We Heard You Loud and Clear

Last quarter, we crowned “Am I audible?” the national anthem of the Indian conference bridge.

So let’s open this edition with the answer corporate India has been waiting for: we heard you. Loud and clear. All 962 of you.

Across 45 days, from July 2 to August 14, 2026, we captured 962 Q1FY27 earnings calls and 954 hours of audio. The platform transcribed the calls, attributed each speaker, extracted guidance, and kept every question tied to the passage it came from.

Because it did the same for 955 calls in April and May, this quarter reads as a delta, not a snapshot.

The delta has a headline nobody was saying out loud: the guidance mood flipped positive. Calls where management raised at least one guidance metric more than tripled to 50 from 16 last season, while explicit cuts stayed flat at around 15. The anxiety vocabulary deflated. Congratulations reached 1,755.

The mic picked up a market quietly turning more confident.

The guidance ledger

Counting which words companies say is table stakes. Every call says “margin” and nearly every call says “growth.” The useful question is which direction management moved its own numbers, compared with what it promised last quarter.

Because each call summary extracts explicit guidance statements, the corpus can keep that ledger across seasons.

The verdict for Q1FY27:

• 50 calls raised at least one guidance metric, up from 16 last season.
• 446 calls explicitly held, retained, or reaffirmed a target.
• About 15 calls lowered at least one target, statistically unchanged from roughly 14 last season.

Last season, raises barely outnumbered cuts. This season, the ratio was better than three to one in favor of upgrades.

The upgrades were not just vague optimism. One pharma company moved FY27 revenue-growth guidance to 14–16% from 12–13% and EBITDA margin guidance to about 23% from 22%. Another raised its revenue target to more than ₹700 crore from ₹650 crore. An infrastructure company lifted its floor for revenue growth to 15% from an earlier 10–15% band.

The cuts were narrower and explainable. Floods forced one cement maker to lower volume-growth guidance to 8–9% from 11–12%. A specialty-chemicals company cut EBITDA margin guidance to 10–12% from 25–28% under conflict, raw-material, and logistics pressure.

This is where the corpus starts to compound. Every revision is timestamped and attributed. Next season, the same ledger can ask whether the upgraded revenue bands were delivered, whether the lower margin range held, and whether the demand deferred to the second half ever arrived.

Guidance stops being a talking point and becomes a track record.

The anxiety complex cooled

With near-identical call counts across the two seasons, theme movement can be measured against a clean baseline.

The macro worry words moved down together. “Geopolitical” fell 26%, from 1,436 utterances to 1,062. The number of calls discussing GST dropped from 318 to 215, crude from 211 to 150, China from 255 to 219, and tariffs from 237 to 207.

Tariffs did not disappear. They became furniture: still discussed on more than one in five calls, but no longer the new thing in the room. Across exporters, the framing was remarkably consistent — near-term hesitation, rerouting in progress, financial impact deferred.

The structural themes barely moved. Exports appeared on 429 calls against 425 last season. Order books went from 309 to 315. AI held at 320 calls after 332 last season. Data centers stayed at roughly 145 calls in both periods.

Those are no longer quarterly stories. They are the permanent grammar of the Indian earnings call.

The themes that rose were seasonal, right on cue. Monsoon talk climbed 64%, from 135 calls to 222. Festive-season references more than doubled, from 43 calls to 100. Flood impact appeared on 49 calls, up from 27, and became a real earnings variable rather than background weather.

One theme quietly died. Labour Code discussion collapsed from 140 calls to 25. Last season’s provisioning wave was absorbed and forgotten.

Themes do not announce their exit. The corpus notices anyway.

AI outgrew its sector

320 calls discussed AI, more than discussed monsoon during a monsoon quarter. The count barely moved from last season, which is the interesting part: AI is no longer a novelty spike.

It now appears in industrial automation, lending models, hospital diagnostics, underwriting, collections, and customer service. Technology companies still talk about it most intensely, but they no longer own the theme.

Data centers showed the same persistence, holding at 146 calls from 142. The market has moved from asking whether AI will matter to asking where the capex, productivity, and revenue will show up.

Almost nobody supplied a clean answer to the last question.

What analysts actually ask

Speaker attribution lets us separate what management volunteers from what investors extract.

Set aside growth and margins, which dominate every Q&A. The differentiated analyst agenda was pricing, volumes, guidance, capex, and demand.

The asymmetry is the finding. Tariffs featured on 207 calls but generated only 113 analyst-question mentions. Managements raised the T-word; analysts attacked its consequences sideways through pricing, demand, volumes, and guidance.

Coverage was concentrated at the top. The most active brokerage asked questions on 149 of 962 calls — roughly one in every six calls in the season, half again as many as the next firm. The two most active individual analysts each worked 18 calls in 45 days: one earnings call every two and a half days, including weekends.

The congratulations economy

Here is the season’s most delightful convergence: the guidance ledger turned positive, and the room noticed before the headlines did.

We recorded 1,755 congratulations across 620 calls, up from 1,641 across 601 calls last season. That is 1.8 congratulations per call, or roughly one every 33 minutes of captured audio.

The anatomy of the ritual:

• It is almost entirely an analyst ritual. Of the attributable congratulations, 1,448 came from analysts and 74 from management. Analyst-spoken congratulations rose 21% season on season.
• The compliment has a fixed grammar. “Congratulations on a good set” appeared 257 times and beat “a great set” at 139. Enthusiasm is calibrated.
• It is a courtesy, not a verdict. Only 565 congratulations specified a “good set of numbers.” More than a thousand were issued without saying what they were for. The compliment opens the door; the question that follows is the real content.
• Sometimes it is earned. The season’s most-congratulated call collected 12 congratulations after reporting total income up 71%, profit after tax up 72%, and a ₹2,208 crore order book.

“Many congratulations” appeared 44 times. “Congratulations to the team” appeared 39 times. “An excellent set” managed nine.

Exactly one person all season said “hearty congratulations.”

The anthem endures

And then there is audibility.

“Am I audible?” was asked 1,182 times across 596 calls, versus 1,235 last season on a near-identical call count. Add 305 variations of “can you hear me,” and audibility anxiety crossed 1,487 utterances.

One call questioned audibility nine times. Across 954 hours of audio, nobody ever answered “no.”

The corpus caught the rest of the liturgy too: “thank you for the opportunity” appeared 1,534 times; “back in the queue” 368 times; “bookkeeping question” 131 times. Conference operators said “ladies and gentlemen” 3,224 times.

This is what full capture produces. Not only the headline numbers, but the texture of the room: the evasions, the rituals, the phrases nobody meant to turn into data.

What two seasons make visible

The point of sitting on two full seasons — 1,917 calls and counting — is not the trivia. Direction only becomes visible when both seasons are held to the same ruler.

A one-season read would have crowned tariffs the breakout story. The two-season read shows tariffs plateauing while guidance quietly moved up: 50 raises against about 15 cuts, compared with a near dead heat last season. It shows AI becoming infrastructure, Labour Code vanishing from the agenda, and the analyst conversation routing around management’s preferred framing.

Every number here is one query away from its source passage, with the speaker’s role attached. That is the system’s real prowess: not that it heard 962 calls loud and clear, but that it can answer questions about any 48 seconds of them, this season or last.

---

The current-season window is July 2 to August 14, 2026 (962 calls); the comparison window is April 1 to May 31, 2026 (955 calls). Sources are live-call audio, diarized transcripts, and structured call summaries. Guidance direction is matched on explicit revision language in extracted guidance items; buckets can overlap when one call moves multiple metrics. Theme counts are keyword-based distinct-call counts. Phrase counts are conservative lower bounds.

Q2FY27 calls begin in October. We will be listening.