{
  "abstract": "Randomized controlled trials, commonly known as A/B tests, are the standard mechanism for inferring causal effects of product innovations. Standard platforms evaluate these tests with fixed-horizon procedures that control the false positive rate at a predetermined time, forcing teams to wait until the end even when early data strongly indicates success or failure. Unfortunately, continuous monitoring (\"peeking\") invalidates these guarantees, while always-valid inference is overly conservative under a strict deadline, and group sequential designs (GSD) have received limited traction in large-scale experimentation due to operational barriers: terminal-rule preservation, autocorrelation, latent effect-size distributions, calendar-duration optimization. We fill in this gap by proposing a Predictive Probability of Success (PPoS) group sequential framework that is autocorrelation-aware and duration-optimal while preserving the platform's fixed-horizon terminal rule. Our key insight is that PPoS thresholds map to closed-form lift boundaries that converge to the standard test at the horizon by construction, admit an effective-sample-size correction estimated from historical A/A tests, and suffice to attain the likelihood-ratio efficiency floor. Experimentally on online A/A and A/B experiments, we show that the proposed policy reduces expected duration by 43.1% (7.97 days vs. 14-day baseline), increasing effective platform capacity by 76%, while retaining the guarantees on FPR and power.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.27899",
  "author_total": 7,
  "authors": [
    {
      "name": "Lorenzo Perini"
    },
    {
      "name": "Mia Garrard"
    },
    {
      "name": "Rina Friedberg"
    },
    {
      "name": "Rohit Kumar"
    },
    {
      "name": "Houssam Nassif"
    },
    {
      "name": "Abbas Zaidi"
    },
    {
      "name": "Pablo Alcain"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/4mf8g8yynjkrh.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.27899",
  "fulltext_url": null,
  "id": "4mf8g8yynjkrh",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.27899",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "stat.AP"
    ],
    [
      "openalex",
      "2613|Statistical Methods in Clinical Trials"
    ]
  ],
  "openalex": "W7223119010",
  "pdf_url": "https://arxiv.org/pdf/2610.27899",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-20",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/4mf8g8yynjkrh.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:34:13Z",
      "native_id": "2610.27899",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.27899"
    }
  ],
  "subjects": [
    {
      "field": "mathematics",
      "origin": "arxiv:stat.AP",
      "subfield": "statistics"
    }
  ],
  "title": "Let Them Peek: Group Sequential Designs for A/B Testing",
  "title_text": "Let Them Peek: Group Sequential Designs for A/B Testing",
  "type": "preprint",
  "updated_at": "2026-10-30T12:34:13Z",
  "url": "https://research.draconitic.com/paper/4mf8g8yynjkrh",
  "venue": "arXiv",
  "versions": []
}