I've been evaluating Mandiant Threat Intel for a potential enterprise-wide license. The data itself seems solid, but the search functionality in their portal is becoming a deal-breaker for my team.
We're running into consistent issues where key threat actor names, malware families, or even specific CVEs referenced *in the body of their own reports* do not return results when searched. For example, searching for a campaign alias they coined might yield zero hits, but you can find a detailed report on it by browsing manually. This isn't a one-off; it's a pattern.
* It forces analysts to waste time browsing or using external search to find Mandiant's own content.
* It undermines confidence in the platform during a critical incident. If we can't trust search to be exhaustive, we can't trust the workflow.
* It significantly impacts the ROI calculation. We're paying for intelligence we can't efficiently retrieve.
I need to know if this is a common experience or if we're configuring something incorrectly. Has anyone else run into this? More importantly, has anyone found a workaround or gotten a clear response from their support on a timeline for a fix?
From a procurement standpoint, a flaw in core platform functionality like this is a major red flag. It directly impacts total cost of ownership by adding labor hours for basic retrieval. I'm currently advising my security team to pause the evaluation until this is resolved.
This is a classic indexing problem, and it's more common than you might think with these specialized platforms. The data is there, but the search engine isn't parsing or weighting the report text bodies correctly.
I've seen this happen when a vendor's search is tuned only for titles, abstracts, or tagged metadata, completely ignoring the full text. Your point about it impacting ROI is spot on - you're essentially paying for a library with a broken card catalog. My advice is to escalate this with their support as a critical procurement blocker. Ask them specifically if their search indexes the full report body and, if so, what their re-indexing schedule is. Sometimes a "known issue" gets prioritized fast when a seven-figure deal is on the line.
If they can't give you a clear fix timeline, that's your answer on platform maturity.
Integrate or die
You're absolutely right about the indexing being the likely culprit, but the "re-indexing schedule" question often gets a canned, non-technical answer. The real problem is usually stemming or tokenization rules that are too aggressive for technical jargon.
In my experience with similar platforms, they'll stem "Conti" to "cont" or treat "CVE-2024-1234" as a single token, but then the search bar splits it on the hyphen. That makes it impossible to match unless the search logic is aware of the corpus's specific format. Asking for their tokenization strategy for malware names and CVEs gets you past the support script.
If they're using a generic cloud search service out of the box, a fix isn't a schedule, it's a redesign.
Show me the benchmarks
That ROI point is critical. We saw the same search problem in our demo and it tanked the perceived value. Our team estimated a 20% time waste just on manual lookups, which killed the business case.
Did support give you any concrete numbers on how their search is supposed to work? Like the percentage of report body text that's indexed? We asked and got a vague "it's being improved" which was a red flag for contract negotiations.
If it's broken during eval, assume it stays broken. We walked away.
You've hit on the exact kind of operational friction that derails adoption, especially during a high-pressure incident. The broken trust is often the hardest part to rebuild.
From a vendor evaluation standpoint, I'd treat this as a blocking issue until they can demonstrate a fix. In your procurement conversations, shift from asking for a timeline to asking for a *demonstration*. Request they run your exact failing searches - the campaign alias, the specific CVEs - in a live session with you. Their ability (or inability) to reproduce the problem and explain the search logic in real-time tells you everything about priority and technical debt.
If they can't make it work during a sales cycle where they're incentivized to, it's very unlikely to improve post-contract. That's a hard line, but it's saved us from similar pitfalls.