How to Measure Graduate Program Success: Checkpoints for 12-Month to 2-Year Programs

A published program length already tells you when to check in. According to Bega Group's RISE graduate page, its program is a 2-year rotational program made up of three 8-month rotations. Fonterra describes an 18-month program on its graduate page, and Asahi Beverages lists 12 to 15 month options on its page for graduates. Those published lengths give you natural points to measure graduate program success, so you do not have to invent a reporting calendar from scratch.

You do not need a data team or a big cohort to do this. A few dated checkpoints and a short list of measures, written down before the first graduate starts, will show whether selection and the program itself are working. Our earlier guide to campus recruitment costs covers cost per hire, and our guide to predictive graduate hiring covers ranking candidates with data, so this one starts after the offer is accepted.

Key Takeaways

• Measure at the points the program already has: the end of each rotation, the mid-point and the program end, plus one reading six months after.
• Six simple measures are enough for a small cohort: offer acceptance, completion, retention, manager rating, placement feedback and time to productivity.
• Keep every selection score on the graduate's record so you can compare it with manager ratings at 90 days and 12 months.
• When a measure looks weak, adjust the selection criteria first rather than only rewriting the advertisement.

Checkpoints by program length

The lengths in the table come from each employer's own page. Fonterra's page gives 18 months, though a graduate quote on it mentions a 2-year period, and it names rotation supervisors without giving rotation lengths, so that row uses 18 months and a mid-point. The checkpoints and measures are our suggestion, not something any of these employers publish. Months count from the first day, and "six months after" means six months after the program ends. Asahi's page shows applications closed, so use it as a length reference only, and check each page for the current intake before you rely on it.

• Bega Group: 2-year program, three 8-month rotations - Suggested checkpoints: Day 90; month 8 (rotation end); month 12; month 16 (rotation end); month 24 (program end); month 30; Measure taken at each: Day 90: manager rating and time to productivity. Month 8: manager rating and placement feedback. Month 12: manager rating, to compare with selection scores. Month 16: manager rating and placement feedback. Month 24: completion and retention at program end. Month 30: retention six months after.
• Fonterra: 18-month program - Suggested checkpoints: Day 90; month 9 (mid-point); month 12; month 18 (program end); month 24; Measure taken at each: Day 90: manager rating and time to productivity. Month 9: manager rating and placement feedback. Month 12: manager rating, to compare with selection scores. Month 18: completion and retention at program end. Month 24: retention six months after.
• Asahi Beverages: 12 to 15 month program options - Suggested checkpoints: Day 90; month 6 (mid-point); month 12 to 15 (program end); six months after the end; Measure taken at each: Day 90: manager rating and time to productivity. Month 6: manager rating and placement feedback. Program end: manager rating, completion and retention at program end. Six months after: retention again.

Offer acceptance is measured before day one, so it sits outside the table.

Two habits apply to every row. Take a reading at each rotation end or, where a program states no rotations, at the mid-point, because feedback is easy to miss when a graduate changes manager or placement, and a mid-point gives a check where there is no such change. And always take a reading six months after the program ends. Six months later the program's support has gone and the graduate's role has to stand on its own.

Six measures that fit a small cohort

Our guide to quality of hire metrics covers hiring in general, and the six measures below are sized for a small graduate cohort. With a handful of graduates, simple counts beat clever formulas. Define each measure in writing before the first graduate starts, and keep the definition the same for every intake.

• Offer acceptance. Graduates who accepted divided by offers made. It tells you whether the offer and the selection experience held up against the alternatives your candidates had.
• Completion. Graduates who finished the program divided by graduates who started it. Write down the reason for every exit.
• Retention at program end and six months after. Graduates still employed by you on that date divided by graduates who started. This is the graduate retention rate in its simplest form. Decide in advance whether a move to another team inside your organisation counts as retained. We suggest it does.
• Manager rating. One rating from the graduate's manager at each checkpoint, against the same few criteria and the same scale every time.
• Placement feedback. A short written check-in with the graduate and the supervisor at the end of each placement: what worked, what was missing, what they would change.
• Time to productivity. The number of weeks until the manager agrees the graduate is handling the core tasks of the role without close supervision. Agree what "core tasks" means for each role beforehand.

These six double as graduate program KPIs if your leadership wants a simple dashboard. Resist adding more. With fewer than a dozen graduates, a single departure moves a rate by a large amount, so read every number next to the reasons people gave.

Linking selection scores to graduate performance

Keep the scores from selection

To link selection scores to graduate performance, you first have to keep the scores. A program that runs a cognitive test, a video interview and an assessment centre can end up filing only the outcome. Asahi's page, for example, lists online cognitive assessments, a one-way video interview and an assessment centre. Whatever stages you run, save the score or rating from each one against the graduate's record, in a sheet the program owner can still open two years later.

If skills tests are one of your stages, a tool such as Righteo's AI skills assessment is one practical option. According to its product page, it tests technical, personality and cognitive skills before an offer and produces shareable reports that you can file with each graduate's record.

What a small cohort can and cannot show

Compare the scores with manager ratings at 90 days and again at 12 months. Put the graduates in order by selection score, put them in order by manager rating, and look at where the two orders disagree. This early performance view of a graduate cohort will not give you a correlation you can trust. It will give you questions worth asking.

It cannot show cause. You only see ratings for people you hired, so you cannot know how the people you turned down would have done. A manager's rating also reflects the manager and the placement, not only the graduate.

Here is a hypothetical cohort of six. Graduate C had the lowest case study score and the highest manager rating at 12 months, while graduates A and B scored highest at selection and were rated in the middle. One graduate like C is a prompt to ask whether the case study rewards polished presenting more than good work. It is not proof that the case study should go.

What to do with the answer

When a measure looks weak, the reflex is to advertise harder or add another interview round. That treats a selection problem as a volume problem. A better first question is which criterion would have changed the decision. This is the practical side of graduate program evaluation: adjust the criteria, not just the ad.

Take a second hypothetical: a cohort of eight in which two graduates leave before the end of the second rotation. In their exit conversations both say the role involved more shift-based planning than the advertisement described. Retention at program end is six of eight, but the number alone says little. The reasons point to a requirement that selection never tested.

Describing the shift work plainly in the advertisement and adding a short work-sample step for that part of the role would change the next intake. Reposting the same advertisement would not.

If the gap is in how you test the work itself, Righteo has a skill tests library where you describe the role and it surfaces the most relevant tests to add to a selection stage, and its AI skills assessment page says you can also build your own. Whatever you change, write it down, date it and note which intake it applies to, so later results are read in context.

Measurement also feeds the calendar for the next intake. Our guide to pipeline planning shows how headcount plans and attrition signals can open a search early, which gives you time to apply what the last cohort taught you.

Related Resources

• AI skills assessment
• Skill tests library
• Campus recruitment costs
• Predictive graduate hiring
• Pipeline planning for predictive hiring
• Bega Group graduate program page
• Fonterra graduate program page

Common Questions About Measuring Graduate Program Success

1. How do you calculate a graduate retention rate?

Divide the number of graduates still employed by you at a set date by the number who started the program. Choose the dates in advance, such as program end and six months after. Also decide whether an internal move to another team counts as retained, and keep that rule the same for every intake so the results can be compared.

2. How many graduates do I need before the numbers mean anything?

There is no safe cut-off to quote here. With a small cohort, treat the figures as prompts for questions rather than proof. One departure can move a rate sharply, so read each number alongside the reasons given and the manager and placement involved.

3. When should the first manager rating be taken?

Take the first rating at 90 days, then at 12 months and at the end of each rotation or, if the program has no stated rotations, at the mid-point. Use the same criteria and the same scale each time, and rate against the role's core tasks.

4. Can selection scores predict who will do well?

They can show a pattern worth a conversation, but a small cohort cannot prove a cause. You only see results for the people you hired, and ratings also reflect the manager and the placement. Use the comparison to question your criteria, then change one thing at a time and watch the next intake.

5. What if my program is a different length from the examples?

Use your own published length. Take a reading at day 90, at each rotation end or the mid-point, at program end, and six months after. Employers can change their programs, so check the current page for any program you use as a reference before you quote its length or structure.

Choose three measures and book the dates

You do not need all six measures on day one. Pick three, such as completion, retention at program end and the manager rating at each checkpoint. Put every checkpoint date for the next cohort in your calendar before they start. Save each graduate's selection scores with their record, so that in a year you have something to compare.