Chief Marketing Officer / China
Choices that made the launch slower and easier to defend
- Sales of Tanqiao's first innovative medicine, 2025
- RMB 860 million
- Its first year on China's national reimbursement drug list
- Hospitals stocking the antibody, June 2026
- 1,140
- Up from 110 specialist centres at the October 2023 launch
- Twelve-month persistence among patients in the support programme
- 63%
- 23,400 patients enrolled; an indication, not a controlled result
Decisions & consequences
01Launching in 110 centres and waiting for the negotiation
Our IL-17A antibody was approved in August 2023 for adults with moderate-to-severe plaque psoriasis. It was Tanqiao's first innovative medicine and the company's answer to four years of falling generics prices. Several domestic and imported biologics were already sold in the class, and we expected to enter the 2024 round of reimbursement negotiations, with any listing taking effect in January 2025. Until then, patients would pay the full price themselves.
Our head of sales, Fang Zhiyuan, proposed launching at once in 600 hospitals, hiring 400 specialist representatives in the first quarter and building share before the negotiation. His argument was serious: early prescribers form habits, and a larger patient base strengthens a company's hand. I proposed launching in 110 grade-A tertiary dermatology centres, staffed by 140 representatives transferred from our dermatology generics team, and hiring the rest only after the negotiation. Self-pay demand outside major centres would be thin; a large field force with little to sell would be tempted into exactly the promotion we were leaving behind; and the money would serve patients better after listing. Gao Shiwen backed my plan after our CFO modelled both.
We gave up about RMB 150 million of self-pay sales: 2024 sales were RMB 186 million against Fang's plan of RMB 340 million. Two competitors gained share in provincial hospitals that year, and some of it has not come back. Fang told the board plainly that he disagreed, which I respected.
In the 110 centres we concentrated on dermatologists' experience with the medicine, injection training and an independent foundation's means-tested assistance programme. The listing took effect in January 2025. We then hired 420 specialist representatives over six months, and the antibody reached 640 hospitals by December 2025 and 1,140 by June 2026. Sales were RMB 860 million in 2025.
What changed is how Tanqiao plans launches: the size of the field force now follows access, not the other way round. I would do one thing differently. Hospital pharmacy committees in second-tier cities meet infrequently, and I should have started applications there during the self-pay year. Waiting cost us roughly a quarter in the provinces where we most needed speed.
Closing 3,800 sponsored department meetings a year
When I joined Tanqiao, marketing controlled an academic promotion budget of RMB 640 million a year across generics and the planned launch. Most of it paid for department meetings in hospitals: a short talk, a meal, and travel or speaker fees for the doctors involved, about 3,800 meetings a year. Each meeting had paperwork. Few had a documented educational need, and our own review found that around 400 doctors received over a third of all speaker payments.
In mid-2023 the campaign against corruption in medicine, led by health and disciplinary authorities, intensified, and many companies froze hospital activity while they waited to see what would be allowed. I did not want a freeze, which assumes the old model will return. In July 2023 I proposed ending marketing-sponsored department meetings entirely; moving RMB 230 million of the budget to medical affairs under Lu Yanting for scientific exchange with documented needs; setting speaker fee bands with a cap of ten paid engagements per speaker a year; and funding an accredited online curriculum through the Yuequan Dermatology Teaching Society, which would control its own content.
Fang Zhiyuan objected strongly. Meetings were his representatives' main reason to be in a hospital, and he expected our hospital generics to suffer. Two regional sales directors left within months. Generics sales in the second half of 2023 came in RMB 210 million below plan, though some of that shortfall would have happened anyway as hospitals restricted visits.
Promotional and educational spending fell to RMB 260 million in 2025. The online curriculum had registered 41,000 physicians by December 2025, about a third of them from county hospitals our meetings had rarely reached. External compliance reviews in 2024 and 2025 found no payment without a documented purpose. Tanqiao now keeps a register of every paid engagement, reviewed twice a year by the Audit Committee.
Where I went wrong was the sequence. We closed the meetings before our representatives had anything better to offer, and for about three months many of them visited hospitals with nothing useful to say. I would train and assess the field force on the science first, and only then close the old programme.
Keeping patient data away from the sales force
After listing in January 2025 our antibody's volume grew quickly, and so did the number of patients who stopped. Market research suggested that roughly half of self-paying patients in 2024 had stopped within a year, and reimbursement alone would not remove the reasons: fear of injecting, a minor infection that alarmed them, a missed refill after moving cities, or confusion about collecting a reimbursed injection from a designated pharmacy.
Our digital team proposed a patient service inside a popular messaging app, linked to the sales force's customer system. It would show each representative how many patients from each doctor had enrolled and how long they stayed on treatment. Gao Shiwen liked a second feature: Tanqiao would own a database of treatment outcomes. I opposed both. Linking patients to prescribers in a sales system would turn support into a way of monitoring doctors, and a dataset gathered by marketing would carry neither the consent nor the method that research needs.
We chose a third-party nurse service, Anxu Care Services, under a contract that keeps identifiable data with Anxu. Patients give separate consent for their health information, as the Personal Information Protection Law requires for sensitive personal information, and enrolment is offered by pharmacists and nurses, never by our representatives. Tanqiao sees only aggregated reports by province. We gave up doctor-level visibility, which Fang believed would have helped him deploy his field force, and the outcomes database, which our chief executive regretted.
The programme launched in March 2025 and had enrolled 23,400 patients by June 2026, at about RMB 48 million a year. Twelve-month persistence among enrolled patients was 63 per cent, against 46 per cent among patients we could observe outside the programme through pharmacy data. Patients who choose support are not a random sample, so I present that gap as an indication, not a measured effect.
What changed is that patient services now sit behind a firewall from sales, and our compliance committee audits it every year. What I would do differently is to have asked medical affairs to design an ethics-approved registry at the same time. It began only in 2026, and we lost a year of properly gathered evidence.
Withdrawing a superiority chart before the second launch
In April 2025, a month before our antibody was approved for ankylosing spondylitis, our agency delivered a refreshed psoriasis campaign. Its centrepiece was a detail aid comparing our week-12 PASI 100 rate with a competitor's, taken from two separate trials with different populations and designs. The chart suggested we cleared skin faster. The same campaign included short videos in which a lifestyle influencer described her skin before and after treatment, aimed at patients on social platforms.
Our medical reviewer flagged both. A cross-trial comparison cannot support a superiority claim, and Chinese law confines advertising of prescription medicines to professional medical publications, so a patient-facing brand campaign was not open to us at all. Fang Zhiyuan argued that competitors used similar charts and that his representatives needed a reason for a dermatologist to switch. He was right that the field force needed something to say. He was wrong that this was it.
I withdrew the material. We pulped 38,000 printed detail aids, removed the digital versions from representatives' tablets, cancelled the influencer contract and wrote off RMB 6.8 million. The psoriasis campaign slipped by seven weeks, in the quarter we were also preparing the spondylitis launch. The replacement reported our own pivotal trial against placebo: 76 per cent of patients reached PASI 90 at week 12, 41 per cent reached PASI 100, about one in nine did not reach PASI 75, and the common adverse events, including candida infections, sat on the same page. Patient-facing content became unbranded disease education reviewed by medical affairs.
The outcome is harder to measure than the write-off. Our share among new psoriasis patients in hospitals we already served held steady through the delay. Several hospital pharmacy committees reviewing the spondylitis indication cited our evidence file directly, and we have had no complaint about our promotional claims since.
What changed is the order of review: medical and legal colleagues now approve the list of claims at the brief, before any agency designs anything. That is also what I would do differently in hindsight. The chart should never have reached artwork, and the RMB 6.8 million was the price of reviewing too late.