A season that reorganized the audience and lifted the front door.

Through May 14, 2026, the Society sold 2,240 tickets across the seven mainstage concerts in Susan Andrews’s official tally, an increase of five percent over last season. The larger movements sit below.

Three measurements describe the 2025/26 season more accurately than ticket count alone. The first is that subscriber participation grew from 433 ticket holders in 2024/25 to 706 in 2025/26, a sixty‑three percent increase, while membership-tier ticket holders declined by nineteen percent. The funnel did not simply expand; it consolidated. Patrons who had previously bought single tickets or held member-tier seats moved into season subscriptions, and total seats sold rose modestly while the share of those seats committed to the whole season rose considerably.

The second is that the website received 33,015 distinct users between September 1, 2025 and May 14, 2026, sixty‑four percent more than the previous year and a step change against 2023/24, which the previous year had run effectively flat against. User growth outpaced session growth by sixteen points, which is how an audience widens rather than how an existing audience repeats. The Society’s digital front door brought new humans through.

The third is the money that arrived alongside ticket revenue and that the dashboard above this report does not usually surface: $577,694 of non‑ticket revenue across three seasons in Susan Andrews’s May 14 workbook, with five of the ten source tabs still unpopulated. The figure is a floor, not a ceiling, and the floor itself is a meaningful share of the Society’s sustaining revenue.

What follows is a record of those movements against the contracted scope of the 2025/26 Marketing Strategy SOW and the Fractional CMO Partnership Proposal. It is also a record of what the available data cannot answer, which is a longer list than the cleaned chart-friendly numbers would suggest.

I.The season, concert by concert

Seven mainstage concerts sold 2,240 tickets. The finale, Carnival of the Animals on April 23, sold out at 252; Charm and Reflexion drew the heaviest single-ticket sales of the season.

The Society’s 2025/26 mainstage runs from November through April. Susan Andrews’s SUMMARY tab in the May 13 ticket workbook records the canonical category counts for the seven mainstage concerts, including the sold-out Carnival of the Animals finale on April 23. Three additional events sit outside that tab and are noted separately below.

Read row by row, the seven concerts hold a consistent subscriber base of between 89 and 104 each, contributing roughly thirty percent of every concert’s seats. Single-ticket volume varies more widely: 128 for Charm and Reflexion in February at the high end, 46 for Carnival of the Animals at the low (the finale was sold out weeks ahead, so single sales effectively closed before late buyers could reach the door). Comps cluster heavily at the season opener: 61 of the 355 Joyful Beethoven seats were complimentary, an order of magnitude above the rest of the season.

II.Three seasons compared

Subscribers rose sixty-three percent from last season; members fell nineteen. The funnel did not just grow, it consolidated.

The SUMMARY tab carries the same row structure across all three seasons, seven mainstage concerts each. (One 23/24 row is mislabeled in the source as 23-23 #4, a typo flagged in the 2026-05-14 reporting boundary memo; the build script normalizes it back into the 23/24 column so totals balance to Susan’s own SUM row.)

Subscribers rose from 433 in 2024/25 to 706 in 2025/26. Members held at 673 against 830 the year prior, a nineteen percent decline. Single ticket purchases declined modestly; comps grew sharply against a low 2024/25 base; Young Friends declined to 68 from 80. The aggregate of 2,240 represents a five percent gain over the prior season’s 2,136. The verbal estimate Brandon gave Lisa on May 1, "thirty-three percent new subscribers, twenty percent premium tier growth," understates the actual subscriber-ticket movement and does not name the offsetting member decline. The dashboard above reports the larger and more accurate numbers.

III.The audience that showed up online

Sessions up forty-eight percent. Users up sixty-four. The current season is a step change against a flat 2024/25.

The Society’s Google Analytics 4 property (301992134) registered 42,383 sessions and 33,015 users between September 1, 2025 and May 14, 2026, against the same windows in the two prior seasons:

2024/25 ran effectively flat against 2023/24. The current season is a step change. The +64% gain in users against +48% sessions indicates that the lift came largely from new humans reaching the site rather than from heavier behavior by the same returning audience.

Most of that traffic arrived on mobile. Of the 25/26 season’s 42,383 sessions, about fifty‑six percent were on a phone, with the remainder split between desktop and a thin sliver of tablet. The website’s mobile presentation is the presentation the largest share of the audience encountered, a point Lisa raised at the March 31 standup and the data corroborates.

Channel attribution on the main property is plain enough to recite in a sentence. Direct traffic leads, followed by organic search, referral, and email. A large share of direct traffic, alongside a meaningful share from organic search, is the signature of an established brand whose audience reaches the site by typing the URL, returning to a bookmark, or finding the Society through a search query, rather than through paid acquisition.

A different question, whether those website visits converted into ticket purchases, the data does not cleanly answer. The ThunderTix purchase platform runs inside an iframe embedded on the CMSPB site. GA4 cannot reliably trace the source of a ticket purchase across that boundary. The ThunderTix GA4 property (363702623) registered over a thousand sessions and eleven thousand events of its own through the season, evidence of real interaction within the iframe but not a chain of attribution back to the originating channel. Closing that gap is among the items proposed in the Strategic Intelligence Sprint described in section seven.

IV.The inbox

Across three seasons the email program has held a thirty‑five percent unique open rate, above the published nonprofit benchmark band. Unsubscribes sit under one in a thousand sent. Bounce rate has climbed from six percent to nearly twelve percent, a list‑hygiene story the Sprint takes up directly.

The email program is the channel where CMSPB owns its own audience. The list reaches the inbox without paid placement and without a platform’s algorithmic interference. The three‑season picture, drawn from the Constant Contact v3 reporting API on May 15, 2026, is in the table below. Sends are the number of messages successfully delivered; unique opens count one open per recipient regardless of how many times they opened; unique clicks similarly count one click per recipient.

Three observations from the table. First, the unique open rate has held in the 34–35% band across all three seasons. The Constant Contact published benchmarks for the Nonprofit Services and Arts/Entertainment sectors sit between twenty‑two and twenty‑nine percent. The CMSPB list is reading above its sector. Second, the unsubscribe rate has held at around one‑tenth of one percent. The list is not fatiguing; recipients want to hear from the Society. Third, the bounce rate has moved from six percent in 2023/24 to nearly twelve percent in 2025/26. Bounces are a list‑hygiene measurement, not a campaign one: it tells us the share of stored addresses that are no longer deliverable. Removing inactive and undeliverable contacts is one of the standard outputs of the Sprint’s Universal Customer Record protocol.

The click‑through behavior of the list is the more interesting story underneath the headline. Across all 2025/26 sends, the click‑to‑open rate (the share of recipients who opened the email and then clicked a link inside) sat at two‑point‑four percent; the click‑through rate (clicks against sends) at under one percent. The signal those two numbers carry together is that the audience is reading rather than transacting through the email. A large share of opens land on informational content (announcements, reminders, newsletters) where the recipient does not need to click to receive value. Where the call to action is sharp and targeted, the click rates rise substantially, which the next subsection demonstrates.

Segmentation and targeted sends — where engagement concentrates

The campaigns sent to a narrowly defined segment of the list, primarily current members and subscribers, produced the highest engagement of the season. The Constant Contact API reports unique open rates by individual campaign; the table below lists the six 2025/26 campaigns with the highest unique open rate among campaigns delivered to a hundred or more recipients.

The pattern is consistent. When the audience definition tightens, the open rate moves from the mid‑thirties to the mid‑sixties. The membership and presale communications around Humor & Harmony, which were sent only to current members and subscribers, opened at over sixty percent on every send. The Membership Perks email opened at sixty‑six percent. The Members‑only Humor & Harmony presale announcement opened at sixty‑four percent. The same dynamic appears in the click‑to‑open behavior of the Artist Circle dinner invitations, which were sent to roughly fifty recipients each and produced click‑to‑open rates between twenty‑seven and forty‑five percent. Segmentation works on this list.

Translating segmentation effects into ticket sales attribution requires the cross‑system join the Sprint delivers. The current Constant Contact reporting does not connect a send to a downstream ticket purchase. The Sprint’s reconciled patron foundation closes that loop by joining campaign sends to ticket and donation records on a canonical patron identity, which is the technical work that converts the engagement numbers in this section into measurable revenue attribution.

By category

The list also rewards different content categories differently. Concert reminders to the full list (sent in the days leading up to a performance) and short‑form communications around special events outperform the recurring monthly newsletter on a per‑send basis. The table below aggregates the 2025/26 sends into seven categories derived from the campaign names.

The categories with the highest open rates are the ones most directly tied to a specific event in a specific week. The lower open rates on Membership and Subscription content reflect a broader audience definition: these campaigns reach the full list, not the member‑only segment, and so include recipients for whom the call to renew or upgrade is not immediately actionable. This is the data Lisa asked for on May 1: a measurable answer to "what did the digital campaign yield" that does not depend on Susan’s ticket files or on iframe attribution. The yield is in the inbox, and it is documented.

V.Money beyond the box office

A $22,696 surplus on the media budget. $577,694 in non-ticket revenue across three seasons, of which thirty-one percent of 2024/25 arrived in the previous fiscal year.

The 2025/26 marketing budget was set at $60,000 and executed at $37,304. The unspent $22,696 is in part the result of a discipline applied at the start of the season that removed roughly $24,900 in low-attribution lines from the prior year’s budget (the WLRN and Legends 100.3 radio packages, a $12,000 PR retainer, the PB Illustrated advertorials, and the season postcard). The same restructured budget absorbed three additional campaigns mid-season at no incremental cost: the Rising Artists applications campaign, the Rising Artists concert promotion, and the Humor & Harmony rollout.

The non-ticket workbook

Susan Andrews delivered a second workbook on May 14, the non-ticket revenue export committed at the May 12 Data Reporting Strategy Meeting. Across five populated tabs the workbook records $577,694 in revenue attached to a fund and an intended season:

Five additional tabs remain unpopulated in the source: 24/25 Concert Sponsor, 24/25 Government Grants, 24/25 Annual Fund (labeled "Annual Funf" in the file, sic), 25 Fund a Need, and 25 Gala Donations. Until those rows are filled, this report cannot speak to gala fundraising revenue, concert sponsorship revenue, or grant revenue. The build script re-reads the workbook automatically when new tabs are populated.

A note on the fiscal year

CMSPB’s fiscal year runs from July 1 through June 30. Patrons routinely buy memberships and subscriptions for the next season during the final weeks of the current fiscal year, a pattern Brandon described to Susan on May 12 and that the data confirms. Of the $448,636 attributed in Susan’s tab structure to the 2024/25 season, $139,903 (thirty‑one percent) was actually transacted in calendar fiscal year 2023/24. Trend analysis should use Susan’s intended-season attribution, which groups revenue by the season the patron was paying for; accounting and audit close should use the calendar fiscal year, which groups by the date the check arrived. The two views are not in conflict, but they are not interchangeable.

VI.What the numbers do not capture

Two structural facts sit behind every figure above: capped attendance from emotional mid-flight ad cancellations, and four years of unbooked strategic work that built the funnel the dashboard headlines.

This report is an account of outputs delivered against the contracted scope. It is not a complete account of the value those outputs represent, and it is necessary to say why in writing.

A standing constraint on paid attendance

Across multiple recent seasons, paid social campaigns have been canceled mid‑flight at the direction of CMSPB leadership, against the marketing recommendation, on grounds that were emotional rather than evidentiary. Several concerts have ended with paid attendance in the sixty‑to‑seventy percent of capacity band, a fact available to CMSPB’s box office staff through the ThunderTix barcode‑scan attendance reports that Susan herself describes as among the more reliable sources the organization holds. The ticket counts in section one are the counts the Society sold. They are not the counts the marketing program, allowed to run uninterrupted, was reasonably capable of producing. That structural cap sits invisibly behind every figure above.

Work performed outside contracted scope

The 2025/26 SOW contracts for execution work across email, paid social, and print, alongside creative oversight of the season’s recurring fundraising and seasonal initiatives. The contract does not capture four years of brand stewardship, channel discipline, and email-program design that produced the audience figures in section three; the mid-season scope absorption of the Rising Artists campaigns and Humor & Harmony at no incremental cost; the emergency vendor sourcing and donor-relationship triage that have occupied a portion of every season; the system design that eliminated $24,900 of low-attribution media spend at the start of 2025/26 without loss of performance. The +64% YoY users figure that the dashboard headlines is as much the cumulative result of four seasons of unbooked strategic work as it is the result of paid activity during 2025/26 specifically.

The market for the work

The professional-services market in which this work was contracted is structurally different from a vendor market. Specialists in marketing strategy, creative direction, and applied AI who choose to discount their rates for arts-and-culture clients do so on the basis of mission alignment, not pricing strategy. The same specialists serve private-sector and government clients at full rate. The discount is therefore a subsidy from the contractor’s other margins to the Society’s budget, paid in exchange for an expectation of strategic partnership and reciprocal investment in the contractor’s recommendations.

The published research on the dynamic is consistent across multiple sources. Stanford Graduate School of Business’s 2015 survey of 924 nonprofit directors found that sixty-nine percent of directors report their organizations have faced one or more serious governance-related problems in the prior decade, more than a quarter of directors lack a deep understanding of their organization’s mission and strategy, and a majority do not believe their fellow board members are sufficiently engaged.¹ Mission Edge, surveying the broader sector, finds nonprofit-sector turnover sits at roughly nineteen percent against twelve percent in other industries, and that burnout in nonprofits is almost never a problem with individual employees but a signal that something in the organization’s operating logic needs to change.²

Harvard Business Review’s research on independent consultants finds that ninety percent of independents chose independence proactively, not reactively, citing project quality and strategic alignment with clients as the controlling factors above compensation.³ The same publication’s research on passion-driven work identifies mission-driven executives, nonprofit employees, teachers, nurses, and physicians as among the populations most at risk for burnout, with the mechanism being the gap between the worker’s identity investment in the mission and the organization’s reciprocal investment back into the worker.⁴ A 2025 follow-up extended the finding by surveying when passion-driven professionals decide it is time to leave careers they care about.⁵

The market consequence is that top-tier specialists, the kind who deliver outputs above contracted scope and discount rates below their market, migrate toward clients who reciprocate with strategic partnership rather than transactional management. The receiving clients are increasingly in the private sector or in better-governed nonprofits whose boards conform to the Stanford recommendations (focused missions, rigorous evaluation, engaged directors, formal executive-evaluation procedures). Compensation differential matters; it is not the controlling factor. The controlling factor is whether the contractor’s strategic recommendations are heard, evaluated against data, and either implemented or refused on grounds that the contractor can cross-check against evidence.

This is context for the figures in this report. The growth recorded across the 2025/26 season was produced by a specialist team operating against a budget that was capped, a campaign portfolio that was overridden on several occasions on emotional rather than evidentiary grounds, and a board structure that the Stanford research identifies as broadly typical of the sector. The growth is what it is. The cost of producing it under those conditions, against an unsubsidized baseline, is meaningfully larger than the SOW line items disclose. A change in leadership that reciprocates strategic partnership rather than treating specialist labor as a transactional purchase is the condition under which the cost compresses and the upside expands. The current managing director’s posture, documented across the May meetings as systems-minded and operationally rigorous, is consistent with that condition.

Data still pending

One source remains outside this report. The Meta Ads Manager CSV for 2025/26 campaign performance has been described verbally on the May 1 review call but has not been exported as a file. When it lands in the project’s data folder, the build script reads it on the next run; until then, the dashboard does not display estimates as facts. The Constant Contact data that previously sat in this section has been ingested and now appears in section four.

VII.The unanswered questions, and the Sprint that answers them

Households, attribution, lifecycle, and renewal economics. One sprint, $4,800, thirty days, fixed fee.

The list of items the May 12 Data Reporting Strategy Meeting identified, restated in the language of that meeting:

How many of the tickets sold were actually scanned at the door. Which campaign brought which patron through to a purchase. How a couple appearing in ThunderTix as two separate customer numbers should be counted as one household. How much each named household has given the organization across all categories of contribution over three seasons. Brandon’s formulation to Susan on May 12 was specific: "I have, to this day, not been able to go into any system and pull a report of John Raymond and Beverly Raymond’s spending history." That report is the first deliverable of the proposed Strategic Intelligence Sprint.

The Sprint resolves four problems in sequence. First, a Universal Customer Record protocol that defines how households, multi-account patrons, and name variants collapse into a single identity (CMSPB has 3.7% of its 1,575 patron records using the formal Household field, another fifteen percent carrying signal in compound first names or free-text notes, and zero percent populated in the two fields ThunderTix actually provides for the purpose). Second, a reconciled three-season patron foundation joining tickets, donations, and dues by Customer ID where present and by email with last-name and zip fallback where it is not. Third, a choke-point map of the points at which data quality degrades, principally the iframe boundary between cmspb.org and ThunderTix and the manual customer-name editing in the ThunderTix admin UI. Fourth, a live renewal campaign launched against the cleaned data with measurable targeting and verifiable attribution.

The Sprint is $4,800, a fixed-fee thirty-day engagement delivered ahead of the July 15 membership-renewal cutoff. The written scope lives at cmspb-proposal.skippi.ai.