What stayed the same

The big picture has not moved. Weddings still dominate the top of every table. Private parties still sit 11% below weddings for the same lineup. Pubs still anchor the floor. And the city hierarchy (Edinburgh at the top, Newcastle at the bottom) held for the second consecutive month.

One caveat before the tables: May’s levels are not directly comparable to Issue 01 — this issue’s rebuild moved every rate-card fee onto a consistent take-home basis, which re-based headline medians roughly a quarter lower. The structure is what carried through, and that is the useful signal: the same gig types pay most, the same cities lead, the same band-size curve applies. Here is where each segment landed:

Wedding and private-party cells are deep: thousands of observations each. Corporate, festival and theatre cells are still thin (single digits). Directionally useful, but treat them as early signals rather than settled medians. They will firm up issue by issue as the dataset grows.

What changed in the numbers

Two shifts worth noting: one in volume, one in the city league.

Data volume: 2,381 → 3,847 (+62%)

April’s Issue 01 launched the Index with 2,381 observations from 2 rate-card feeds. May nearly doubled that — mostly by widening category coverage across the two existing feeds, with a third source adding its first 73 rows on top. The extra data mostly landed in the wedding and private party cells: the segments where the new source lists the most acts. That is why the observation counts on those cells jumped from the low thousands to 6,929 and 3,242 respectively.

You will notice the cell count actually dropped from 345 to 313 and the city count from 14 to 13. That is not a regression. It is a quality tightening. The May refresh raised the bar on how much data a cell needs before it publishes. Fewer cells, but each one is backed by more observations and is more trustworthy. The cities that dropped were ones where the data was too thin to publish with confidence; those now fall back to the regional or UK figure.

City league: Edinburgh still on top

Edinburgh held the top spot at £1,186 median for a 3–4 piece wedding band. Brighton climbed to second (£1,050), overtaking Glasgow. London’s median (£903) sits mid-table but its p25-to-p90 range is the widest of any city we publish: everything from registry-office solos to Mayfair black-tie galas.

Scotland punches well above its weight on wedding fees: Edinburgh and Glasgow both sit above London. That might surprise people. The Scottish wedding market is smaller but high-end, and the premium wedding venues (Gleneagles, Prestonfield, Archerfield) set a floor that lifts the whole city median. Scotland’s top 2 cities both clear the London median, and we publish 13 cities in all.

What changed in the methodology

This is where the real work happened in May. Three infrastructure changes that do not change today’s numbers much, but make the whole system more trustworthy going forward.

1. Third rate-card source

April drew rate-card data from 2 published sources. May added a third, bringing wider category coverage: particularly across function bands, party bands and tribute acts. The first batch from the new feed produced 73 new observations at an average net fee of £1,312, landing squarely in the wedding and function-band segments.

Why does a third source matter? Not because of the extra 73 rows. It matters because it diversifies the data mix. When only two sources feed the index, any pricing quirk in either one has an outsized effect on the median. A third source dilutes that.

2. The double-haircut fix

This was a proper bug and I want to be transparent about it.

Published rate cards quote gross fees: what the client pays, including commission. The Index converts these to net fees (what the artist actually takes home) by applying a documented 20% haircut when the data enters the Index.

The problem: a code change in late April accidentally introduced a second 20% haircut further along the process. So rate-card fees were being reduced by 20% twice: once correctly on entry, once incorrectly downstream. The result was net amounts being under-reported by about 20%.

The good news: the blast radius was small. Only 11 rows (entered between 27 April and 4 May) were affected. The 2,646 rows from the initial seed were already correct. We caught it during the May reliability sprint, patched the issue, and corrected all 11 rows. If you pulled data from the Index between those dates, the affected cells would have been slightly low, but the error was within the noise for most use cases.

3. Re-sighting mechanism restored

When the weekly refresh re-checks a listing that already exists in the dataset, it should increment that listing’s observation count rather than creating a duplicate row. This lets us track which listings persist month-to-month (a signal of price stability) without inflating the sample size.

That mechanism had been silently broken since launch. A code change overwrote the version that handled deduplication. May restored it. The practical effect: re-sighted listings now track properly, and the n counts you see in the tables above are genuine unique observations, not double-counted refreshes.

How the data mix looks now

Transparency is the whole point of the Index. Here is exactly what fed the May numbers:

Rate cards dominate by raw count (85%), but each one only carries 0.48 of a full vote. A single confirmed booking (weight 1.00) outweighs two rate-card listings. As real GigXchange bookings start closing in volume (expected Q3 2026), the weighted picture will shift meaningfully toward transacted fees rather than listed asking prices. The May Index spans 13 cities and 12 regions over the past 24 months, with rate cards refreshed every 7 days, prices weighted on a 12 month half-life, and every cell benchmarked against the Musicians’ Union floor of £167.16 per musician; a listing seen nine times counts for far less than 9 gigs.

The honest gap right now: no confirmed bookings and no post-event submissions have entered the index yet. That means every published cell is built from asking prices (rate cards, profiles), recommended rates (MU/Equity) and venue budgets, not from money that has actually changed hands. The rates are directionally right, but they will get more precise once booking data starts flowing.

The band-size cost curve

One of the most useful outputs of the Index is the relationship between band size and fee. More members does not mean proportionally more money. The uplift flattens out.

The steepest jump is duo to 3–4 piece: you cross the threshold from background music to a headline act with a rhythm section, and the market prices that shift. Beyond five musicians, the incremental cost per head drops because logistics (PA, transport, stage plot) are already covered. Stepping from a 2 piece to a 4 piece roughly doubles the fee; a 4 piece to a 5 piece adds far less.

What we honestly cannot tell you yet

I would rather be upfront about the gaps than pretend the data is complete.

• Confirmed booking data: zero rows so far. The gold-standard weight-1.0 source has not yet produced any volume. Once it does, every cell will get more precise.
• Festival and theatre cells: fewer than 10 observations each. The medians are directionally right but could shift significantly with a handful of new data points.
• Corporate bookings: only 8 observations for 3–4 piece. Entertainment budgets are rarely public. This cell will take longer to mature.
• Genre-specific rates: every observation is tagged with a genre, but we do not yet have enough volume to publish separate folk vs. rock vs. soul vs. jazz cells. That is coming. The data structure is ready.
• Scotland beyond weddings: Edinburgh and Glasgow are well-covered for wedding fees but thin on pub and corporate data. Those cells still fall back to regional figures.
• Rate-card dominance: 85% of observations come from published rate cards. That is a known limitation: rate cards represent asking prices, not clearing prices. The weighting system is designed to correct for this as higher-weight sources (bookings, verified submissions) start producing volume.

What comes next

Three things are on the roadmap for Issue 03 (June 2026):

• Club fee model goes live. Club gigs do not have a flat fee: they are split doors or a percentage of ticket revenue. We have designed a 4-archetype economic model that derives club fees from ticket price, venue capacity, occupancy and artist share. It produces a blended median of £296 per act, aligned with grassroots reality, not the £600+ that union rate cards suggest. This will enter the Index in shadow mode (computed but not published) until we have enough confirmed bookings to calibrate it.
• Venue capacity data. The ticket-heuristic model needs to know how many people each venue holds. We are building a capacity dataset that will be independently useful for the gig directory and open-mic listings. Expected delivery: early June.
• First confirmed-booking observations. Completed bookings on the platform now feed the Index automatically. Once bookings start closing at volume, weight-1.0 rows will start reshaping the percentiles. The integration is built and tested. It is waiting on transaction volume, not code.

Issue 03 is now published: read the June 2026 write-up or the full June report, including an honest scorecard on the three promises above. The live index updates nightly. You do not need to wait for the monthly report to see current data.

How to contribute

Every gig you have played is an observation we need. The Index gets more accurate with more data, especially from sources beyond published rate cards.

Head to /rates and submit three gig fees anonymously. No email, no signup, no account required. Submit 3 fees in under 2 minutes and you unlock every cell, every trend line, and every city drilldown for 30 days, while we re-poll the same listings every 7 days. Three observations from you makes the whole system better for everyone.

If you are a venue booker, music journalist, or academic researcher: the Index is free to cite and free to build on under CC BY 4.0. Attribution: “GigXchange Index, gigxchange.app/rates”.

Annual refresh commitment

This guide was published on 22 May 2026 and is refreshed every May. UK live music rate data moves monthly, so annual verification matters. We re-verify every reference, recommendation, and data point once a year. Next scheduled refresh: May 2027. If any claim is outdated before then, email support@gigxchange.app and we aim to update it within one working day.