By Douglas G. Wood
Over the past 20 years, I’ve sat across the table from detectives, sergeants and commanders at dozens of agencies, working through the same set of confidential informant problems narcotics units and task forces run into everywhere. Different departments, different regions, different budgets, but the pattern is almost always the same: The agencies that struggle aren’t lacking good informants. They’re lacking a program disciplined enough to survive the people who built it.
Here’s what actually separates the best units from the rest, and where I’d tell a new commander to start.
Trust is built one contact at a time
Hollywood loves a dramatic informant story. Real investigations are usually a lot less dramatic than that. Most productive informants don’t become valuable overnight. The relationship develops over months, sometimes years. One good address leads to another. One accurate phone number supports a search warrant. One controlled buy establishes credibility, and investigators start trusting the source because experience keeps backing it up, not because of a gut feeling.
That confidence comes from evidence, not personality. Every meeting adds a data point. Every payment becomes part of the history. Every successful operation strengthens the documented record, and every bad tip becomes part of that record, too. Years later, nobody should have to guess whether a source was reliable. The file should already answer that.
What the best agencies do differently
A few habits show up consistently in the strongest CI programs I’ve seen, and none of them are exotic.
They treat the informant as the agency’s asset, not the handler’s. The file, the history and the relationship are documented well enough that a different detective could step in tomorrow without starting from zero.
They build the renewal date into the record itself instead of relying on a supervisor’s memory. Suitability reviews, background check renewals and agreement expirations get flagged 30 or 60 days out, automatically, so nothing slips because a source has gone quiet for a few months.
They log every contact, not just the ones tied to a payment. A phone call, a debrief, a safety concern mentioned in passing — all of it goes into a running, time-stamped record, because the value of that log isn’t just protecting today’s case. It’s answering a reliability question with evidence six months from now instead of a gut feeling.
They separate day-to-day disbursements from the approval that authorizes them. Individual handler wallets are tied to receipts and reconciled against a central fund that requires sign-off, with a second set of eyes on anything larger. That structure catches a problem while it’s still small enough to fix quietly.
And their supervisors actually open the file. Not once a year before an inspection. On a real schedule.
What a supervisor should be reviewing, and how often
This is where a lot of programs quietly fail, not because nobody cares, but because “review the CI files” never gets a schedule attached to it. Here’s the schedule I’d tell a sergeant or lieutenant to follow:
- Weekly: New informant registrations and any payments made in the prior week, reconciled against receipts.
- Monthly: Contact frequency for active sources, so a file that’s gone quiet gets noticed before it becomes a surprise.
- Quarterly: Reliability trends across the full roster, comparing what’s been corroborated against what hasn’t.
- Annually, at minimum: Full suitability reassessment and background check renewal for every active informant, with results attached to the record, not summarized from memory.
If none of that is happening on a real calendar right now, that’s the first thing to fix, before any conversation about new software or new policy language.
A quick health check for chiefs
If I were asked to gauge the health of a CI program in a single afternoon, I’d ask for direct answers to these questions, not a policy binder:
- Pull three CI files at random. Can you tell me, without calling anyone, when each was last reviewed, what each informant has been paid this year and who’s currently handling them?
- Ask to see the last completed audit of confidential funds. If the honest answer is “we’ve never really been audited,” that’s the finding.
- Ask a detective, not a supervisor, to walk you through how they’d hand off an informant to another investigator tomorrow. If the answer involves a notebook only they can read, you have your answer.
- Check whether any active informant has gone more than six months without a documented contact or review. There’s usually at least one, and it’s rarely the one everyone’s been watching closely.
The weaknesses I see most often
Visiting agencies of every size over the years, the same handful of gaps show up again and again, almost regardless of budget or agency size. Documentation that exists but lives in three different places: a notebook, a spreadsheet and whatever a retired detective remembers. Nobody consolidated it, so nobody can actually answer a question about the program quickly.
Payment records that would embarrass the agency in an audit, not because anyone did anything wrong, but because the paper trail was never designed to answer six basic questions on demand: Who authorized it, who received it, why, how much, which case and where’s the documentation?
Supervisors who review the CI program the same way they review anything else that isn’t urgent: eventually, and usually only when something forces the issue.
Files that never get formally closed. An informant goes inactive, and instead of a deactivation with a reason attached, the file just stops being touched, which makes it impossible to tell an abandoned source from an active one six months later.
Individual handlers can create risks of their own, usually from good intentions rather than bad judgment. Treating a source as “mine” rather than the unit’s feels protective in the moment but becomes a liability the day that detective transfers, retires or is simply unavailable.
Writing thin notes because a meeting felt routine is another common mistake. The meetings that feel routine are exactly the ones that get forgotten first, and exactly the ones a prosecutor asks about 18 months later.
Then there’s letting reliability become a reputation instead of a documented track record. “He’s always been solid” is not the same as a file that shows which specific tips were corroborated and which weren’t. And skipping the paperwork on a payment because the amount was small? Small, undocumented payments are how audits start.
What the file actually needs to hold
A CI file with just a name and a phone number isn’t a file. It’s a business card. A file that protects the informant, the investigation and the agency needs the background check and vetting results on hand, along with the date they need to be redone, since a vetting result from four years ago says almost nothing about who this person is today. It needs the signed agreement itself, any waivers, any restrictions a supervisor placed on the source, photographs, known vehicles and aliases, emergency contacts, and a full history of every handler this informant has ever had, not just whoever is working with them now.
Left scattered across a desk drawer, that’s a liability waiting to surface in discovery or an audit. Kept in one place, it’s exactly what a supervisor needs at 10 at night when a detective calls asking whether it’s safe to send this source into a specific situation.
Where I’d tell a new commander to start
If someone taking over an investigations unit asked me where to begin building or fixing a CI program, I wouldn’t start with technology. I’d start here, in order:
- Pull every active CI file that currently exists, wherever it lives, and consolidate it into one place with one format. You cannot fix what you can’t see in front of you.
- Put a renewal date on every single one, and build a habit, weekly if you have to, of checking what’s coming due.
- Fix the money before anything else. If you can’t answer who authorized a payment and why within 60 seconds, that’s priority one, not a someday project.
- Put a real schedule on supervisory review, the one outlined above, and stick to it even when the unit is busy. Especially when the unit is busy, since that’s exactly when reviews get skipped.
- Only after those four are solid, look at whether your current system, paper or digital, is helping you sustain that discipline or working against it. Technology should reinforce a good process. It can’t create one that doesn’t already exist.
Compliance is earned, not installed
No piece of software makes an agency compliant. People, policy and supervision do that. Agencies working with criminal intelligence should understand how their systems align with 28 CFR Part 23, where applicable, including requirements for reasonable suspicion, controlled dissemination and periodic review and purging of outdated intelligence.
The FBI’s CJIS Security Policy also establishes requirements for protecting criminal justice information, including applicable safeguards for access, authentication, encryption and auditing. Agencies should evaluate which requirements apply to their systems and operations. These safeguards exist to protect the investigation, the agency and the public.
Software doesn’t review files. Sergeants do. Technology just gives them better information to look at when they sit down to do it.
Three questions worth asking
Before the next audit, promotion, retirement or major case, it’s worth asking three questions honestly. If your lead investigator retired tomorrow, could another detective inherit every active informant without losing anything important? Could your agency account for every confidential fund payment quickly and with confidence? Would you actually welcome an unannounced audit next Monday morning?
If the answer to any of those is no, the fix isn’t better technology. It’s a better process.
Final thought
The agencies with the strongest confidential informant programs usually aren’t the largest or the best funded. They’re the ones with a supervisor who actually reviews the file, a system where every payment can be accounted for on demand, and a record that survives the people who built it. That’s not a technology problem to solve. It’s a discipline to maintain, and it’s one any agency can start building tomorrow, regardless of budget.
Further reading: https://caseclosedsoftware.com/all-posts-page/
About the author
Douglas G. Wood is president and CEO of Crime Tech Solutions, LLC, developer of Case Closed Software, an investigative case management platform used by law enforcement agencies across the United States. He has worked with dozens of agencies and met with hundreds of detectives over the past 20 years, focusing on confidential informant management, investigative workflows and CJIS-oriented information systems.
