Procurement at an association, chamber or trade publisher usually needs a reference: proof that the software has already run a real prize competition. Vendors answer with a case study — two pages, a customer logo, three figures, one quote. It was written by the vendor, approved by the customer and shortened by marketing. That does not make it worthless; it makes it a source you assess like any other.
What an award software case study leaves out
Three weaknesses are built into almost every study.
Selection. Studies appear when the customer is happy and consents. Organisers who switched tools after one season never feature, so a study shows the favourable case, not the typical one.
No baseline. “Entries up 40 per cent” says nothing without last year's figure. Twenty to twenty-eight is a different story from 500 to 700 — and neither can be read if the same year doubled the categories or dropped the fee.
Attribution. An award changes every year: new categories, a new media partner, a longer entry window. The software is one change among several, yet the study credits it with the whole effect.
Which figures survive scrutiny
Some numbers are hard to dress up because they sit on the operation itself; others sound precise and are not. The table sorts the usual claims by what they prove and the question that decides it.
| Claim in the study | Evidential value | The question that decides it |
|---|---|---|
| “Entries up X%” | Only with the prior-year figure | Absolute numbers for both years, category count, fee, entry window? |
| “X hours saved” | Only with the process scope | Which steps were counted — recorded or estimated, and by whom? |
| Drop-off rate in the entry form | Good, if defined | When does an entry count as started: page view or saved draft? |
| Share of jurors finished on time | High, hard to massage | How many jurors, how many rounds, how long was the window? |
| “Satisfaction 4.8 out of 5” | None while n is missing | How many responses, collected when, sent out by whom? |
| Support response time | Good, if stated separately | First reply or resolution — and does it hold in deadline week? |
| Cost per entry | Only on a complete cost base | Are setup, add-on modules and internal staff hours included? |
| Seasons with the same vendor | The most reliable of all | Since when, on which plan, without a gap? |
The last row matters most: whether an organiser renews after the first season is the one figure a vendor can barely reinterpret. Do not ask how many customers there are — ask how many are in their third season.
Nine questions for a reference call
A reference call beats any study because you can push back. It is still a prepared conversation — the vendor picks who you speak to. Ask about friction, not satisfaction: people remember friction precisely.
- What did you use before, and how many entries did you handle with it last year?
- How many weeks passed between signing and the entry window actually opening?
- What did you have to change in your own process to fit the software?
- What took longer than you were told it would?
- Who did the configuration — your team or the vendor — and which parts were billable?
- What went wrong in the final week before the deadline, and how quickly did you get an answer?
- Which feature did you expect and not receive?
- How many of your jurors needed help to submit their scores?
- Would you sign again on the same plan — and if not, on which one?
Seven signs a case study has been tidied up
- Percentages with no absolute starting figures.
- No year or time frame, so the study may describe a product version that no longer exists.
- Quotes without a name and role, or written in the vendor's own vocabulary.
- Not a single point of friction. Every real award season has at least one.
- Figures that do not fit the size of the award: 200 hours saved on 30 entries.
- The same number appearing in studies about different customers.
- No jury size, category count or number of scoring rounds, although those determine the workload.
The check takes ten minutes: recalculate the figures against the stated parameters. If the parameters are missing, the check is already over.
How the HR Award process changed — and why no figures appear here
Laureo grew out of running the HR Award at the HR Inside Summit, so we apply the rule to ourselves: the 2026 season's result figures have not been released by the organiser, and we will not publish them before that happens. What can be described is the process before and after — qualitatively, without any claim about effect.
Before, entries arrived as email attachments and were copied into spreadsheets. Incomplete documents were chased one by one, scoring packs assembled and sent to each juror, the returns merged by hand. Who had scored what sat across several files.
After, the same sequence runs in one system. The form enforces completeness, an eligibility check runs before submission, entrants work in a draft mode. Jurors see only the entries assigned to them, conflicts of interest are handled at assignment, score normalisation evens out strict and lenient jurors. Results are released in stages, feedback PDFs are anonymised, certificates generated automatically. That describes a workflow; it is not a success report.
What can replace a case study
Until reliable figures exist, a model calculation with its assumptions on the table is more honest. Ours assumes 100 entries, 10 jurors and an internal rate of € 50 per hour — conservative for association staff in German-speaking Europe: roughly 200 hours (≈ € 10,000) on email and spreadsheets, roughly 120 hours (≈ € 6,000) with an online form, € 25,000+ one-off for custom development, roughly 10 hours plus € 1,790–3,490 in platform fees with Laureo. Every assumption is named and replaceable; the breakdown sits under Resources, the derivation in what an award really costs.
The argument against us belongs here too: if your rules require twenty named references and three years of figures, a young product cannot supply them — not us, and not most European vendors. A trial instance then beats any PDF: create your categories, invite two colleagues as jurors, run one scoring round end to end. Which criteria matter is set out in the award software comparison, and setting up a round cleanly in the article on the jury process.
Model calculation with disclosed assumptions, July 2026. No result figures for the HR Award 2026, as the organiser has not released them. Laureo is a product of State of Innovation GmbH, Vienna.
Frequently asked questions
How do you spot an exaggerated software case study?
Look for three gaps: percentages without absolute starting figures, no year or time frame, and not a single point of friction. A credible study states the prior-year number, the jury size and the category count. Without those parameters nothing can be recalculated, so the document is copy rather than evidence.
What questions should you ask a software vendor's reference customer?
Ask about friction rather than satisfaction. How many weeks passed between signing and the entry window opening? What took longer than promised? What went wrong in the final week before the deadline, and how fast was the reply? Which feature was missing? Would they sign the same plan again?
Is hours saved a reliable metric in a software case study?
Only if the process scope is stated. Ask which steps were counted, whether the hours were recorded or estimated, and who did the measuring. Without the number of entries, the jury size and the number of scoring rounds, an hours figure cannot be transferred to your own award.
What if an award software vendor has no references yet?
Then test the product instead of the story. Set up a trial instance with your own categories, invite two colleagues as jurors and run one scoring round end to end. A model calculation whose assumptions you can replace with your own figures is also more verifiable than a PDF.
Related reading
Planning an award right now?
In 20 minutes we walk through your process: entry, jury round, invoice.
Book a demo