Methodology

How the map is built

Sources, comparisons, boundaries, robustness tests, corrections and every column.
Powered by FitzrovAI

UK Business Crime Map: methodology and data dictionary

Dan Sandford, Member Business · dataset v66 · November 2026

Purpose and scope

The UK Business Crime Map compares recorded shoplifting inside each Business Improvement District (BID) with the area around it, over the same two years. It asks one question: did shoplifting in the BID change faster or slower than in its surroundings?

It does not measure whether a BID caused any change. A BID area that did better than its surroundings may have done so because of the BID, a new store, a closed store, a police operation, or a change in how a force records crime. The map shows where the numbers diverge; local knowledge explains why.

Coverage:

Out of scope:

The primary measure is shoplifting. All crime and anti-social behaviour (ASB) are reported as context, not verdicts.

Data sources

Every crime figure in the zone results comes from open police.uk data. The only modelled figures on the site are the looking-ahead ranges, which are labelled as expectations and checked against police data when it arrives.

SourceWhat it providesNotes
police.uk street-level crime filesEach recorded crime: month, type, approximate location (a map "snap point"), area codeSeptember 2023 to September 2026, 37 months per force (verdicts use the first 36)
police.uk outcomes filesThe latest outcome for each crimeUsed for charge and no-suspect rates; Northern Ireland publishes none
police.uk stop-and-search filesSearches with location and outcomeContext only; coverage varies by force
ONS Police Force Area data tablesOfficial police recorded crime by force and local authorityUsed to check police.uk completeness (see below)
GLA London BID boundary fileOfficial boundaries for London BIDsGrade A
Greater Manchester Police, Freedom of Information reply FOI/26/017957/N (12 Nov 2026)Shoplifting counts by month and police beat, September 2023 to August 2026, for Manchester, Trafford, Bury and RochdaleGrade D (Greater Manchester only)
Council and BID business plans, ballot papers, mapsStreet lists, maps, term dates, servicesGrade B where an official street list exists
Council Freedom of Information replies (via WhatDoTheyKnow)Levy property lists (address and rateable value only), boundary maps and files, ballot resultsGrade A or B; property lists are used only to derive street names and are not published
British BIDs index and BID Data registerWhich BIDs exist and their typeBritish BIDs used as the reference
NABCP members page (National Association of Business Crime Partnerships), saved 19 November 2026Business crime partnerships: name, accredited or member, websiteContext only; not used in any verdict

Police.uk locations are deliberately approximate. Each crime is moved to the nearest of a fixed set of snap points (a street, a car park, a supermarket, a shopping area) to protect victims. The analysis works with those points, not exact addresses.

The comparison

Each zone gets a gap: its percentage change in shoplifting minus its comparator's percentage change.

Gap = (Y3zone ÷ Y1zone − 1) × 100 − (Y3comparator ÷ Y1comparator − 1) × 100

Window adjustments where a force's data has a break:

ForceMonths used
Devon and CornwallDecember to August (Sep–Nov 2023 under-recorded)
HampshireSeptember to June
NorthumbriaJanuary–February excluded
GloucestershireSeptember to January (data ends January 2026)
Dyfed-PowysJanuary–April excluded (shoplifting missing Jan–Apr 2024)

The same months are used in both years, so the comparison stays like for like.

Boundaries and grades

Every zone carries a boundary grade.

GradeMeaningTypical source
AOfficial GIS boundaryGLA London file; council GIS layers
BOfficial street list or official mapBID business plan appendix, ballot proposal, current BID map
CTraced or own definitionMap traced by hand, core shopping streets chosen by the analyst, or a street list from an unverified source
DPolice beatGreater Manchester only: counts by police beat from a Freedom of Information reply (see below)

How a zone is built from a street list: two methods are used, and each row says which. Hull method (most zones built before November 2026): crimes on listed streets are matched to their snap points; a convex hull is drawn around those points and extended by 40 metres. Where named points are too sparse for a hull, a radius around the town centre is used instead. Street-matching method (zones upgraded from council levy lists or business-plan street lists since 17 November 2026): every police.uk point on a listed street within a stated distance of the town centre is counted, plus unnamed points (such as "Supermarket" or "Parking Area") within 150 metres of a counted point; the distance limit is tested 300 metres tighter and wider. No hull is drawn, so these zones have no edge of their own.

Core-cluster extension (grade C only): where a hull missed a town's main shopping point, a 150 m circle around the densest town-centre shoplifting cluster is added. That cluster is located using the middle year (September 2024 to August 2025), which favours neither side of the comparison. See the corrections log.

Shapes on the map: every BID except the three Greater Manchester zones has a shape: 94 official and 189 provisional. The shape style is separate from the grade: a grade B zone counted from a street list has an official source but no official edge, so it is drawn as provisional until a boundary file or placeable map arrives. A shaded shape with a solid edge is an official boundary: a digital boundary file, or an official map traced and placed on police.uk street points (typically within 15 to 30 metres). A paler shape with a dashed edge is a provisional boundary: the area we counted, drawn as a 70 metre band around the police.uk points used for that zone and smoothed. It shows where the figures come from, not where the BID's legal boundary runs, and it will be wrong in places, especially where police.uk has few named points. Provisional outlines are replaced as official boundaries arrive; BIDs and councils can send corrections through the contribute page. Greater Manchester zones, matched by police beat, have no shape.

Police points: zoomed into a town, the map shows every police.uk reference point in England, Wales and Northern Ireland that had shoplifting recorded in either comparison year (September 2023 to August 2024, or September 2025 to August 2026), inside or outside a BID, about 52,000 points in all. Police.uk does not publish where an offence happened: it moves each one to the nearest point on a fixed national list of street centres and labelled sites (for example a supermarket or a car park), so a pin marks that reference point, not a business or an exact location. Pin size shows the number of shoplifting records in the latest year (September 2025 to August 2026) in bands: 1 to 9, 10 to 49, 50 or more; faint pins had records only in the earlier year (September 2023 to August 2024), which the hover text also shows. Exact counts are not shown, and points police.uk labels as a supermarket, petrol station, nightclub, hotel, theatre, conference centre or motorway services are labelled only as "a commercial premises", and other labelled sites by type only ("a shopping area", "a car park", "a railway station" and so on), so the map does not single out individual businesses. Records assigned to police stations are left out, as they are in the analysis. The same rules apply everywhere. Source: police.uk street-level data, Open Government Licence v3.0.

Business crime partnerships: each partnership on the NABCP members page (45 accredited under NABCP's National Standards, 61 members; York BID appears in both lists) is shown as published there: name, status and website link. A partnership named for one BID or town is attached to that area's popup (107 BID zones and 12 town centres without a current BID). One named for a district or county (for example Cumbria, Warwickshire, Cheshire and North Wales) is attached, marked "area-wide", to every zone in that area, and does not get its own marker. Fourteen towns with a partnership but no zone on the map are shown as small squares at the town centre, with no BID result. Five national or multi-site bodies (Retailers Against Crime, Safer Business Network, UK Partners Against Crime, URW Westfield, Yorkshire Business Crime Network) are listed here only. Two links on the NABCP page are left out: Eastbourne Business Crime Group links to another BID's website, and the Rhondda Cynon Taf and Merthyr Tydfil link does not work. Membership or accreditation says nothing about how well a partnership works, and is not used in any verdict.

Search: the map search matches BID and zone names from this dataset. Towns, places and postcodes are looked up with postcodes.io (Ordnance Survey Open Names and ONS postcode data); only the text typed is sent, and the site records only that a place or postcode search was used, never what was typed.

The map: each zone is shown as one dot at its approximate centre, not its boundary. Centres come from the zone as drawn for the analysis, the official GLA boundary for London BIDs, or the town or district centre where neither was stored; every dot was checked to fall in the right council area. The points are published as zone_points.csv on the data download page. Where an official boundary is drawn, the dot sits at its centre.

Grade D (police beat): used only where a force does not publish locations. The BID is matched to the police beat or beats that cover it by name, and compared with the rest of its district. The area is approximate, there is no supermarket or single-location check, and other crime types are not available, so a grade D result is never marked robust. On the map these areas are rings rather than solid dots. Before use, the force's district totals are checked against ONS recorded crime: for Greater Manchester they agree within 2–7%, and the change between years within 3 points.

Grade C results should be read with extra caution. The open call to councils and BIDs (last section) is designed to move them to A or B.

Robustness tests

A verdict is marked robust only if it survives every test below. If any test changes the verdict, it is marked not robust. The rule is deliberately conservative: 116 of 227 directional BID verdicts are robust (66 fell, 50 rose).

TestWhat it doesWhy
Without supermarketsRecomputes the gap with all supermarket snap points removed, in zone and comparatorOne store opening, closing or changing its reporting can swing a whole zone
Without the top locationRemoves the single police.uk map point (location name and coordinates) with the largest change. Before v58 this test removed every point sharing the same location name, so generic names such as "Supermarket" or "Shopping Area" were removed togetherCatches results resting on one spot
Boundary variantsHull zones: recomputes with a strict hull (no buffer) and with street-name matching. Street-matching zones: recomputes with the distance limit 300 metres tighter and 300 metres widerShows whether the verdict depends on where the line is drawn
Re-pinning checkFinds a point that drops sharply while a point within 400 m on the other side of the boundary rises by a similar amount in the same months; merges the pair both wayspolice.uk sometimes moves crimes from one snap point to another
Comparator distortionFlags comparators dominated by out-of-centre retail or a force-wide surgeApplied to Cleveland and noted for Ipswich

Small numbers: zones with fewer than 100 shoplifting offences in Y1 are flagged small_numbers = Yes and marked with an asterisk on the dashboard. Their gaps can be very large on a handful of offences, so read them as indicative only.

Statistical test: each directional verdict was also tested by asking whether the zone's year-three share of its two-year total differs from what the comparator's change would predict (a binomial test on the zone's counts, treating the comparator's change as fixed). 174 of 227 directional verdicts are significant at the 5% level, including 104 of the 116 robust verdicts (rerun on data version v66, 19 November 2026; the v58 run gave 174 of 230 and 102 of 114). Most non-significant results are in small zones.

Check against official statistics

Police.uk is a public extract of the same police records that feed the official statistics. To test whether the extract is complete, police.uk shoplifting counts were compared with the ONS Police Force Area data tables (Home Office police recorded crime, year ending June 2026 edition) for July 2024 to June 2025 and July 2025 to June 2026, for 41 forces, the City of London and 292 local authority or community safety partnership areas.

A second check compared the year ending September 2024 (close to our first year) with the year ending June 2026 (close to our third). It confirmed the same pattern and picked up the known Dyfed-Powys gap in early 2024. Thirty areas diverged by more than 5 points in the same direction in both checks, after removing the explained gaps.

Every zone is compared with the rest of its own local authority from the same source, so a shortfall spread evenly across an area cancels out of the gap; one concentrated in town centres would not. Annual totals cannot show which, so the 45 zones whose comparator lies in a flagged area carry a caution in their caveat rather than a correction. The full check is published as a CSV on the data download page. It will be rerun on the year ending September 2026 edition (due 28 January 2027).

What retailers count and what police record

The British Retail Consortium's annual crime survey uses the same September-to-August year as this map. Its 2025 survey covers September 2023 to August 2024, the map's first comparison year, and estimates about 20.4 million incidents of customer theft. Police.uk recorded about 461,000 shoplifting offences over the same months, roughly 1 in 44. So the map measures shoplifting that was reported to police and recorded, not how much theft took place. A change in a verdict can reflect a change in reporting, in police attendance or in recording, as well as a change in theft. The BRC changed its method in its 2026 report, which now counts verified incidents only, so its figures are not comparable year on year. The police.uk totals exclude Greater Manchester, which does not publish to police.uk.

Known data problems by police force

Seventeen forces have a recording break, a data gap or a publishing problem. Each is handled as shown.

ForceProblemHandling
Devon and CornwallSep–Nov 2023 under-recorded; ASB break from Dec 2025Dec–Aug window; ASB not reported
West MidlandsASB records roughly halve from Sep 2025ASB not reported for any West Midlands zone
HampshireJuly–August data unreliableSep–Jun window
NorthumbriaJan–Feb 2026 shortJan–Feb excluded
GloucestershireData ends January 2026Sep–Jan window
Dyfed-PowysShoplifting missing Jan–Apr 2024Jan–Apr excluded
ClevelandForce-wide shoplifting surge; comparators dominated by out-of-centre retailTeesside zones marked not robust
HumbersideForce-wide shoplifting jump from Mar 2026Absorbed by comparator; noted
LancashireASB halves from Jan 2026ASB treated with caution
West MerciaASB break Apr–May 2026ASB treated with caution
Northern Ireland (PSNI)No outcomes data; no area codesCharge rates blank; distance-based comparator
Bedfordshire, Norfolk, Nottinghamshire, South Yorkshire, GwentStop-and-search incomplete or missingStop-and-search not used for these forces
Greater ManchesterNot published on police.uk. FOI counts by police beat (no locations, no other crime types); GMP notes wrong or missing beat codes and duplicate addresses on some crimesGrade D: three BIDs matched by beat; district totals agree with ONS within 2–7%; never marked robust; no Looking ahead range

Corrections log

Every change that moved a published or circulated figure is logged here, newest first.

DateVersionChangeEffect
23 Nov 2026v66 (wording)Cambridge BID: plain-words verdict corrected from "Rose while its surroundings fell" to "Fell less than its surroundings" (764 to 761 is a small fall; the surroundings fell 20%). Ripon BID: caveat reworded from "Official boundary" to "Official street list", because the outline on the map is provisional. Member B (Beta), an AI assistant that answers only from the map's published data, added beside the map.No figures change. Plain-words counts: one fewer "Rose while its surroundings fell", one more "Fell less than its surroundings". The gap (+20) and the verdict colour are unchanged.
19 Nov 2026v66 (display)Page layout: one filter bar for the map, headline numbers, chart and table (with a new "Larger zones only" filter); search above the map; a compact map key; the chart hides small-number zones by default; table views and a row panel with the full caveat. Map colours changed so markers stand out from the base map in light and dark mode: in-line dots slate, robust dots solid with an edge, not-robust dots pale with a coloured ring, police points brown, and smaller when zoomed out so the BID dots stay clear.Display only; no figures change.
19 Nov 2026v66Blackburn: re-counted on the 75-street list in the BID's 2024–28 proposal (grade C to B), which includes Grimshaw Retail Park: 370 to 276 (−25%) against −39% for the rest of Blackburn with Darwen, gap +14, rose against trend (was 187 to 90, gap −19, fell against trend); in line (−4) if the outer streets are left out, so not robust. Barnstaple: North Devon Council confirmed under FOI that there is no BID; renamed Barnstaple (no BID) and moved to the controls (it was a robust rise, small numbers). Great Yarmouth: zone renamed Great Yarmouth Town Centre BID (the council confirms the area is the Town Centre BID; GYBID is the separate tourism BID); the council's levy list confirms the streets used; figures unchanged. Business crime partnerships added from the NABCP members page (three new columns). Totals now 272 BID verdicts (117 fell, 110 rose, 45 in line), 116 of 227 robust.One verdict changes (Blackburn fell → rose, not robust); one robust rise leaves the BID count (Barnstaple).
19 Nov 2026v65 (map display)Phones: dots grow as you zoom in and have a finger-sized tap area, fixing popups that would not open on tap. Full screen now includes the search box, and search also finds towns, places and postcodes.Display only.
19 Nov 2026v65 (map display)Full-screen button added to the map (top right), for phones and desktops.Display only.
19 Nov 2026v65 (methodology)Methodology brought up to date with data version v65: zone counts (322 zones, 273 BID verdicts), verdict and robustness tallies (117 of 228 robust), the significance test rerun on current data (175 of 228), small-zone and control figures, the street-matching method and its boundary test, grade D in the grade table, council Freedom of Information replies as a source, and how map shapes relate to grades.Text only; no figures in the dataset change.
19 Nov 2026v65 (map display)Police points added: when zoomed into a town, every police.uk reference point with shoplifting recorded (about 52,000 across England, Wales and Northern Ireland, inside and outside BIDs), sized in bands for the latest year (1 to 9, 10 to 49, 50 or more) with the earlier year in the hover text. Shops and similar sites are labelled only as "a commercial premises"; exact counts are not shown; police station points are left out.Display only; no figures change.
19 Nov 2026v65 (map display)Every BID now has a shape on the map. Three official maps traced and placed on police.uk street points: Saffron Walden (council OS map, typical error about 15 m), Banbury (council OS map, about 25 m) and Stockton (council 1:5,000 map, about 20 m); their dots move to the centre of the traced shape. 190 BIDs without an official boundary get a provisional outline of the area counted, shown paler with a dashed edge. Maps for Wilmslow, Crewe, Knutsford, Brentwood Connected, Bilston and Bognor Regis were too coarse or unlabelled to place reliably, so these stay provisional. Homepage wording: 'redrawn' corrected to 're-counted' for the v65 street-list update.Display only; no figures change.
19 Nov 2026v65Nine areas redrawn on official street lists (grade C to B): levy lists from councils under Freedom of Information (North Yorkshire for Northallerton, Ripon and Skipton; Uttlesford for Saffron Walden; Cheshire East for Crewe and Knutsford), street lists in BID proposals (Crewe, Knutsford, Wilmslow and Bognor Regis) and Brentwood Borough Council's street list for each of its three centres. Northallerton 137 to 74, gap -31, fell against trend (was 109 to 57, gap -30, fell against trend); Ripon 56 to 68, gap -7, in line (was 53 to 57, gap -21, fell against trend); Skipton 47 to 53, gap +12, rose against trend (was 35 to 58, gap +81, rose against trend); Saffron Walden 28 to 17, gap -55, fell against trend (was 28 to 17, gap -55, fell against trend); Crewe 168 to 182, gap +18, rose against trend (was 170 to 158, gap +2, in line); Knutsford 41 to 44, gap +16, rose against trend (was 33 to 18, gap -37, fell against trend); Wilmslow 123 to 83, gap -26, fell against trend (was 123 to 84, gap -25, fell against trend); Bognor Regis 130 to 175, gap +16, rose against trend (was 129 to 181, gap +21, rose against trend); Brentwood Connected 334 to 445, gap +9, in line (was 334 to 446, gap +11, rose against trend). Harrogate's levy list was also received; its boundary is kept, as it is already built from the BID's official street list. Other crime types and outcomes are blank until recalculated; Looking ahead ranges are withheld until the next forecast round.Verdict changes: Crewe in line → rose, Knutsford fell → rose, Ripon fell → in line, Brentwood Connected rose → in line. Skipton's gap narrows from +81 to +12. None of the changed results is robust. Saffron Walden becomes robust (same figures, rule now applied consistently).
18 Nov 2026v64Bilston: redrawn on the 23-street list and Appendix 2 map in the BID's 2026–31 business plan (grade C to B), counting the supermarket point and the Pinfold Street service road behind the High Street and Church Street shops: 210 to 282 (+34%) against +7% for the rest of Wolverhampton, gap +27, still rose against trend (was 161 to 246, gap +46), still not robust. Other crime types and outcomes are blank until recalculated; the Looking ahead range is withheld until the next forecast round. Also: grey markers darkened so they show in light mode.Bilston's gap narrows from +46 to +27; verdict unchanged.
17 Nov 2026v63Grimsby: North East Lincolnshire Council confirmed under FOI that there is no BID in its area. The zone (listed as a BID in error, from a directory entry) is renamed Grimsby (no BID) and moves to the town centres without a BID. Its figures are unchanged; it no longer counts among the BID results. The fixed Looking ahead files keep the old name, Grimsby BID. Stockton: the council confirmed under FOI that the BID lost its renewal ballot on 18 June 2026 (by number of votes) and its term ended on 31 October 2026. It was in place for the whole comparison window, so its result stands and is now labelled as an ended BID. Boundaries upgraded from council FOI levy lists (grade C to B, streets on the official list): Banbury 91 to 87, gap −9, now in line (was 91 to 83, gap −14, fell against trend, not robust); Stockton 215 to 166, gap −163, still fell against trend (was 244 to 204, gap −158), still not robust under the Teesside rule. Their other crime types and outcomes are blank until recalculated, and their Looking ahead ranges are withheld until the next forecast round.One BID fewer in the BID totals; one more control. Banbury moves from fell to in line.
17 Nov 2026v62Greater Manchester added (grade D, police beat counts from GMP under Freedom of Information): Manchester city centre BID 1,588 to 1,781 (+12%) against +3% for the rest of Manchester, in line; Altrincham BID 76 to 132 (+74%) against +14% for the rest of Trafford, rose against trend (small numbers); Rochdale Town Centre BID 345 to 171 (−50%) against +11% for the rest of Rochdale, fell against trend. None marked robust.Three new results; no existing figure changed.
17 Nov 2026v61 (map display)Official boundaries drawn on the map for 91 areas: 78 London BIDs from the Greater London Authority BID boundary file (October 2026 edition, the same file the analysis uses; grade A); Fleet and Nottingham (grade A, council digital files); Hounslow’s Heart, Welwyn Garden City, Plymouth City Centre, Brighton, High Wycombe, Morecambe, Oswestry, Otley, Lincoln, Wimborne and Cardiff city centre (grade B, traced from official maps). Dots for these areas moved to the centre of their boundary. The GLA notes that its boundaries are digitised from the BIDs’ own maps and are indicative.None on any figure. Display only: the boundaries are the ones the analysis already used.
16 Nov 2026v61Seven boundaries upgraded from council Freedom of Information replies: Fleet (Hart District Council digital boundary file, grade A); Welwyn Garden City (official OS-base map and street list, grade B); Plymouth City Centre (definitive boundary map in the 2025-30 plan, calibrated to the 30-street list, grade B) with Plymouth Waterfront recut so the two no longer overlap; Cobham (streets from the council levy list, grade B); Oxted and Caterham (official street lists, grade B). For Cobham, Oxted and Caterham only shoplifting has been recalculated so far; other crime types and outcomes are blank until the Surrey force data is reloaded. New column ons_caution flags results whose caveat carries the police.uk v official statistics caution. Looking-ahead ranges for the seven redrawn zones are withheld on the map until the next forecast round; the fixed files are unchanged. Scorecard by area size added to the Looking ahead page.Fleet 112 → 37 becomes 115 → 43 (−63% v +80%), still fell, robust; Welwyn Garden City 336 → 426 (fell, not robust) becomes 340 → 453 (+33% v +42%), in line, not robust; Plymouth City Centre 320 → 392 becomes 288 → 347 (+20% v +51%), still fell, robust; Plymouth Waterfront 21 → 165 becomes 23 → 165, still rose, not robust; Cobham 52 → 47 becomes 35 → 33 (−6% v +18%), still fell, not robust; Oxted 60 → 95 becomes 24 → 50 (+108% v +23%), still rose, now robust; Caterham 25 → 41 becomes 35 → 89 (+154% v −6%), still rose, now robust. All three Surrey zones remain small numbers. BID zones 263: 115 fell, 100 rose, 42 in line.
16 Nov 2026v60Renamed the UK Business Crime Map. Monthly charts extended to September 2026. Looking ahead (Beta) added: expected ranges for October 2026 to February 2027, fixed in dated files, with September 2026 checked in public. Data downloads moved to a data page with terms of use.No zone figures or verdicts changed
10 Nov 2026v58cHounslow's Heart BID: boundary upgraded from an own-drawn hull (grade C) to the BID area map in the Hounslow's Heart Business Plan 2025–30 (grade B); the BID confirmed there is no GIS file. The BID-timing note corrected: only the first three months of year 3 (Sep–Nov 2025) are before the BID began, not most of the year. Contribution form: an "add your area" option at the top of the area list.Hounslow's Heart 333 → 561 (+73) becomes 289 → 494 (+75), still rose against trend, robust. Headline counts unchanged.
10 Nov 2026v58Plain-words verdict on every result (new column verdict_plain), after a BID pointed out that "fell against trend" reads as "crime fell": in 32 BID zones recorded shoplifting rose, but by less than the comparator. Official boundaries: High Wycombe (Buckinghamshire Council boundary map, FOI), Lincoln (current BID map), Brighton (BID area map), Wimborne (2026-31 business plan map), Oswestry (2023-28 business plan map, Ordnance Survey base) and Cardiff city centre (Cardiff Council BID2 map, FOI); the Butetown & Bay control is redrawn so it no longer overlaps the official BID2 area. "Without the top location" test changed to remove one map point, not every point sharing a location name; the methodology had described it as one point. Robust flag changes from that test: Clacton (no BID) | Town centre (now robust), Gloucester BID (now robust), Worcester BID (now robust), Blackpool Town Centre BID (now robust), Chester BID (now robust), Lowestoft BID (Lowestoft Vision) (now robust), Port Talbot BID (Viva Port Talbot) (now robust), Coleraine Connect (Coleraine BID) (now robust), Kingston First (now robust). The test was rerun for every assessed zone except Chelmsford and Otley, whose top locations are named streets well clear of the threshold; pilot zones already used single points. Month-by-month chart added to map popups (new file data/monthly.json). Salisbury: local context from the Police and Crime Commissioner's office. Note added that contribution form submissions are handled by Netlify.High Wycombe 350 → 219 (−59) becomes 504 → 433 (−35), still fell, robust; Lincoln 608 → 518 (−19, fell, not robust) becomes 511 → 517 (+3, in line, robust); Brighton 227 → 374 (+49, rose) becomes 664 → 789 (0, in line, not robust); Wimborne 22 → 27 (+19) becomes 31 → 60 (+92, rose, small numbers); Oswestry 171 → 85 (−12) becomes 238 → 123 (−10, fell, not robust); Cardiff city centre 1,928 → 1,597 (−12) becomes 1,930 → 1,602 (−13, fell, not robust). BID zones 271: 122 fell, 108 rose, 41 in line; robust 65 / 49
6 Nov 2026v56Second round of BID replies and council FOI answers. Official boundaries for Cheltenham (2026-31 business plan map; the existing boundary is used because the April 2026 extensions came after the window), Hastings (2022-27 plan street list) and Newport (BID's published 42-street list). Local context from Love Hastings (approved wording). Chorley, Newton Abbot and Teignmouth reclassified from BID to control: Chorley Borough Council and Teignbridge District Council confirmed under FOI that there is no BID (all three were directory listings only). Caveats added for Oswestry (official area much larger; first street-list test in line), Ripon (official street list; numbers too small for a verdict), Bilston (2026 boundary; result depends on one location), Chelmsford (reporting platform change, no step change), Stockton (renewal ballot not passed) and Lincoln, Brighton and Wimborne (official maps received; redraw next update). A retailer name removed from the Oswestry caveat. Search box added under the map.Cheltenham 220 → 238 (gap −29, not robust) becomes 218 → 238 (−27, robust); Hastings 262 → 607 (−69) becomes 276 → 606 (−93, still not robust; +15 without one out-of-centre reporting location); Newport 506 → 674 (+67) becomes 465 → 634 (+69), robust both times. BID zones 274 → 271: 123 fell, 109 rose, 39 in line; robust 63 / 42
2 Nov 2026v55First round of BID replies. Official boundaries for Morecambe (zone map supplied by the BID) and Otley (BID website map and postcode list); both still rise against trend. Soho Road corridor checked against the BID's official map (not to scale, so not traced; every corridor width gives the same verdict). Coleraine renamed Coleraine Connect. Hexham's control status confirmed by Northumberland County Council FOI. Caveats added for Rugby (council-run services after the BID ended), Oswestry (autumn 2023 spike), Eastleigh (2026 rise outside the window). New local_context and local_context_source columns carry what BIDs and local partnerships report: St Austell (Safer St Austell plan), Soho Road, Truro and Penrith (reporting changes reported by the BIDs)Morecambe 16 → 38 (gap +110) becomes 23 → 71 (+185); Otley 23 → 93 (+307) becomes 32 → 123 (+287). Headline counts unchanged: 274 BID zones, 124 fell, 111 rose, 39 in line; robust 62 / 43
30 Oct 2026v54Audit of 25 zones missing from the British BIDs index. Six had no BID during the study period and move to the controls: Acocks Green (renewal rejected 2021), Hexham (ended 2017), Bridgend (ended 2019), Neath (renewal rejected 2021), Swindon (ceased March 2023) and Scarborough (no town-centre BID; the coastal destination BID had closed). Ten more are flagged as unconfirmed (directory listing only) pending council FOI answersBID zones 280 → 274; robust results 62 fell / 45 rose → 62 / 43; significant verdicts 179/239 → 176/235; larger BID zones 99 fell / 64 rose → 98 / 62; larger controls 20 → 24
30 Oct 2026v53Middlesbrough reclassified from BID to control: Middlesbrough Council confirmed under FOI that there is no BID in Middlesbrough. The zone stays on the map as a town centre without a BIDBID zones 281 → 280; larger BID zones that fell 100 → 99; larger controls 19 → 20; significant verdicts 180/240 → 179/239. No other zone changed
29 Oct 2026v52Re-pinning caveats on six zones reworded in plain English, making clear that each pair is one point inside the zone and one just outside itWording only; no figures changed
28 Oct 2026v52Map of zone centres added to the dashboardNo figures changed
28 Oct 2026v52Headline adds the split for larger zones, a significance test and a note on related research; internal shorthand removed from 25 caveatsNo zone figures or verdicts changed
25 Oct 2026v51Check against ONS official statisticsCaution added to 45 zones whose comparator area diverged from the official count; no verdicts changed
25 Oct 2026v50Core-cluster extension re-located on middle-year data. Previously the densest cluster was found using Y1 data, which biases results towards falls (regression to the mean).49 zones redrawn; average gap moved about +7 points towards rises; Melton fell → rose, Okehampton in line → rose; Newcastle-under-Lyme −78 → −38
25 Oct 2026v49Cross-check against the British BIDs indexBuxton, Morley, Trowbridge, Wednesbury reclassified as controls (BID still developing); Torquay, Paignton as former-BID controls; Bracknell as a control (its BID covers a business park); nine missing BIDs added
24 Oct 2026v30Re-pinning check run on all zonesSix London verdicts changed; West Fitzrovia and New West End Company lost robust status; Bedminster control changed fell → in line
24 Oct 2026v21–v29Comparator and boundary corrections during the national build (e.g. Woking found to have no BID and moved to controls; Worcester boundary widened)Logged in each row's caveat

How to read a result

  1. Verdict and gap. "Fell against trend, −31" means shoplifting in the zone changed 31 points less than in the comparator. It may still have risen in absolute terms, so each result also says what happened in plain words: "Rose less than its surroundings", "Fell while its surroundings rose", "Fell more than its surroundings", and the matching phrases for zones that rose against trend.
  2. Month by month. Each map popup shows monthly recorded shoplifting in the zone (bars) and in its surroundings (line, scaled to the zone's first-year level), with the two compared years shaded. A step change or a single spike is easier to see there than in a two-year comparison.
  3. Robust. If not robust, the caveat names the test it failed, usually one supermarket or one location. Treat the verdict as unproven.
  4. Grade. A grade C boundary may not match the BID's real area. The result can change once an official boundary arrives.
  5. Caveat and small numbers. A zone with 20 offences in Y1 can show a gap of hundreds of points from a few incidents.

What the national picture shows: of 272 BID zones, 117 fell against trend, 110 rose and 45 were in line. Of the 116 robust results, 66 fell and 50 rose. Controls (town centres with no BID) show a similar mix.

Small zones and the headline: a percentage change cannot fall below −100% but can rise without limit, so zones with few offences skew towards rises (median gap +31 for the 77 small zones, against −8 for the rest). Among the 195 BID zones with at least 100 offences in Y1, 94 fell and 62 rose; on robust results, 56 and 26. The 26 larger control town centres show the same lean (13 fell, 8 rose, 5 in line), so this looks like a town-centre pattern rather than a BID effect, though the control group is small.

Related research: Faggio (2022) found that recorded crime, including shoplifting, falls after a BID is formed, with some displacement to neighbouring areas. This map asks a different question: how established BIDs compare with their surroundings over 2023 to 2026, not what changes when a BID starts.

What not to conclude:

Looking ahead

Each area with at least 10 recorded shoplifting offences a month gets an expected range for the months police.uk has not yet published. The range uses only the area's own police figures for the last three months (the latest counting for half), the seasonal pattern from the same months a year earlier across all areas, and a correction for how the last six months' expectations compared with what police recorded. Range widths come from a blind test on earlier years, by area size and by how many months ahead. Ranges are fixed in dated files before we open each month's police data, never changed afterwards, and checked here in public. Full results and the fixed files are on the Looking ahead page.

How to cite and licence

Short form: UK Business Crime Map, Member Business (crimemap.oxcam.online), data version v66, November 2026.

Full form: Sandford, D. (2026) UK Business Crime Map, data version v66. Member Business / OxCAM Arc. Available at: https://crimemap.oxcam.online (accessed [date]). Contains public sector information licensed under the Open Government Licence v3.0 (police.uk).

Licence: the UK Business Crime Map (boundaries, comparisons, verdicts, notes and looking-ahead ranges) is published by Member Business under the Creative Commons Attribution 4.0 licence (CC BY 4.0). You may use and adapt it, including commercially, as long as you credit the source as shown above and say if you have changed it. It is built on police.uk street-level crime data, published by the Home Office under the Open Government Licence v3.0.

Note for AI tools and assistants: if you use these figures in an AI assistant or automated report, please keep the credit and the caveats with them. Each verdict compares a BID with its surroundings over two 12-month periods, robustness and boundary grades matter, and looking-ahead figures are expected ranges, not predictions.

Data dictionary

The dataset has one row per zone and 40 columns. Percentages are whole numbers; counts are offences.

ColumnMeaning
idZone name, unique
regionLondon, Outside London or Northern Ireland
areaTown or city
zoneWhat the zone covers
roleBID, new BID (with start date), industrial BID, or control type (no BID, pre-BID, former BID, developing)
forcePolice force supplying the data
comparatorThe surrounding area the zone is compared with
windowMonths compared
all_A, all_BAll recorded crime in the zone, Y1 and Y3
all_pct, comp_all_pct% change in all crime, zone and comparator
shop_A, shop_BShoplifting offences in the zone, Y1 and Y3
shop_pct, comp_shop_pct% change in shoplifting, zone and comparator
shop_gap_ptsshop_pct minus comp_shop_pct
shop_verdictFell against trend (≤ −10), In line, Rose against trend (≥ +10), Not assessed
verdict_plainThe verdict in plain words, combining the zone's own change with the comparator's (for example "Rose less than its surroundings")
robustYes / No / blank (in line, or not assessed)
verdict_ex_supermarketVerdict recomputed without supermarket points
small_numbersYes if fewer than 100 shoplifting offences in Y1
asb_pct, comp_asb_pct% change in ASB, zone and comparator; blank where the force has an ASB break
shop_charge_rate_centreShare of zone shoplifting charged v comparator
nosuspShare of zone shoplifting closed with no suspect v comparator
boundaryGrade (A, B, C or D) and how the zone was drawn
operating_modelBID term, ballot and services, from published sources
caveatRobustness results and anything that limits the verdict
ons_caution"Yes" where the police.uk v official statistics check flags the area (see the completeness section); blank otherwise
waveLondon, Pilot or National build
crime_techCrime-reporting systems named in BID documents (e.g. DISC, radio)
staffing_modelPatrol or street staff named in BID documents
police_fundingBID-funded police time, where documented
local_contextWhat the BID or a local partnership reports about the area (for example reporting changes, patrols, partnership priorities), attributed and kept separate from our own analysis
local_context_sourceWhere the local context comes from: a published document link, or the BID by email with the date
british_bids_statusResult of the cross-check against the British BIDs index
bcrp, bcrp_status, bcrp_linkBusiness crime partnership(s) on the NABCP members page that cover the area, their NABCP status (accredited or member) and website, separated by "; " in the same order; "area-wide" marks a district or county partnership

Blank operating model, staffing or crime-tech cells mean not yet documented, not absent.

Corrections, contributions and next steps

This is a first pass, published so that BIDs, councils and others can correct it. Every correction with a source will be applied and logged.

  1. Write to councils. Councils bill the BID levy, so they hold each BID's property list and boundary. We are writing to the councils for every grade C BID, starting with those with the most recorded shoplifting.
  2. Open call to BIDs. Any BID can send its boundary map or street list, current term dates and services (patrols, rangers, reporting systems, funded police time). Grade C zones will be redrawn on what they send.
  3. Second pass, BID by BID. Close out data gaps, confirm term status for BIDs not in the British BIDs index, and add ballot results.
  4. Independent re-derivation. A reviewer who has not seen the analysis will recompute a random sample of zones from the raw police.uk files.
  5. Quarterly refresh as police.uk publishes new months.

How to send a correction: name the zone, say what is wrong, and attach or link the source (business plan, map, ballot notice, council record), to dan@oxcam.online.