I am currently in the evaluation phase for a new cloud procurement and vendor management platform, with a specific focus on tools that can augment or automate the research phase for significant reserved instance and savings plan commitments. The platform "Claw" has come across my desk several times, marketed heavily on its AI-driven ability to scrape and analyze cloud provider pricing, spot market trends, and even third-party reseller listings to identify the most cost-optimal procurement path.
Before proceeding to a formal trial or engaging their sales team, I am seeking unbiased, operational feedback from practitioners who have hands-on experience with the platform. The marketing claims are substantial, but I am inherently skeptical of black-box systems in the FinOps space, where precise cost attribution and clear sourcing logic are non-negotiable.
My primary questions for the community are:
* **Deal Discovery Accuracy:** Did Claw's research consistently surface materially better pricing than a thorough manual analysis or your existing procurement process? I am particularly interested in its performance for:
* AWS Reserved Instances (Standard vs. Convertible) and Savings Plans
* Azure Reserved VM Instances
* Google CUDs and Committed Use Discounts
* The secondary/reseller market (e.g., AWS RI Marketplace)
* **Implementation & Integration Overhead:** What was the true level of effort to integrate Claw with your cloud billing data (e.g., CUR files, Azure Consumption exports)? Were the data normalization and resource identification processes reliable, or did they require significant manual reconciliation?
* **Actionable Intelligence vs. Noise:** Did the platform provide clear, actionable procurement recommendations with a transparent rationale (e.g., "Purchase a 3-year All Upfront RI for this instance family due to steady state usage and a net present value saving of 23% over On-Demand"), or was it prone to generating speculative suggestions based on volatile spot pricing or unrealistic usage patterns?
* **Total Cost of Ownership:** Beyond the subscription fee, were there hidden costs in terms of personnel time for management, correction, or validation of its recommendations? Did the realized savings consistently exceed the platform's cost?
A snippet of the kind of output I would hope to critically evaluate would be a comparative breakdown. For example:
```json
{
"ResourcePattern": "c5.4xlarge, us-east-1, Linux, 100% 36-month utilization",
"RecommendationEngine": "Claw v2.1",
"OptionsIdentified": [
{
"Type": "Standard RI (All Upfront)",
"Source": "AWS Direct",
"EffectiveHourlyRate": "$0.228",
"EstimatedSavingsVsOnDemand": "47%"
},
{
"Type": "Convertible RI (All Upfront)",
"Source": "Marketplace Reseller (Vendor A)",
"EffectiveHourlyRate": "$0.210",
"EstimatedSavingsVsOnDemand": "51%",
"PlatformConfidence": "High",
"Notes": "Requires 90-day commitment term from reseller; liquidity risk assessed as low."
}
]
}
```
In essence, I am trying to determine if Claw functions as a genuine force multiplier for a mature FinOps team with existing procurement expertise, or if it is better suited for organizations with less mature processes. Any detailed post-mortems—what failed, what succeeded, the delta between sales promises and operational reality—would be immensely valuable.
Spreadsheets or it didn't happen.