Highly trusted
Strong source confidence plus repeated confirmation from operational use.
Product
Data Quality
Happypath is transparent about data quality. Every relevant data point is scored on a scale from 0 to 100 so customers, partners, and internal teams can understand how much confidence to place in the data that powers real-world arrivals.
We do not treat quality as a hidden internal metric. We use it to communicate trustworthiness clearly, and we actively improve freshness and relevancy through source review, operational feedback, boots-on-the-ground operations, and targeted follow-up when conditions change.
Production threshold
Below 50 means not trustworthy
Any quality score below 50 is considered not trustworthy and should not be used in production routes or delivery situations.
These records are subject to continual review by Happypath staff until the data has been revalidated, corrected, or replaced.
Freshness and relevancy
Keeping data fresh and relevant is part of the product, not an afterthought. Positive confirmations keep trusted data strong, negative signals reduce trust quickly, and low-confidence records trigger proactive review so outdated paths, doors, or metadata do not stay in production unnoticed.
Happypath runs boots-on-the-ground operations in all coverage areas so the dataset can be checked, corrected, and restored when real-world conditions change.
Operational quality loop
Freshness is actively maintained
Happypath quality starts with the source, but it does not stop there. Feedback changes trust over time, thresholds make declining confidence visible, and operations teams step in when the data is no longer good enough for production use.
This is a core part of the product. In every coverage area, Happypath runs boots-on-the-ground operations that verify assets in the real world and restore data quality when entrances, paths, elevators, or access conditions have changed.
How to interpret the score
The score is meant to be understandable at a glance. It reflects both the initial source quality and what later operational feedback says about whether the data is still current and relevant.
Operationally strong
Reliable enough for production workflows, while still open to further improvement.
Use with care
Visible and reviewable, but should be validated in the right operational context.
Not trustworthy
Not suitable for production routes or delivery situations and queued for review.
Source-based initial score
Happypath collects information from multiple sources, and those sources are not treated equally. Each data type starts with an initial score based on the relative trustworthiness, proximity, and assumed competence of the source.
Street address
Base address data starts with the strongest trust in direct public-source integrations.
| Source | Initial quality | Notes |
|---|---|---|
| Public registry | 100 | Direct integrations with public sources. |
| Address list vendor | 90 | Purchased lists of addresses. |
| Manually entered address | 50 | Not currently supported. |
Entry door
Doors are judged by both source trustworthiness and how directly the source knows the property.
| Source | Initial quality | Notes |
|---|---|---|
| Happypath remote mapping staff | 70 | Entered by Happypath staff using our remote mapping tools. |
| Property manager | 90 | Data entered via one of our portals for property managers. |
| Courier / Happypath in-field mapping | 80 | Data is approved and QA'd by Happypath staff before entering the dataset. |
| Recipient / tenant | 90 | Data sourced from collection flows completed by end users. |
Entry door path
Paths change more often than doors, so their starting scores are slightly more conservative.
| Source | Initial quality | Notes |
|---|---|---|
| Happypath remote mapping staff | 60 | Entered by Happypath staff using our remote mapping tools. |
| Property manager | 80 | Data entered via one of our portals for property managers. |
| Courier / Happypath in-field mapping | 70 | Data is approved and QA'd by Happypath staff before entering the dataset. |
| Recipient / tenant | 80 | Data sourced from collection flows completed by end users. |
Metadata and generic assets
Access metadata and generic assets are trusted most when they come from people closest to the property.
| Source | Initial quality | Notes |
|---|---|---|
| Happypath remote mapping staff | 70 | Entered by Happypath staff using our remote mapping tools. |
| Property manager | 90 | Data entered via one of our portals for property managers. |
| Courier / Happypath in-field mapping | 80 | Data is approved and QA'd by Happypath staff before entering the dataset. |
| Recipient / tenant | 90 | Data sourced from collection flows completed by end users. |
Feedback-based scoring
Quality stays dynamic after creation
After the initial quality score is assigned, the current quality for any asset or metadata record is recalculated based on user feedback. Each data point can receive either positive or negative feedback from real operational use.
Positive feedback
Each positive signal adds 5% to the current quality score, rounded to a whole number and capped at 100.
Negative feedback
Each negative signal cuts the current quality score by 50%, with a floor of 0.
Review policy
Low scores trigger action
A low score is not a passive warning. It is an operational signal that tells us the data needs attention.
- Scores below 50 are queued for revision by Happypath.
- Revisions may include contacting the original source, revalidating remotely, dispatching boots-on-the-ground operations, or sending the data back through field or portal workflows.
- This is how we keep the dataset relevant over time instead of pretending that old records remain trustworthy forever.
Worked example
If a Happypath staffer adds a door and a path, repeated positive outcomes steadily raise trust. But one strong negative signal can immediately reveal that a path is no longer fresh enough for production use.
| Event | Door score | Path score |
|---|---|---|
| Initial creation | 70 | 60 |
| Positive feedback #1 | 70 + 5% = 74 | 60 + 5% = 63 |
| Positive feedback #2 | 74 + 5% = 78 | 63 + 5% = 66 |
| Positive feedback #3 | 78 + 5% = 82 | 66 + 5% = 69 |
| Positive feedback #4 | 82 + 5% = 86 | 69 + 5% = 72 |
| Positive feedback #5 | 86 + 5% = 90 | 72 + 5% = 76 |
| Negative feedback #1 | No negative feedback for the door | 76 - 50% = 38 |
In this example, the path falls to 38 after negative feedback. That immediately puts the path below the production threshold and queues it for revision by Happypath. The door remains trusted because it did not receive the same negative signal.
Transparent quality and fresh data are part of the product
Happypath does not ask customers to blindly trust hidden data pipelines. We expose quality clearly, respond when signals show that data is going stale, and keep improving the dataset so it stays useful in real operational environments. That includes boots-on-the-ground operations in every coverage area to keep data fresh, relevant, and production-ready.