THRIVE Launch Success Code™.
Decoding the DNA of drug launches.
What a medicine brings to its launch before the first decision is made: 198 FDA approvals, read through their DNA.
The Launch Success Code™ is not a report to read once. It is a diagnostic that turns a generic success benchmark into an asset-specific launch agenda.
Research across 198 FDA novel approvals (2016 to 2024), tested across 33 launch variables. To our knowledge, the broadest factor analysis of launch performance published to date.
A launch arrives with odds already attached.
Eleven traits, most of them fixed before approval, separate launches that tend to beat their forecast from launches that tend to miss. Because the fixed traits travel together, a launch rarely carries just one. Launches with a predominantly favorable profile beat their forecast about two-thirds of the time and missed one time in four. Launches with a predominantly unfavorable profile showed the mirror image: one beat in four, two misses in three, and more than 40 percent came in below 40 percent of forecast. The unfavorable profile is also the most common: 117 of the 198 launches.
The eleven markers, in four groups, shown by the level that, in this sample, more often beat or more often missed forecast. Eight are fixed at approval; three describe the organization and are levers a company can pull. Dashed: consistent in direction with the published literature, below our screening threshold on its own.
About half of new drug launches miss the expectations set for them before approval.
Six independent studies, covering launch classes from 2003 to 2025 and samples from 46 to 340 products, put the share of launches that underperform their forecast between 31 and 66 percent. The median across the six is 47 percent. Study windows and definitions differ, and the finding does not.The Launch Success Code™ asks the question the literature leaves unasked: before any launch decision is made, what does the asset itself bring, and does it shift the odds?
Six studies, one finding. Each study defines performance slightly differently, so the rates are not directly comparable; the definition each used, and the citation, are in the Executive Summary endnotes. The chart shows industry-wide launch outcomes, not any service provider’s launch portfolio.
THRIVE’s Launch Success Code™ diagnoses and charts the course to beating the odds.
There is a large body of work on why launches miss. THRIVE reviewed more than 50 studies from 22 organizations, published between 2013 and 2026. Read together, they converge on a small number of explanations: market access secured too late, organizations that under-invest or under-commit, patient populations sized larger than they could be found, products not different enough to move prescribers. These are careful studies, and their advice is sound.
They also share a frame. Almost all of them ask what the launching company did: when it engaged payers, how much it spent, how it found patients, how it positioned the product. That is the right question in the last eighteen months before approval. It leaves a higher-level question unasked: before any of those decisions is made, what does the asset itself bring to the launch, and does it shift the odds? We think about a launch the way a biologist thinks about an organism.
What the asset arrives with: its modality, how it is taken and how often, whether treatment is chronic or a single course, the disease it treats, the safety obligations the regulator attached to it. None of this can be changed once the product is approved, and all of it is knowable from public information, often years before approval.
Expression: the decisions that switch those traits on or off. This is where the published literature lives, and where the launch team earns its keep.
Payers, policy, competitors, the practice of medicine. It surrounds both, and it is changing fast. A retrospective analysis cannot say on its own what will work next; we address that layer in our THRIVE Futures Architecture™ work.
The launch that results is what the genome and its expression produce in that environment. This report is about the first layer: 198 FDA novel approvals from 2016 to 2024, 33 variables coded from public sources and consolidated to 18 for testing, each tested statistically against how the launch performed relative to the forecast set for it before approval. What follows reflects tested associations rather than impressions, and the non-findings matter as much as the findings.
198 launches. One honest distribution.
Of 198 launches with clean pre-launch forecasts and reported sales: 34% beat their forecast, 58% missed it, 9% landed within the band, and 35% were severe misses, delivering less than 40% of forecast. At that level, a miss can blow a hole in a big-pharma P&L; for a single-asset company, it can be fatal.
Most industry narratives do not survive testing.
Oncology is the safer bet. Oncology launches performed no differently from non-oncology launches.
A Fast Track or Breakthrough badge signals differentiation. Neither designation separated beats from misses, and neither did being first in class.
Competition decides the launch. Not whether a rival launched within six months of approval, not how crowded the indication already was.
A Complete Response Letter leaves a scar. A CRL on the way to approval left no mark on the launch that followed.
Price explains performance. Launch price, tested on the 133 launches with a filed price, showed no association with performance on its own.
Several badges the industry uses as proxies for a strong launch were not the traits associated with the odds in this sample.
Eleven markers, in four groups, shift the odds.
Same disease, two genomes.
Skyrizi and Sotyktu
Both launched into plaque psoriasis, both from large, experienced organizations with deep immunology franchises, into the same payers and the same prescribers. Skyrizi is an antibody, injected once a quarter after loading, chronic, without a REMS, with full approval, a broad label and head-to-head superiority over two incumbents, from a very experienced team: six of the eleven markers on the favorable side and none on the unfavorable. AbbVie’s own outlook before approval put 2025 sales at about $5 billion; the company raised that to more than $7.5 billion in 2022 and reported $17.6 billion for 2025.
Sotyktu is an oral, daily small molecule, chronic, without a REMS: the treatment-design cluster on the unfavorable side, with the same favorable marks on evidence, label and team. Bristol Myers Squibb’s own outlook, filed before approval, put its revenue potential at more than $4 billion by 2029; reported sales in 2025, its third full year, were $291 million. Company quality does not explain the gap; both companies know how to launch. The environment does not explain it; it was the same environment. What differed was the genome.
One genome, expressed twice: Krystexxa
This launch sits outside the sample, approved in 2010, and isolates the second layer. Krystexxa is an intravenous biologic for chronic gout that has stopped responding to conventional therapy: a specialty infusion for a small refractory population, the same genome from the day it was approved. Its first owner launched it at the three million people with gout rather than the fifty thousand who were eligible, with a sales force still being hired, late planning and a value story payers did not accept. Sales reached $6 million in 2011 against an $18 million forecast; the company sold in bankruptcy three years after approval.
Horizon acquired the product at the end of 2015 and relaunched it at the refractory population, through rheumatologists, with a field force quadrupled to nearly 200, account managers built for reimbursement, new evidence and, in 2022, a label expansion. Sales grew from about $35 million in 2015 to more than $1 billion by 2023. The genome never changed; the expression did.
Neither case is an argument that the genome decides. Skyrizi was also launched extremely well, and Krystexxa shows what expression alone can move. Genome, expression and environment produce the phenotype together, and all three matter. Sources: company disclosures and published sell-side research, cited in the Executive Summary endnotes.
Read the genome before writing the plan.
Every launch plan we have seen starts from the product’s strengths. The markers add a second starting point: the traits the product cannot change, and the odds that accompanied them in this sample. A daily oral in a crowded category is not a worse medicine than a quarterly injectable; it is a launch whose forecast, in this sample, was more likely to outrun its reality, and whose plan may need to spend more, earlier, on the two places where daily orals are lost: persistence and access.
Modality, route, frequency, duration, orphan status and the likely safety obligations are known when an asset is licensed or a program is prioritized, long before a launch team exists. A portfolio carrying several unfavorable stacks may warrant closer scrutiny of its forecasts and underlying assumptions. The markers are a way to see that early and to price it into the plan, the deal or the hiring.
Because performance was measured against the forecast, the markers describe where expectations tend to outrun results as much as where launches go wrong. The published literature already knows that pre-launch forecasts run high on average. What it has not said is which assets get over-forecast, and by how much. A forecast built with that in mind is a better forecast; a plan built against it is a better plan.
Expression matters most where the genome is hardest.
The markers tell a team, well before approval, which odds it is starting with and where its expression has to work hardest. They are a starting point, and they are not a verdict. In the most unfavorable profiles in the sample, launches from experienced companies with very experienced teams missed about half the time; first-time launchers with the same profile missed three times in four.
Download the Executive Summary
The Phase I findings of the THRIVE Launch Success Code™ in twelve pages: what is well established, what is new, what the DNA says, two launches read through their DNA, what this means for the launch team, for portfolio and business development, and for forecasting. Method and sources in the appendix.
Download the PDFRead the DNA of your next launch
A guided conversation, about an hour, in which we compare your asset’s fixed traits with the 198-launch sample, show how launches with similar traits performed, and discuss where your launch plan may need to work hardest. No confidential information needed to start.
Schedule the conversationTell us what you are launching, and when.
We come back with two or three times for the hour, and what to have in hand. Modality, route, dosing, disease and regulatory posture are enough to read the genome; nothing confidential is needed to start.
Methodology and caveats. From about 460 FDA CDER/CBER novel approvals (2016 to 2024), we retained 198 launches with a sell-side forecast published 6 to 24 months before approval and product-level reported sales. Performance was measured against the forecast on its own terms and sorted into bands: above 120% of forecast, beat; 80 to 120%, met; below 80%, miss; below 40%, severe miss. 33 variables were coded and consolidated to 18 for testing, one at a time, with an effect-size measure and a significance screen; the screen was repeated on four alternative samples and every marker kept its direction. The markers are associations, tested one at a time; they are not causes, and the effect sizes are moderate. Throughout, success and failure mean performance relative to the sell-side forecast published before approval, not absolute revenue, so the markers describe where expectations tend to outrun results as much as where launches go wrong.
Launch with Precision. Thrive by Design.
If your organization is launching a new therapy where failure is not an option, let’s talk.
Start a conversation