Perse modelled every half-hourly metered site across a national UK garden centre and retail group, then turned the result into a phased, fundable investment plan. 43 sites, site-by-site, no surveys.
“Having worked with Perse since early 2026, what really stands out is their agility and appetite for continuous improvement—we’ve raised a change and found it tested and deployed within a day. The team is passionate about what their capabilities make possible, and they’ve given us the opportunity to scale how we work with our customers to unlock value from energy efficiency and the energy transition.”
The Client wanted to evaluate solar and battery options across a large half-hourly metered estate. Every site has a different load shape, tariff and roof, so the right size of system differs at each one — and committing to the whole estate at once was never realistic. They needed evidence for what to build where, and in what order.
On the Perse platform each site was run through 25 solar and battery scenarios, priced against its current tariff and three alternatives including time-of-use pass-through contracts. That gives the best-suited solar, battery and tariff combination for every site, and payback modelled at site and portfolio level — which becomes a phased, payback-ranked plan rather than an all-or-nothing recommendation.
Every site is assessed from its half-hourly data and roof geometry, so the estate can be screened and ranked before committing survey time. Detailed site visits go only to the sites the modelling says are worth them.
43 sites × 25 scenarios × four tariffs is over 4,000 modelled outcomes. On the platform that runs in hours, not the months a consultancy exercise of this size would take — and it can be re-run whenever tariffs or install costs move.
Phase 1 and Phase 2 together deliver 81% of the total annual saving for 72% of the capital — the core of the investment case.
Payback varied by a factor of four across the estate, which is exactly why the capital is released in phases. Site identities are withheld; figures are aggregated.
Each rail spans the payback range for that group; the marker is the group average.