Training is one of the few corporate investments whose return is rarely measured with any rigour. According to research by the Association for Talent Development (ATD), only a minority of organisations calculate the ROI of their training programmes. The rest settle for participant satisfaction (the familiar "happy sheet") without knowing whether the training changed anything at all.
With budgets under pressure and internal resources fiercely contested, demonstrating the business impact of your customer relations training is no longer a luxury. It is a survival condition for the programme itself.
The reference model for evaluating training impact is Kirkpatrick's, which distinguishes four levels.
Did participants enjoy the training? This is the easiest level to measure (a satisfaction questionnaire) and the least predictive of real impact. A 95% satisfaction score is no guarantee that practices will change.
Did participants acquire the skills targeted? This can be measured through before-and-after tests, assessed simulations or certification. AI simulation platforms measure this level automatically: progress on each skill, session by session.
Are participants applying those skills in their daily work? This is the hardest level to measure and the most important. Methods include mystery calls, managerial observation and analysis of customer interactions (CRM records, call recordings). This is where the teaching method makes the difference: practice-based training produces significantly better behavioural transfer than theory-led training.
What is the impact on business indicators? Revenue, customer satisfaction, retention, fewer complaints, regulatory compliance. This is ROI in the strict sense.
The basic formula is simple.
ROI (%) = [(Benefits attributable to the training − Total programme cost) / Total programme cost] × 100
The difficulty lies in estimating attributable benefits. Here is how to calculate them by type of impact.
Measure the change in conversion rate, average basket or closing rate before and after the training, comparing against an untrained control group. In our banking case study, the increase in closing rate was converted into additional revenue in exactly this way.
Customer satisfaction converts into economic value through two mechanisms: retention (a satisfied customer stays longer) and recommendation (NPS). In public transport, the same logic applies to repeat journeys.
Lower staff turnover (replacement cost avoided), fewer complaints (handling cost avoided), fewer incidents of abusive behaviour (absence cost avoided). In our retail case study, lower turnover generated measurable savings on recruitment and training.
In regulated sectors (banking, insurance, healthcare), the cost of non-compliance is measurable: penalties, remediation costs and reputational damage. Improving compliance in the banking case study reduced a regulatory exposure of considerable value.
Direct costs: the training provision itself, software licences, technical integration. Use our pricing guide to estimate these.
Indirect costs: training time (the salary cost of hours spent in training), project management time (HR, line managers) and opportunity cost (productivity lost while people are training).
For a digital programme (AI simulation, microlearning), indirect costs fall sharply: fifteen five-minute sessions amount to 75 minutes per learner, against 14 to 16 hours for a two-day classroom programme.
Pick three to five measurable indicators aligned with your business objectives. The most common ones by sector are as follows.
Retail: average basket, conversion rate, customer satisfaction index, loyalty rate, product return rate.
Banking and insurance: closing rate, average assets per client, compliance rate, complaint rate, attrition rate.
Transport and hospitality: NPS, reported incident rate, complaint rate, online reviews.
Contact centres: first contact resolution (FCR), average handling time, CSAT, call-back rate.
Before launching the programme, measure current performance on your key indicators. That baseline is the reference point for everything that follows. Without it, no ROI calculation is possible.
This is the trickiest step. An indicator can improve for many reasons: a new marketing campaign, a change in product range, seasonality. Three methods help isolate the training effect.
The control group. Compare the performance of a trained group with an untrained one over the same period and scope. This is the most robust method, and the one used in our case studies (transport, banking, retail).
Managerial estimation. Ask managers to estimate the share of the improvement attributable to the training (the Kirkpatrick/Phillips approach). Apply a discount factor for prudence, typically keeping 60-70% of the estimate.
Trend analysis. Compare the indicator's trajectory before and after the training. If it was flat and improves sharply after rollout, attribution is plausible.
Every improvement must be converted into currency. For example: +1 point of conversion rate × number of transactions × average basket = X. −1% turnover × headcount × replacement cost = Y. −5% complaints × average cost of handling a complaint = Z.
Apply the formula: ROI = [(Total benefits − Total costs) / Total costs] × 100. An ROI of 200% means every euro invested generated two euros of net benefit.
To put your own results in context, here is how the main formats generally compare.
Classroom training alone: limited behavioural transfer (only part of what is learned gets applied) holds the return down.
Standard e-learning: low completion rates weigh heavily on the outcome.
Simulation and conversational learning: higher completion and stronger behavioural transfer support a better return.
Blended (classroom plus simulation): combining the two approaches tends to deliver the best yield, as set out in our guide to blended learning.
Measuring satisfaction only. A 95% satisfaction score is gratifying but says nothing about business impact. Go at least as far as Kirkpatrick level 3 (behaviour).
Ignoring indirect costs. Counting only the provider's invoice understates the real investment considerably.
Measuring too early. Behaviour change takes time. Measure behavioural ROI at three months and business ROI at six to twelve months.
Attributing 100% of the improvement to training. Always apply a discount factor or use a control group to isolate the real effect.
A solid business case combines documented results from a comparable context, a realistic estimate of benefits (with prudence factors applied), a full cost calculation covering direct and indirect items, and a measurement plan with milestones at one, three and six months.
Where training funding schemes are involved, a documented ROI considerably strengthens the application. It is also a decisive argument for executive teams arbitrating between competing investment projects.
Need help building your business case? Get in touch with Face Up for a tailored estimate of your programme's ROI.
