Tenant Satisfaction Measures (TSMs) are the 22 indicators that every English social landlord must report annually to the Regulator of Social Housing. Twelve come from a direct survey of tenants; ten come from your own operational records. They cover satisfaction with the home, repairs, safety, complaints handling and neighbourhood; and the underlying gas, fire, asbestos, water, lift and Decent Homes compliance figures that sit underneath those tenant perceptions. This guide explains the regime as it stands in May 2026, walks through the methodology choices that disproportionately shape what your published score looks like, and surfaces the pitfalls that have led the Regulator to publicly exclude named landlords from its statistics.
We have written this with research and insights managers in mind — the people who will defend a methodology decision to a board. Where the Regulator’s own wording matters, we quote it verbatim. Where there is no authoritative answer published, we say so and explain what practitioner consensus looks like instead.
1. The TSM regime in 2026 — what changed and what’s next
The TSM framework came into force on 1 April 2023, with the first full reporting year covering 2023/24. As of May 2026, two annual statistical releases have been published by the Regulator of Social Housing (RSH) — the most recent on 4 November 2025 (2024/25 headline report).
Three updates to watch right now:
- Awaab’s Law and RP02.On 3 March 2026 RSH updated the Technical Requirements document to clarify how Awaab’s Law hazard categorisation must be applied to the responsive repairs TSM. Emergency-hazard defects must be treated as emergency responsive repairs; significant hazards default to non-emergency. If your repairs reporting still uses pre-March 2026 categorisation, your published RP02 figure is now stale.
- Proposed electrical safety TSM. Driven by the 2025 extension of EICR rules into the social rented sector, RSH has proposed adding electrical safety as a new building-safety measure. First reporting year proposed: 2026/27. Start tracking now.
- Consumer Standards consultation closed 3 March 2026. RSH ran a consultation on the updated Transparency, Influence and Accountability Standard, the Consumer Standards Code of Practice, and the TSMs themselves. The outcome was not yet published at the time of writing. Revised standards are intended to apply from 1 October 2026.
Who is in scope
Two thresholds matter. Registered providers owning 1,000 or more units of social housing (combined Low Cost Rental Accommodation and Low Cost Home Ownership) submit all 22 TSMs to RSH annually and publish them on their own website. Providers with fewer than 1,000 units must still generate and publish TSMs but are subject to proportionate requirements: a perception survey at least once every two years, a high-level representativeness check rather than formal weighting, and a census approach is acceptable for very small stocks.
2. The 22 measures
Twelve tenant perception measures (TP01–TP12)
These are sourced from a direct survey of tenants. The Regulator publishes the exact wording in Tenant Survey Requirements (March 2023, updated April 2024). Verbatim wording is not optional — see Section 6 below for what happens when providers materially alter it. The questions cover:
- TP01 — Overall satisfaction with the service provided by the landlord
- TP02, TP03 — Satisfaction with repairs and with repair timeliness (TP02 must precede TP03 in the questionnaire)
- TP04, TP05 — Home well-maintained and home safe
- TP06, TP07, TP08 — Listens and acts, keeps informed, treats fairly and with respect (TP08 is the only agreement-scale item rather than satisfaction-scale)
- TP09 — Complaints handling (asked only of tenants who complained in the past 12 months)
- TP10 — Communal areas (asked only of tenants who have them)
- TP11, TP12 — Neighbourhood contribution and anti-social behaviour handling
One rule that catches a surprising number of providers: “The question to generate overall satisfaction (TP01) must appear as the first question in any perception survey questionnaire used to generate TSMs.” No warm-up questions, no demographic preamble, no “rate your most recent service interaction” lead-in.
Ten management information measures
These come from your own records, not the survey: five building safety measures (BS01–BS05: gas, fire, asbestos, legionella, communal lifts), two repairs measures (RP01 non-Decent Homes proportion; RP02 responsive repairs within target), two complaints measures (CH01 complaint volume per 1,000 homes; CH02 within Ombudsman Complaint Handling Code timescales), and NM01 covering anti-social behaviour case volume. The Regulator is publicly direct that BS01–BS05 are a regulatory trigger, not a benchmark.
“You should be hitting 100% compliance on all of your safety indicators. If you’re not and you’re missing them by quite a margin, we will come and ask… because that is one of your fundamental responsibilities — keeping your tenants absolutely safe.” — Will Perry, Director, RSH
3. Methodology — the choice that disproportionately shapes your score
The 2024/25 sector statistics show how much the survey method matters. Average overall (TP01) satisfaction by method, sector-wide:
| Survey mode | Share of all responses | Average TP01 satisfaction |
|---|---|---|
| Postal | 8% | 80.2% |
| Telephone (human) | 63% | 69.4% |
| Face-to-face | 11% | ~70% |
| Online / email | 16% | 60.1% |
| SMS | 2% | — |
Postal returns the highest published satisfaction — twenty percentage points above an online-only sample for the same landlord population. That is not because postal respondents are objectively more satisfied; it is because modes differ in who replies, in interviewer effects, and in self-selection. The Regulator says so directly: “Survey collection methods where staff or contractors ask questions… can generate higher satisfaction scores than other methods.”
The honest pros and cons of each mode:
Postal
Highest published satisfaction figures and the strongest inclusion of digitally excluded tenants. Operationally the slowest mode (8–12 weeks typical, with two reminders). The cost-per-response is materially higher than commonly assumed once print, freepost, reminders and data entry are loaded in. The strongest risk is acquiescence bias — tenants tend to agree with a paper questionnaire that arrives on the landlord’s letterhead.
Telephone (human interviewer)
The dominant mode in the sector (63% of responses) and the most expensive on a per-completed-interview basis once reasonable quotas are met. Strong reach across age bands when call patterns are managed. The classic risk is the interviewer effect — respondents overstate satisfaction when speaking to a polite voice in real time.
Online and email
The cheapest mode by unit cost and the fastest cycle (2–4 weeks). Also the mode with the largest published downward bias on satisfaction — used alone, it can make an otherwise comparable landlord look 15–20 points worse than a postal-only peer. Heavy under-representation of older, lower-income and digitally excluded tenants (particularly sheltered and supported housing populations) is the structural problem.
AI-phone (automated voice / conversational AI)
Fits within the Regulator’s telephone category and within the MRS Code of Conduct when fieldwork is transparent, question wording is identical to the TSM specification, and the methodology is documented in the summary of approach. Reduces the interviewer-effect bias that human phone introduces. The Regulator has not published explicit AI-phone guidance as of May 2026, so providers using AI fieldwork should be especially careful with documented rationale and mode-mix reporting. The sector has no published TSM-specific response-rate benchmark for AI-phone yet.
Mixed-mode (push-to-web + postal + phone follow-up)
The Regulator and sector benchmarkers now describe mixed-mode as the methodological gold standard for representativeness. Each mode catches different cohorts. The operational cost is the inclusion of a postal sweep; the analytical cost is that mode effects within the same dataset must be acknowledged and (where material) weighted out.
4. Sample size, response rate, and weighting
There is no minimum response rate. The Regulator anchors statistical accuracy on margin of error against your population, not on response percentage:
| Stock size (LCRA or LCHO, calculated separately) | Required margin of error at 95% confidence |
|---|---|
| Under 2,500 units | ±5% |
| 2,500–9,999 units | ±4% |
| 10,000–24,999 units | ±3% |
| 25,000+ units | ±2% |
In practical terms a provider with 10,000 LCRA and 2,500 LCHO units needs roughly 965 LCRA and 485 LCHO completed responses to hit ±3% and ±4% respectively. Smaller providers (under 1,000 units) can satisfy the requirement with a census approach.
Weighting is required when achieved sample materially under- or over-represents tenant groups likely to have different satisfaction levels. The Regulator names six representativeness characteristics: stock type (general needs / sheltered / supported), age, ethnicity, building type, property or household size, and geographic area. Households are capped at one response regardless of occupants, and partial responses count for the questions that were answered.
5. Reporting and publication
The reporting year is 1 April to 31 March. Large registered providers submit annual returns to RSH via NROSH+, typically in late June (the exact window is set each year in the CEO data-requirements letter). Sector-level statistics are published by the Regulator in late autumn — 2024/25 came out 4 November 2025; 2023/24 came out 26 November 2024.
On your own website you must publish, alongside the 22 TSMs, a summary of approach covering: achieved sample size, survey timing, collection method(s), sampling method, representativeness assessment, any weighting applied, named external contractors, households excluded from the sample frame with reasons, reasons for not hitting minimum samples, any incentives used, and methodological issues that had material impact. The Regulator does not mandate a tenant-facing template — the format is up to you, provided every required element is there. The Regulator also does not publish tenant-facing TSM data on its own portal beyond the annual statistical release.
6. The five pitfalls that lead to data being excluded
The Regulator named two landlords in the 2024/25 release and excluded their perception data because “perception surveys… were worded in ways that were materially different from the TSM requirements.” The named landlords were Stoke-on-Trent City Council and Cannock Chase District Council. Several councils had management-information measures removed for inability to provide quality-assurance evidence in time for publication.
The five recurrent mistakes:
- Materially altered question wording.The regulator’s March 2023 Tenant Survey Requirements document is the verbatim source. Departures from that wording — even well-intentioned ones — risk exclusion.
- Incomplete summary of approach. Failing to publish either the questionnaire actually used or the full summary-of-approach document.
- Variable representativeness assessments. Under-sampling sheltered, supported, or ethnic-minority cohorts and not weighting to compensate.
- Management information gaps.Particularly for the building safety BS01–BS05 measures, which are the Regulator’s declared trigger for further questioning.
- Method-changing to flatter scores. Switching to a higher-scoring mode mid-cycle without documenting the rationale. The Regulator and Housemark have publicly stated they will see through this:
“You’re not going to fool the regulator by changing collection method.” — Jonathan Cox, Chief Data Officer, Housemark
7. What to look for in a survey provider
There is no regulatory requirement to use a third party; providers may run TSM surveys in-house or outsource. Where outsourced, fieldwork must follow the MRS Code of Conduct and accuracy remains the board’s responsibility. When evaluating a provider — including SignalLine — the questions worth asking are:
- Wording fidelity. Do you use the verbatim TP01–TP12 wording, including the TP01-first and TP02-before-TP03 ordering rules?
- Mode mix transparency. Do you report per-mode response counts and per-mode satisfaction in your hand-back to the board, so we can publish a defensible summary of approach?
- Representativeness. Against which characteristics do you assess representativeness, and what weighting approach do you apply when achieved sample skews?
- MRS Code compliance and data handling. How are household-cap and one-response-per-household enforced, and where does data live?
- Independence from operations. Where does fieldwork sit organisationally? Customer Services ownership creates a perceived conflict; insights or research ownership is cleaner from a regulator perspective.
Frequently asked questions
Is there a minimum response rate?
No. RSH does not set a response-rate floor; the bands above are anchored on margin of error against the tenant population. Anyone telling you “30% is the minimum” is overstating the regulator’s text.
Are AI-phone surveys an approved TSM method?
They fit within the Regulator’s telephone category and the MRS Code, provided question wording is identical to the TSM specification and methodology is documented. The Regulator has not published explicit AI-phone guidance as of May 2026, so document your rationale and report mode mix clearly.
Can I change survey method between years to improve scores?
You can — but document the rationale, and expect to be asked. The Regulator and Housemark have publicly stated they will see through method-driven score changes.
What is the publication deadline?
Reporting year ends 31 March. Large providers submit annual returns to RSH typically in late June; the sector-level statistical release is published in late autumn (4 November 2025 for 2024/25).
Sources and further reading
- TSM survey methodology: postal vs phone vs online vs AI — how to choose — companion guide. Decision-support for the methodology choice that drives a 20-point gap in your published score. Cost-per-response components, mode-effect mechanics, decision framework by stock size.
- How to run a TSM perception survey programme — a 12-week operational guide — operational walkthrough. Week-by-week timeline from board approval to NROSH+ submission, decision gates, and the anti-patterns that compress your analysis window.
- Tenant Satisfaction Measures: Tenant Survey Requirements — the verbatim source for question wording, response scales, and ordering rules (Regulator of Social Housing, March 2023, updated April 2024).
- Tenant Satisfaction Measures: Technical Requirements — sample size accuracy bands, weighting rules, management-information measure definitions (updated 3 March 2026 for Awaab’s Law).
- TSM 2024/25 Headline Report — the most recent sector-level breakdown by mode and measure (4 November 2025).
- RSH TSM collection page — landing page for all current and prior publications.