training effectiveness with practice scores?

You measure sales training effectiveness with practice scores by having every rep practise the same scenario out loud before the training and again after it, then scoring both attempts against the same checklist.

This is question two: did practice scores rise? It is the cleanest measure you have, because you control the buyer, the offer and the scoring. On real calls, a rep can have a bad day because the buyer was in a bad mood. In practice, the only thing that changes is the rep.

The practice link is the part of the method the sales training guide explains under practice: say it out loud, get feedback, repeat until it sticks. Reading about an opening and saying one to a stranger are different skills, and only the second one shows up in your score.

Score the opening out of 10

The opening checklist is worth ten points. Seven out of ten is a good opening, and a tough buyer needs eight or nine. Note that one item, what your company does, earns nothing on its own: it must be said, but it scores 0. The item worth the most is why the call matters to the buyer, at 3 points.

The opening checklist: what each part is worth (out of 10)

Who you are 1
Your company 1
What your company does (must be said) 0
What the call is about 2
How you got their details 1
What you want from this call 2
Why it matters to them 3
Source: ShiftMate sales training guide, by Mike Steenkamp.

ShiftMate Sim Studio

Practise this on an AI buyer who starts at no

Your team practises sales calls out loud on AI buyers built from your own offer, and every practice call gets a report with one thing to fix first.

ShiftMate is rated 4.5 out of 5 on Google (2,156+ Google reviews)

Here is how the score moves in a made-up example. Grant sells printer toner and stationery for a made-up firm called Dassiebos Stationery. Before training, his practised opening to an office administrator went like this:

"Hi, Grant from Dassiebos Stationery. We sell toner and stationery at great prices. Do you have a few minutes?"

Scored against the checklist: who he is (1), his company (1), what the company does (said, but 0). Nothing on what the call is about, how he got her details, what he wants, or why it matters to her. His baseline is 2 out of 10.

After practising the opening out loud with feedback, the same scenario sounded like this (also made up):

"Hello Mrs Naidoo, this is Grant Kemp from Dassiebos Stationery. I'm calling about how your office reorders toner. I'd like ten minutes next week to show you a way to cut the time your team spends chasing toner orders, and to send you a one-page comparison first. Would Thursday suit you?"

Who (1), company (1), what the call is about (2), what he wants (2), why it matters to her (3): 9 out of 10. He lost the point for how he got her details. That is a fixable gap, and it is only visible because the score exists. Notice also that Grant led with Ease for an office administrator, which is the right place to start for that role.

Repeat the measure, because people forget

Score once right after the training and you will see the best version of your team. Score again two and six weeks later, because skills fade without practice. A 2015 replication of Ebbinghaus' forgetting curve, in which one subject spent 70 hours learning and relearning lists at intervals from 20 minutes to 31 days, produced results similar to the 1880 original. It was a study of memorised word lists in a lab, not of selling, but it is a fair reason to measure again rather than once.

There is also evidence that practising in simulations works for complex skills. A meta-analysis of 145 studies in higher education found a large positive effect (g = 0.85). Those studies were in classrooms and clinics, not sales floors, and they are not South African, so use the finding as encouragement, not as a promise for your team.

What to practise, and where

Practise the opening first, because it is the most measurable and the most fixable. Then add a price objection and a request for a next step. Your reps can practise on AI buyers in the sales call simulator, or run live pairs using the scripts in the sales role play examples. Either way, keep the scenario identical between the baseline and the retest.

Did real calls change? Question three

Real calls have changed if the same rep, scored on the same scorecard against the same method, now opens better, matches the right reason to buy to the buyer, and ends more calls with a next step.

This is the question that separates a good practice score from a better salesperson. Plenty of reps shine in role play and revert on a real call, when a gatekeeper is rude or the buyer says "send me an email". So you must go and listen.

A simple routine for how to measure sales training effectiveness on live calls

Here is how to measure sales training effectiveness on real calls without drowning in recordings:

  1. Pick five calls per rep from the week before training. Pick five more from weeks three to four after it.
  2. Score each on the same scorecard: opening points, which principle the rep led with, how they handled the first objection, and whether a next step was agreed.
  3. Average each rep's five scores before and after. Compare the team average, and look at each rep separately, because the team average can hide one person who got worse.
  4. Share the numbers with each rep in a short one-to-one. Pick one thing to fix first.

Five calls per rep is small. It is not a statistical study. It is enough to hear whether the opening changed and whether the rep is still reading from the old habits, and that is all you need at this stage.

Check the principle fit, not only the polish

A smooth call that leads with the wrong reason to buy still fails. The reason depends on who you are speaking to. The figure below shows what to lead with by the buyer's role: office staff care about Ease and Speed, finance cares about Price, Safety and Speed, owners care about Price and Volume, and so on.

What to lead with, by the buyer's role

Buyer's roleLead with
Admin and office staffEase, Speed
FinancePrice, Safety, Speed
OwnerPrice, Volume
Marketing and salesVolume, Quality
OperationsSpeed, Ease, Quality
ProcurementPrice, Quality
Source: ShiftMate sales training guide, by Mike Steenkamp.

Then count what the rep actually put forward. One research paper on customer value propositions found that listing every benefit, or every difference from a competitor, is weaker than what the authors call a "resonating focus": the one or two differences that matter most to the customer, shown against the customer's next best alternative and backed with evidence (Anderson, Narus and van Rossum, Harvard Business Review, 2006). So on your scorecard, a rep who leads with one or two fitting principles scores higher than one who reads out all six.

A South African example: business insurance

In some industries one principle acts as a gate. In insurance, nothing else counts until the buyer is sure the cover is at least as safe as what they have now. Here is how much each reason to buy counts in that industry.

Insurance: how much each reason to buy counts (out of 10)

Safety 4
Quality 1
Ease 2
Speed 1
Volume 0
Price 2

Safety gate: nothing else counts until the buyer is sure it is safe.

Source: ShiftMate sales training guide, by Mike Steenkamp.

Ravi sells business cover for a made-up broker called Mooirivier Business Cover. He is calling Johan Pretorius, the finance manager at a small made-up logistics firm. On a scored call, here is the difference the scorecard should catch (both lines made up):

What Ravi says Scorecard result
"We can save you R6,000 a year on your premium, Johan." Led with Price before the safety gate. Johan asks "is it the same cover?" and the call stalls.
"Johan, before we talk price: I'd like to show you, line by line, that this cover protects your vehicles and stock at least as well as what you have now. Then we can look at the yearly cost. Can I send you a side-by-side and call you on Thursday?" Safety first, Price second, a set time to talk again agreed. Gate passed, next step earned.

One practical note: insurance, financial services, healthcare and software are heavily regulated, and what you may say to a buyer in them is not for a sales trainer to decide. Check what applies with the person responsible for compliance in your business, and build that into your scorecard.

Check the objection and the next step

Two more items belong on every real-call scorecard. The first is how the rep handled the first objection. The method is to agree with the objection, then redirect to what you do better, and to be honest about what switching involves. Listen for both halves. The objection handling guide has eight answers you can score against.

The second is the next step. You do not close on the first call. A good call ends with information to be sent and a set time to talk again. Count how many of each rep's scored calls end that way before and after training.

How do you measure sales training ROI in rand?

You measure sales training ROI by taking the extra gross profit the training produced over a year, subtracting what the training cost, and dividing the result by what the training cost.

This is question four: did results change? It is the last question for a reason. Results depend on the market, the season and the stock on the shelf as well as on the training, and they arrive later than the scores do. If you ask question four first, you will credit or blame the training for things it did not do.

A made-up worked example

Karin manages six reps at a made-up office-supplies firm called Dassiebos Stationery. Everything below is invented to show the sums, not to predict your result.

  • The training, practice time and scoring cost R45,000 all in.
  • Before training, each rep booked 12 meetings a month. After, each books 13: one extra meeting per rep, so six extra meetings a month.
  • One meeting in six becomes a sale, so the team wins one extra sale a month.
  • Each sale is worth R9,000 in first-year gross profit.

One extra sale a month is 12 a year, which is R108,000 of first-year gross profit. Subtract the R45,000 cost and you get R63,000. Divide R63,000 by R45,000 and the sales training ROI is 1.4, or 140%.

Line Made-up figure
Extra gross profit over a yearR108,000
Cost of training and practiceR45,000
Net gainR63,000
Return (net gain divided by cost)1.4, or 140%

Count first-year gross profit, not turnover, and count it over the same twelve months for the cost of the training. If the training also helped to keep a customer for a second year, leave that out of the headline figure and mention it separately. A cautious number you can defend is worth more than an exciting one you cannot.

Give the results time to arrive

Do not judge the rand figure in week three. Sales cycles take time, and new habits take time to show up in the pipeline. The Bridge Group, in a 2026 survey of 158 B2B companies, found that account executives took 6.2 months to reach full productivity and that 48% of reps were at quota. Those are B2B companies in a survey outside South Africa, so use them as a sense of how slow ramp can be, not as your target.

A sensible rhythm is practice scores at weeks 0, 2 and 6, real-call scores at weeks 0 and 4, and business numbers monthly for six months. Then you have a story: the opening score rose first, the real calls followed, and the meetings and sales came after.

Match the numbers to the chain

The four questions work as a chain. If the practice score rose but real calls did not change, the problem is transfer: reps are not using the method live, so they need more practice and a manager listening in. If real calls changed but results did not, look at lead quality, price or the product, because training cannot fix those. Knowing where the chain breaks is the real reward of measuring sales training this way.

What this does not solve

This approach does not prove cause, it does not fix a weak offer, and it does not replace a manager who listens to calls.

  • It cannot prove the training caused the result. A before-and-after comparison is evidence, not proof. A new price list, a quiet month or a loyal customer returning can all move your numbers. Write down anything else that changed.
  • Small teams give noisy numbers. With three reps and five calls each, one bad day moves the average. Look at the direction over several months rather than a single figure.
  • A practice score is not a real call. A rep can master the opening in practice and still freeze with a hostile buyer. That is why question three exists.
  • Scorecards do not fix a weak offer. If your product is not safer, better, easier, faster, bigger in volume or cheaper than what the buyer uses now, no opening will change that for long.
  • The research is from elsewhere. Most of the studies cited here come from the United States or Europe. They are reasons to test the idea in your own team, not South African guarantees.
  • Marking takes honesty. If the same manager marks and also gets measured on the result, scores drift upwards. Have a second person mark a sample.

It also does not tell you who to hire. That is a separate job, and it is one reason ShiftMate exists.

How ShiftMate and ShiftMate Sim Studio help

ShiftMate Sim Studio is where a sales team practises calls out loud on AI buyers built from the company's own offer. Every practice call gets a report with one thing to fix first, so each rep and manager always has a single clear next job, and you can check whether that thing is still the first fix a week later. Use it for the baseline and the retests in question two, and read the AI sales training for teams page for how it fits together. ShiftMate is the South African marketplace where businesses hire call centre agents who have already been assessed.

Three internal pages round this out: the sales training guide to the six value principles for the method, the telesales script examples for lines to score against, and the cold calling tips for South Africa for the opening in a first call.

Frequently asked questions

How do you measure sales training effectiveness?

Take a baseline before the training, then ask four questions in order: did they like it, did practice scores rise, did real calls change, and did results change. Use the same scenario and the same scorecard each time, and measure readiness and results rather than hours completed.

What are the best sales training metrics?

Start with an opening score out of 10 on a practised call, a scorecard average on five real calls per rep, the share of calls that end with a next step agreed, and business numbers in rand: meetings booked, quotes sent, sales won and first-year gross profit per sale. These sales training metrics cover both readiness and results.

How do you calculate sales training ROI?

Take the extra first-year gross profit the training produced, subtract the total cost of the training and practice time, and divide by that cost. In a made-up example, R108,000 of extra gross profit against R45,000 of cost gives R63,000 net, a return of 140%. Be cautious about how much of the gain you credit to the training.

How long before you see results from sales training?

Practice scores can move within days, real-call scores within three to four weeks, and sales numbers usually take months because sales cycles are slow. In a 2026 Bridge Group survey of 158 B2B companies, account executives took 6.2 months to reach full productivity, though that is a non-South African benchmark.

Can you measure sales training without a CRM?

Yes. A shared spreadsheet is enough. Record the date, rep, call score, whether a next step was agreed, and the monthly meetings, quotes and sales for each rep. The skill is not in the software but in scoring every call the same way and writing the numbers down on a regular day.

How many calls should you score per rep?

Five real calls per rep before the training and five after is a workable start. It is enough to hear changes in the opening, the reason to buy and the next step, but it is not a statistical study, so look at the trend across several months rather than one score.

Should I measure hours of training completed?

Record it if your business needs it, but do not treat it as a result. Hours completed shows attendance, not whether a rep can open a call, match a reason to buy to the buyer, or agree a next step. Measure the opening score, the real-call scorecard and the business numbers in rand instead.

ShiftMate Sim Studio

Practise this on an AI buyer who starts at no

Your team practises sales calls out loud on AI buyers built from your own offer, and every practice call gets a report with one thing to fix first.

See AI sales training

Sources & References

  • https://www.salesforce.com/news/stories/sales-research-2023/ - only 26% of sales professionals get one-to-one coaching at least weekly
  • https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0120644 - replication of the forgetting curve, reason to retest
  • https://journals.sagepub.com/doi/10.3102/0034654320933544 - simulation-based learning meta-analysis, g = 0.85
  • https://www.bridgegroupinc.com/research/2026-ae-models-motions-metrics - account executive ramp of 6.2 months and 48% at quota
  • https://hbr.org/2006/03/customer-value-propositions-in-business-markets - resonating focus beats listing every benefit

All legal information verified as of 16 November 2026. Consult with a labour lawyer for specific cases.

Keep reading