Ground-floor retail is required by policy and often unlettable

Both approaches to ground floor uses work. What differs is what they cost you, and the cost is what this sets out.

The policy is right and the application often is not

Requiring active uses at ground level exists because blank frontage measurably deadens streets, which is well established. The requirement is frequently interpreted as retail specifically, on streets that have no catchment to support a shop. Retail demand is concentrated on a small number of pitches within any centre and falls away sharply beyond them.

Mandating shops beyond that boundary produces units that never let, which achieves the opposite of the policy intent.

Vacancy is worse than the alternative

An empty unit with a shuttered front is a longer stretch of dead frontage than a well-designed residential or workspace ground floor would be. It also signals decline, which affects the letting prospects of the units either side of it. Long-term vacancy attracts meanwhile uses, hoardings and vinyl graphics that satisfy nobody.

Measured properly, the comparison that matters is between a likely-empty shop and a likely-occupied alternative, not between a shop and a blank wall.

Structural requirements make retail expensive to provide

Retail floor-to-floor heights are greater than residential, servicing needs are different and column spacing has to suit an open plan. Providing that on the ground floor of a residential building adds cost that is recovered from the flats above. Where the retail then does not let, that cost delivered nothing, which is a straightforward waste.

Measured properly, conversely, a ground floor built to residential dimensions cannot become retail later, which forecloses the option permanently.

The useful requirement is activity, not retail

Workspace, nurseries, clinics, community rooms, gyms, studios and residential entrances all produce frontage activity and comings and goings. Design codes that specify entrance frequency, transparency and floor-to-ceiling height achieve the frontage objective without dictating the use class. This gives the building a chance of being occupied continuously through changes in the retail market.

It also allows the same space to become a shop later if the catchment develops.

Online retail changed the arithmetic permanently

Retail floorspace demand has fallen in many markets as sales shifted online, and the shift accelerated in the early 2020s. The remaining demand concentrates further into prime locations, which widens the zone where mandated retail cannot let. Convenience, food, personal services and experiences have held up better than comparison goods retail, which affects what kind of unit is viable.

Measured properly, policy written before these shifts, and not revised since, is producing units for a market that has contracted.

What works in a dense grid may not transfer to a low-density suburb.

Flexibility should be designed in

A ground floor with generous height, a level threshold, provision for separate servicing and a facade that can be adapted keeps its options open. The additional cost of that flexibility is modest compared with the cost of a permanently vacant or permanently unconvertible space. Where planning use class systems allow changes without a fresh application, this flexibility is realisable; where they do not, it is theoretical.

Measured properly, checking what change of use is permitted locally is therefore part of the design decision.

Side by side

The takeaway

Ask what the street can support before asking what the policy requires.

Cities are built by a thousand small permissions, not one big plan.