How Insight helps electoral services teams turn existing council data into reliable matching, during the annual canvass
- DC Marketing Team
- Jul 31
- 6 min read
Using local data to create a more efficient annual canvass.
Insight helps electoral services teams turn existing council data into reliable matches, reduce avoidable Route 2 activity and strengthening the electoral database.
Using local data to create a more efficient annual canvass
How Insight helps electoral services teams turn existing council data into reliable matches, reduce avoidable Route 2 activity and strengthen the electoral database.
Electoral services teams already have access to valuable information across their authority. The challenge is bringing that information together in a structured, controlled way so it can support accurate electoral matching and better canvass decisions.
Insight, available within Elector8, is designed for this purpose. It creates a separate data warehouse from local sources, removes duplication and then compares the resulting records with the electoral database. For annual canvass preparation, this gives teams another evidence base for confirming electors locally and improving the number of properties that can follow the more efficient Route 1 process.
Why local data matching matters
At the start of canvass preparation, electors are treated as not matched until evidence allows that status to be upgraded. The national data match is a key part of the process, but it is not the only information available to an electoral registration team. Local authority records such as council tax and housing benefit data can provide further evidence that a person is associated with an address.
Insight allows that local evidence to be used in a consistent and manageable way. When a master record in Insight is linked to an existing elector, that individual can be treated as matched for canvass purposes regardless of the result returned by the national data match.
An important qualification A successful local match does not automatically create a Route 1 property. Every elector at the address must be matched. If one person remains unmatched, the property can still fall into Route 2. The value therefore comes from improving the overall quality and completeness of matching, not from treating every individual match as a one-for-one route change. |
A data warehouse, not simply a search tool
Insight is based on data warehousing rather than data mining. A data-mining process usually takes a file and applies direct instructions to each row. Insight takes a broader approach: it builds a new database from one or more local sources and creates a set of master records that can be compared with Elector8.
This distinction matters because the same person may appear in several source files. A resident could be present in both council tax and housing benefit data, for example. Insight is designed to recognise that these records may represent one individual, combine them into a single master record and avoid introducing competing duplicates into the matching process.
Creating one reliable view from different sources
The first stage is to import local data and translate it into the field structure used by Elector8. Incoming files may use different names for the same information, so the import wizard allows each useful column to be mapped to the appropriate person, property or address field. Irrelevant fields can be excluded.
The quality of the source data has a direct effect on the amount of automatic matching that can be achieved. Separate fields for forenames, surname and address lines make matching easier than a single full-name or full-address field. Insight can still work with less structured data, but more manual refinement may be required.
The Unique Property Reference Number is especially valuable. UPRN is not mandatory, but it provides the strongest link between property records held in different systems. Where it is available and reliable, it can substantially reduce the property-matching work required before individual residents are compared.
How Insight supports canvass preparation
For teams using Insight primarily ahead of the annual canvass, the objective is to establish as many dependable links as possible between local master records and existing electors. This can begin before the national data match has taken place, giving teams more time to obtain, prepare and process local files.
Insight uses a two-stage matching process:
1 | Find the property Property matching rules identify the relevant address first, using fields such as UPRN, postcode and address lines. |
2 | Find the person at that property Once a property is identified, person matching rules compare the incoming resident record only with electors linked to that address. This keeps processing focused and reduces the risk of matching to the wrong person elsewhere in the authority area. |
Rules can begin with exact, high-confidence combinations such as full name and date of birth, then become progressively more flexible where required. For example, contained-name matching can help with common shortened names where one version appears within another. High-confidence matches can be accepted in bulk, while less certain results remain available for review.
A controlled and iterative process
Insight is intended to support professional judgement rather than replace it. A practical matching exercise normally starts with a small number of strict rules that can identify a large volume of safe matches quickly. After those results are accepted, the remaining records can be reviewed to understand why they failed and whether carefully adjusted rules can resolve them.
This iterative approach is important for both accuracy and performance. A very large set of rules can slow the process significantly because each property rule and person rule must be tested in order. Starting with the strongest evidence, clearing confirmed results and then focusing on exceptions gives teams greater control over the balance between automation and review.
Supporting empty-property decisions
Empty properties are a particularly useful application of local data. They would otherwise default to Route 2, but accurate local empty-property information can support a Route 1 canvass where the relevant conditions are met. Insight keeps empty-property data separate from occupied-resident files and applies different processing to it, reflecting the fact that there is no resident to match at the address.
A dedicated empty-property match indicator also helps teams avoid repeatedly trying person-matching rules where there is nobody recorded as living at the property. This makes the final stages of exception handling more focused.
The right data makes the process more effective
The most useful source file is complete, consistent and structured at individual-resident level. For recurring imports, Insight expects a full supply of current residents rather than only a list of people who became newly liable during that period. A partial file can create a false impression that everyone who appeared previously but is absent from the latest file has left the property.
Good practice therefore includes:
· supplying the full current population represented by that source each time, rather than only recent changes
· using one row per resident at each address
· providing occupied and empty properties as separate imports
· keeping file structures and field names consistent over time
· including UPRN and separated name and address fields wherever possible
For a new or cleared Insight database, the largest and most complete file should normally be imported first. It becomes the foundation of the master-record database, with smaller sources then compared against it to prevent duplication.
More than a once-a-year canvass tool
Many electoral services teams will use Insight as an annual canvass preparation tool, but its wider value is in continuous data improvement. Regular weekly or monthly imports allow the master-record database to evolve as local records change. Insight can recognise consistent appearances of the same person, connect new source information and create a developing view of occupancy over time.
This ongoing model can help teams identify potential new electors, possible amendments and records that may require deletion or investigation. The elector record in Elector8 can also show the source files in which a linked person has appeared, giving administrators useful context when reviewing the evidence behind a match.
A stronger use of data teams already hold
Insight helps electoral administrators make structured use of data that already exists within the authority. Its value is not simply in producing more matches, but in creating a reliable method for combining sources, removing duplication, applying clear matching rules and directing human attention to the records that genuinely need review.
Used ahead of the annual canvass, this can reduce avoidable Route 2 activity and support a more efficient workload. Used throughout the year, it can become part of a broader approach to maintaining a more complete, current and evidence-based electoral database.
To discuss how Insight could support local data matching and canvass preparation within your authority, contact the Democracy Counts team.
Comments