A notification that does not arrive because the address is not correct. A person who appears duplicated in two municipal registers. A file that is difficult to locate because it has not been classified correctly. A piece of equipment that is listed as an asset, but has not been in service for a long time. An economic data that does not match between two reports.
These are some situations that are sometimes common in the administration. They don't always make noise, they don't always have an impact on citizens and are often resolved with manual effort, calls, checks and a lot of experience from municipal staff.
But they all have one thing in common: behind them there is a data quality problem.
In previous posts we have talked about why data governance is necessary, the importance of managing it well, knowing what data we have and describing it with metadata. The next step is equally important: ensuring that this data is reliable enough for the use we want to make of it.
Incorrect data is not just an error in a database, but can end up generating a poor public service.
What does it mean for data to have quality?
When we talk about data quality, we're not talking about absolute perfection. We're talking about something more practical: that the data is suitable for the use that the city council needs to make of it.
A piece of data may be sufficient for internal indicative use, but not be good enough to make a formal notification, grant aid, publish information on the municipal website or build a management indicator.
For example, a roughly written address may be useful for someone in the municipality to identify an area, but it may not be vàlidto send an electronic or postal notification. A list of activities can be used to prepare an internal agenda, but if it is not updated it can generate confusion when published to the public. A list of facilities may seem correct, but if it does not indicate who manages them, what services they offer or whether they are open, it may not be very useful for planning.
Therefore, quality data is data that is correct, complete, up-to-date, coherent, understandable, unique when it needs to be, and Useful for the intended purpose. Simply put: good data is data that we can rely on to work better.
Data quality is not just a technical issue
In a city council, it is common to think that the quality of data depends on the computer program or the provider. And it is true that tools can help a lot: they can validate fields, avoid duplicates, normalize formats or generate warnings. But the quality of data is not only the responsibility of technology.
Quality begins the moment a person enters, modifies, interprets or reuses information. This affects the census, registry, secretariat, intervention, urban planning, social services, citizen services, culture, sports, local police, archives, document management and any other municipal service.
If each service records information with different criteria, if it is not clear who should update a piece of data, if no one reviews recurring errors or if meanings are not documented, technology alone will not solve the problem.
Data quality is a shared, achievable, practical and proportional responsibility.
Common mistakes that end up costing time and trust
Data quality problems tend to appear in very specific forms.
- There may be data. duplicates: the same person, company, entity or facility appears registered more than once.
- There may be data. incomplete: An important field is missing, such as an address, identifier, date, status, or classification.
- There may be data. outdated: information that was correct at the time, but no longer reflects reality.
- There may be data. incoherent: two municipal systems contain different information about the same element.
- There may be data. incomprehensible: codes, abbreviations or fields that only a specific person understands.
- There may be data. misclassified: files, documents, activities or incidents that do not follow a common criterion.
- And there may be data that are not traceable: we don't know where they come from, who modified them or when they were updated.
These problems have consequences. They waste time, generate distrust, make internal coordination difficult, complicate transparència and can directly affect citizens.
When data quality affects public service
Data quality is not an internal issue with no impact. It has direct effects on the way the city council delivers services.
- If the data of have contacted they are not correct or shared, communication with citizens fails.
- If the data of the files are not well informed, processing time increases.
- If a person's data is duplicates, there may be errors in the care or management of aid.
- If the data of the equipments are not updated, incorrect information may be provided about schedules, services or availability.
- If the data of procurement, subsidies or budget are not coherent, making efficient management and accountability more difficult.
- If the data territorial are not well aligned, it can be more complex to plan urban actions, maintenance or services.
Citizens do not see “data problems”. Citizens see notifications that do not arrive, procedures that take too long, lack of internal coordination, contradictory information or services that do not respond well enough. Therefore, improving the quality of data is improving public service.
Not all data requires the same level of quality
A city council cannot review all data with the same intensity. Nor should it. The key is to prioritize. There is data that is more critical because it affects citizens' rights, essential procedures, legal obligations, sensitive services or important decisions. This data must have more control.
For example, population data, economic data, data linked to social services, data from administrative files, notification data or data necessary for the transparència may require more guarantees than other data for more internal or informative use.
Data quality must be managed based on risk and public value criteria.
This means asking yourself simple questions:
- What data do we use the most?
- What data generates the most errors?
- What data most directly affects citizens?
- What data is necessary to make decisions?
- What data do we share with other administrations?
- What data could have the most impact if it is incorrect?
These questions help decide where to start.
Where can you start?
One can start improving data quality without large projects or sophisticated tools, by following these steps:
- Identify two or three pieces of data that generate common problems. For example: lack of standardization of addresses, duplicate people, equipment, files, activities, incidents or contact details.
- Talk to the people who work with this data every day. They know where there are errors, which fields are left blank, which criteria are unclear, and which checks are done manually.
- agree on basic criteria. For example, how an address should be written, which fields are mandatory, how an entity is identified, who can modify a piece of data or when information needs to be reviewed.
- Review regularly. You don't have to wait until you have a serious problem. You can review duplicates, empty fields, old data, or inconsistencies on a regular basis.
- Document the criteria. Creating a short guide, data sheet or internal instruction can prevent many future mistakes.
These actions are simple, but they have an important effect: they reduce dependencies, improve trust, and make information more useful for everyone.
Quality starts before analysis
More and more town halls want to have indicators, dashboards, open data or artificial intelligence tools.
It is a good direction, but we must keep in mind a basic idea: if the data from which it is based is not reliable, the results will not be reliable either.
- An indicator built with incomplete data can lead to a wrong decision.
- A dashboard with outdated data can create a false sense of control.
- An artificial intelligence tool fed with misclassified data can amplify errors or biases.
That's why data quality is not an afterthought. It's not a final review. It is a prerequisite for generating reliable knowledge..
Before asking ourselves what indicator we want, we need to ask ourselves if the data that feeds it is good enough.
The relationship with inventory and metadata
Data quality is closely related to the steps shared in previous articles.
- The inventory helps us know what data we have.
- Metadata helps to understand what it means, who maintains it, where it is, how often it is updated and what limitations it has.
- Quality helps us know if this data is good enough for the use we want to make of it.
They are three connected pieces.
- If we don't know what data we have, we can't review it.
- If we don't know what they mean or who is responsible for them, we can't correct them properly.
- If we don't know what we want to use them for, we can't decide what level of quality they need.
Therefore, a data governance model must integrate inventory, metadata and quality as part of the same process.
The role of the Smart Local Government Network
The Smart Local Government Network aims to provide value through its working groups, helping to define a common data governance model for the local world. This model must be simple, practical and adapted to the diversity of local entities.
In the area of quality, the Network wants to establish shared criteria on which data are priorities, what minimum controls need to be applied, what quality indicators can be useful and how to detect common problems.
It also wants to help define common models of responsibility: who validates data, who maintains it, who detects incidents, who resolves them and who decides whether data is suitable for a certain use.
Better data, better services
Data quality may seem like an internal issue, but it has a direct impact on citizens.
When data is better, procedures can be more agile, notifications more reliable, reports more useful, services better planned and transparència more understandable.
Improving data quality doesn't mean making everything perfect. It means starting to reduce the errors that most affect municipal operations and public services.
It's not just about having data. It's about being able to trust it.
Because when data is reliable, the city council can make better decisions. And when the city council makes better decisions, citizens receive a better service.
