NIRI Dallas SWC: Managing Sell-Side Consensus and the Power of Credibility

At a recent NIRI Dallas Southwest Chapter session, Andy Detwiler, VP of Strategy and Outcomes at Q4, took the stage to tackle a pain point every investor relations practitioner knows too well: the manual, time-consuming grind of managing sell-side consensus. Drawing from his background on the sell side, consulting for hedge funds, and analyzing thousands of financial models at Virtua, Detwiler walked the audience through a practical playbook designed to bring structure to the madness.

Here is how Detwiler broke down the session, keeping his presentation grounded in real-world IR workflows, practitioner hacks, and the tech landscape shaping the space.

The Core Thesis: Credibility as a Monetizable Asset

Detwiler kicked off by reframing what consensus management is actually about, establishing credibility—first for yourself as an IR professional, and second for your leadership team.

"Credibility is an asset that you can monetize," Detwiler pointed out. He explained that establishing consistent trust with the street directly provides stakeholder value by expanding valuation multiples and driving down the cost of both debt and equity capital.

To back this up, Detwiler highlighted an academic study tracking stock returns and consensus dispersion:

• Lower Dispersion Boosts Returns: Companies sitting in the lowest quartile of analyst estimate dispersion achieved an average 5.2% higher annualized stock return.
• Consistency Magnifies Value: When tight estimate dispersion was paired with consistent operational execution, that outperformance jumped to 8.5% annually.

As Detwiler emphasized, when you compound those returns over time, managing estimate alignment becomes one of the most powerful levers an IR team has to drive value.

The 3-Phase Playbook for the 90-Day Earnings Cadence

To manage this process effectively, Detwiler broke the typical 90-day earnings cycle into three distinct operational subsets, stepping through what IR teams should prioritize during each window.

Subset 1: Pre-Quiet Period (Anchoring Consensus)

Going into the quiet period, Detwiler noted that IR teams have one final chance to proactively engage the sell side and understand where everyone sits before lockdown.

His top execution priority here, is a deep-dive line of analysis. He advised going through analyst models line by line rather than just looking at top- and bottom-line consensus. Look for guidance alignment, track estimate revisions during the quarter, and pinpoint who your outliers are. If an analyst has strayed far from guidance, this is your window within Reg FD to reach out, walk through their logic, and understand their drivers.

Subset 2: Quiet Period & Pre-Call Preparation

Once you enter lockdown, Detwiler framed this phase around four basic questions: What was promised? What was expected? What was delivered? And what is the right framework to contextualize future expectations?

To prepare management for a successful call, he shared several priority steps:

• Audit Peer Transcripts: Go back 12 quarters across your peer group to understand what macro and company-specific themes keep coming up.
• Build a Top-30 FAQ Library: Aggregate questions from inter-quarter interactions, Non-Deal Roadshows (NDRs), and conferences. Use them to predict the top 30 questions you'll get on the call and assign clear ownership for who answers what.
• Prep for Outliers: Identify potential "boogeyman" questions from difficult analysts in advance so management isn't caught off guard.
• Align the C-Suite: Ensure the CEO and CFO are on the exact same page regarding tone, so executive optimism matches financial reality.

Subset 3: Post-Earnings (Resetting Expectations)

Right after the earnings call wraps, Detwiler explained that your immediate focus must shift to shaping the upcoming "flash models" to prevent estimate drift.

He recommended running immediate post-call callbacks to catch line-item errors right away. Detwiler also stressed checking third-party aggregator feeds like FactSet, Bloomberg and others. Because these platforms digest models differently, minor discrepancies in tax rates or GAAP versus non-GAAP numbers can throw off published consensus.

Additionally, he cautioned that if management plans to drop or change a corporate metric, you should give the street 1 to 2 quarters of advance notice so analysts can adjust their models smoothly without assuming something is wrong.

Actionable IR Hacks from the Field

One of the most engaging parts of Detwiler's presentation centered on the specific hacks uncovered during his research conversations with senior IR practitioners, CFOs, and FP&A leaders:

• Two-Tiered Pre-Quiet Surveys: Survey sell-side coverage to see what topics are top-of-mind for the upcoming call. Simultaneously, gauge buy-side sentiment, one practitioner Detwiler spoke with used this technique to catch the buy-side "whisper number" down to a penny, preventing a valuation blow-up.
• Planting Strategic Questions: He suggested teeing up a specific question with a neutral sell-side analyst to ask during live Q&A. This lets you address a point of interest on the call without putting it directly in your prepared script.
• Maintaining an "Internal Police" Model: Keep a clean internal baseline model not only to track sell-side assumptions, but to police internal FP&A teams who might be pushing overly aggressive projections.
• Staggering Your Release and Call: He highlighted a unique strategy used by some teams: release earnings at night, spend the evening on sell-side callback calls to clear out basic housekeeping questions, and host the official earnings call the following morning with a refined script.
• Investing in Junior Analysts: "Never embarrass an analyst, especially a junior analyst," Detwiler reminded the room. Junior analysts do the heavy lifting on models, and when portfolio managers ask for recommendations, a junior analyst who trusts your IR team will be your biggest advocate.

AI Workflows: Deterministic vs. Probabilistic Models

Detwiler closed his talk by addressing where AI fits into consensus management, and where generic AI tools fall short.

He pointed out that generic Large Language Models (LLMs) operate on probabilistic risk. Because they generate variable outcomes every time you query them, generic LLMs struggle with multi-step financial reasoning and carry high hallucination risks when reading complex tables or unstructured PDFs.

Instead, he argued that financial workflows require deterministic models built on explicit, rule-based logic (such as Python). Deterministic systems handle financial data layers without arithmetic error, enforce compliance guardrails by design, allow for MNPI redaction, and maintain clean audit trails.

During the Q&A session, Detwiler acknowledged that model tracking across dozens of sell-side spreadsheets remains a heavily manual process today, which is precisely why dedicated, deterministic AI tools are being developed to automate model ingestion and root-cause error analysis.

The Bottom Line

As Detwiler put it to wrap up the session, the relationships and narrative control remain entirely yours, technology is just there to give you your time back. By tightening up the 90-day cadence, leaning on smart practitioner hacks, and using purpose-built tools to catch model disconnects before they turn into market surprises, IR teams can stop playing defense and start capturing the valuation premium their management teams deserve.