Guide · · 3 min read
Intern evaluations that actually inform the offer decision
Most intern evaluations document performance but can't support a hiring decision. How to design evaluations backward from the return-offer decision.
Jacob Moeckler
Co-founder & CEO
Most intern evaluations are designed to document performance, then get used to make a hiring decision they were never built for. They arrive in the last week, use a vague five-point scale, and somehow every intern "exceeds expectations." Then the offer conversation happens anyway, mostly on the manager's gut.
The fix is to design the evaluation backward from the decision it has to support: what would a manager need to say, clearly, for you to extend an offer or not?
The everyone-is-a-four problem
Rating scales on general traits like "professionalism" or "initiative" cluster at the top, because managers don't want to hurt anyone and the criteria mean something different to each of them. The result looks like data but can't separate a strong hire from a nice summer.
Start from the decision
- Define success criteria per role at the start. Three to five things that matter for that team, shared with the intern on day one. Most employers already work this way when hiring: NACE reports 70% use skills-based hiring.
- Ask for evidence, not adjectives. One concrete example per criterion.
- Ask for the recommendation directly. "Would you hire this person onto your team full-time?" Yes, yes with reservations, or no, with a reason.
- Ask about headcount separately. "Should we hire them" and "can we hire them" are different questions, and mixing them hides your best people.
| Element | Typical version | Decision-ready version |
|---|---|---|
| Ratings | 1 to 5 on general traits | Role-specific criteria with an example each |
| Recommendation | Implied by the average score | Explicit: yes, yes with reservations, no |
| Headcount | Not asked, or blended in | Asked as its own question |
| Timing | Final week only | Midpoint and final |
The midpoint matters more than the final
A midpoint evaluation surfaces problems while there's still time to fix them, and it gives the intern real feedback. NACE lists regular feedback from managers among the strategies employers use to raise conversion. When the final evaluation surprises the intern, the process failed, whatever the performance.
Getting managers to finish them
Keep it under ten minutes. Tie the deadline to the offer date so the reason is obvious. Track completion by name, not by percentage. Agree on an escalation path, usually the manager's manager, before the summer starts rather than in August.
From Jacob: A student told me he used his online evaluation to flag that he didn't have a real project. Nobody responded. He ended up setting up a check-in himself to get it fixed. The form did its job of collecting the information; the process around it didn't do anything with it. That's the difference between an evaluation built to document and one built to decide.
FAQ
Should mentors evaluate too?
Their input is useful, especially on collaboration, but keep the recommendation with the manager who owns the decision.
Should interns see their evaluation?
Yes for the midpoint. Decide in advance how much of the final is shared, and be consistent.
How many criteria is too many?
More than five and managers start skimming.
How teams run this in Springbird
Springbird connects evaluations directly to the decision they're meant to support.
- Midpoint and final evaluations sent automatically on each participant's dates, so nobody has to remember to send them.
- Role-specific success criteria and an explicit recommendation, captured as answers you can compare across interns and seasons.
- Completion tracked by manager name, with reminders that go out before the offer deadline.
- Recommendations beside the conversion decision, with each evaluation on the participant's page when the offer call is made.