Organisation
Foundation, 24 researchers and 8 support staff. Funding: 55% commissioned research, 45% grants and programmes.
Example engagement · Research institute
Literature, applications and reports with a pinpoint reference
An independent research institute of 32 people doing policy research for ministries, municipalities and funds.
employees32
turnover€3.8m
reports a year~45
Profile
The organisation
Work
Policy evaluations, literature reviews, monitoring studies. Every finding has to be traceable; reports are cited in Parliament and in council chambers.
Systems
Zotero per researcher, 1,900 of the institute's own reports on a network drive, Word, Excel, grant portals of 7 funders.
The question
Where the hours went
Every literature review started from zero: 3 to 5 weeks of searching and summarising before the analysis could begin. Reference lists differed per researcher and earlier work by the institute was missed because nobody could find it.
Grant applications took 60 hours each on average, in seven different formats, with a 30% success rate. Researchers wrote the same institute description and method paragraphs again and again.
Working session and scan · Worked example
Four weeks of watching the work
Working session with six researchers and the grants coordinator, then four weeks alongside two running studies and three applications.
3–5 wks
of searching and summarising before a literature review got going on substance.
1,900
of the institute's own reports on the drive, not searchable by content. In 2 of the 2 studies followed, relevant in-house work was missed.
60 hours
per grant application, over 20 hours of it on text that already appeared in earlier applications.
38%
of the source references in drafts had to be fixed or completed at final check.
7
funders, each with its own format and its own reporting requirements.
The pattern: the thinking was good, the supporting work around it ate the time. And the source that existed was not where it was needed.
What we built
Four pipelines on one source base
Everything points to a location: page, paragraph, line. A claim without a source doesn't make it into the draft.
Literature monitor
A weekly overview of new publications in the institute's own fields, with a summary and relevance to running studies. No pinpoint reference, not included.
Human approvesThe researcher decides what stays in.
Own work, searchable
1,900 reports and all Zotero libraries brought together, searchable by content. Finds the passage and the report.
Human approvesThe researcher judges whether the source still holds.
Reference list and first draft
An ordered reference list and a first outline per chapter, with the pinpoint reference on every line. Contradictions between sources are flagged, not smoothed over.
Human approvesThe researcher writes the analysis and conclusion themselves.
Applications and reporting
A draft in each funder's format, filled from approved institute texts and earlier work. Reporting drawn from the project administration.
Human approvesThe director signs.
The three rules
Every result has its source.
If the source is missing, it says: I don't know.
What goes out, a person approves.
Engagement · Worked example
Four months
Agreed before we started. One workflow live each month, with the results measured on the organisation's own work.
Month 1
Source base
1,900 reports and 14 Zotero libraries made searchable. Test set of 40 questions: 87% right source in the top three.
Month 2
Literature monitor and report pipeline
First literature review with the pipeline: 8 working days to analysis instead of 4 weeks. Source check: 6 corrections on 340 references.
Month 3
Application pipeline
Four applications through the pipeline, 22 hours each on average. Two awarded, two pending.
Month 4
Policy and handover
Working policy: what the system may do, what the researcher does. Training for all 24 researchers. Maintenance assigned.
Results · Worked example
After four months
8 days
to start of analysis
was 3–5 weeks
22 u
per grant application
was 60 hours
2%
references fixed at final check
was 38%
0
missed in-house work in the studies followed
was 2 of 2
1 week
new literature known
was by chance
100%
claims with a pinpoint reference
was not measurable
What it's worth · Worked example
€230,000 to €300,000 a year
Internal rate €80 to €105, 75% realisation in the first year. Extra awards not counted as revenue, only the hours.
Freed-up research hours in the literature phase
€120–160k~14 days × 18 studies × €80–105, 75% realisation
Freed-up hours on applications
€60–75k38 hours × 26 applications × €80–105, 75% realisation
Less rework on source references
€30–40kMeasured checking hours last year
Reporting to funders
€20–25kMeasured hours per report × number
Total, calculated conservatively
€230–300kAfterwards
What happened next
The institute now publishes a source appendix with every report, giving the pinpoint reference for each finding. Two funders named that as a reason for a follow-up commission. That is not in the value above.
Zo begin je.
Want to look for yourself first? The supplier scan and the Chain check are free and need no conversation.
The supplier scanA half-hour intro
About where your hours go. No obligation. We won't chase you afterwards.
Book an intro callThis is already running
BBEbuddy. Teachers ask a question and the answer comes from a knowledge base of their own, with the source attached. Runs in three languages.Valora. A valuation that used to take hours now takes minutes, with every calculation step visible.Bookaflow. A booking system that stops the moment a person has to decide.
Read the case