Walk an IR manager at a listed mining company through the last four earnings cycles and ask them to recall the analyst questions from each call. The specifics will blur together, but the themes will be identical. Production guidance trajectory. AISC versus budget. The hedging book and what it implies about realized prices. Capital allocation between sustaining and growth. Reserve life at current production rates. Anything on the balance sheet that is getting close to a covenant.
These questions are not going away. They are the right questions for equity analysts to ask about a producing resource company. The issue is not the questions themselves. The issue is that the IR preparation process treats each one as a fresh problem every four months.
Why the wheel gets reinvented
There are a few mechanisms by which IR teams end up redrafting answers to questions they have answered before.
The most common is that the prior answer is not stored anywhere accessible. It exists in the transcript of the last earnings call, in an email thread from the call preparation sprint, or in someone's memory. It is not in a document that the current IR team member can pull up, review, and use as a starting point. The result is that each cycle's call prep begins from zero even when the answer to the question has not materially changed.
The second mechanism is that prior answers are stored but are not structured for reuse. A document that contains the verbatim transcript of last quarter's call is not the same as a document that contains the approved, polished version of the company's answer to the AISC guidance question. The transcript version includes the verbal filler, the follow-up clarifications, and the context-specific details that were relevant to that specific call. Reusing it directly would require significant editing to extract the durable answer from the call-specific context.
The third mechanism is that even when a structured answer exists, the review process treats it as if it needs full fresh review because no one is confident that the prior version was approved at the right level. Was this answer signed off by the CFO? When exactly? Does it still reflect current policy? The answer to those questions requires tracking information that was not captured when the answer was written.
The structure of a reusable answer
An answer that can actually be reused across multiple earnings cycles has three components: the durable framework, the period-specific data fields, and the approval metadata.
The durable framework is the part of the answer that should not change unless the underlying policy has changed. For an AISC guidance question, the durable framework includes the methodology for calculating AISC, the key cost drivers at this company's specific operation, the approach to cost guidance and the bands around which guidance is set, and the company's stated expectation for cost performance across the cycle. These elements are stable and do not need to be redrafted each quarter.
The period-specific data fields are the placeholders within the framework that update each cycle: the current AISC guidance range, the actual AISC for the most recent period, and the specific factors driving any variance from guidance. These fields are the only parts of the answer that require new input from the mine site and finance team each quarter.
The approval metadata records who reviewed the durable framework, at what level, and when. This is the piece that answers the question "was this signed off?" without requiring anyone to search through email chains. It is also the piece that makes it safe to reuse the framework: if the policy has not changed and the framework was approved by the CFO six months ago, the current IR manager can proceed on that basis with confidence.
Building the system in practice
The practical question is how to actually implement this without creating more administrative overhead than it saves. At a four-person IR function, adding a complex knowledge management system would be counterproductive.
The minimum viable version looks like this: a shared document or folder containing one file per recurring question type, organized by topic category. Each file contains the approved durable framework for that question, the data fields that need to be updated each cycle, the approval record at the bottom, and the version history showing when it was last updated and what changed.
At the start of each quarterly prep cycle, the IR manager opens the file for each question type, reviews whether the durable framework is still accurate, updates the data fields with current-period numbers, and routes for approval if anything in the framework has changed. If nothing has changed in the framework, only the data fields need fresh review and sign-off. The CFO does not need to re-read the entire AISC explanation if all that has changed is the guidance range from the number that was set four months ago.
The question log practice from the prior article on Q&A banks applies here as well: reviewing the most recent call transcript to identify whether any questions were asked that do not yet have a corresponding file in the system. New files get created as new question types appear, so the coverage improves over time without requiring a one-time build-from-scratch effort.
The role of AI in maintaining the system
A structured Q&A management system is where an AI drafting tool adds consistent operational value, not just at initial setup but at every quarterly cycle. When the IR manager is reviewing the prior answer for the AISC question and needs to update the data fields with current-period numbers, a tool that has access to the prior approved version and the current-period data can generate the updated draft answer automatically, leaving the IR manager to verify that the numbers are correct and that the durable framework still reflects current policy.
This is not the same as the tool generating a novel answer from scratch. It is more constrained and more useful: applying the approved structure to current-period data. The quality bar for this kind of output is higher because the template is known and deviations from it are immediately visible. If the generated output changes a phrase in the durable framework without a clear reason, the IR manager will catch it because they know what the approved framework says.
The system also creates better inputs for voice calibration. A structured Q&A library of approved answers, reviewed and signed off over multiple cycles, is a higher-quality training corpus for voice calibration than raw quarterly reports. The answers have been edited to reflect the company's preferred phrasing rather than first-draft language, and the approval metadata confirms that they represent the company's actual communication standards rather than intermediate drafts.
What this does not solve
A Q&A management system is useful precisely for the questions that recur. It does not help with the questions that appear for the first time because operating conditions have changed, a new analyst is asking from a different analytical framework, or a macro event has suddenly made a topic newsworthy that was not before. Those require the same fresh thinking they always have. The system's value is in freeing up the IR manager's time for those novel situations by reducing the cycle time on the recurring ones. A team that spends 40 percent of its call prep time redrafting answers to questions it has answered before four times has more room to prepare for the genuinely hard questions than a team where that 40 percent is already handled.
