Altair
Skip to content

Trading Analysis Report: CRM

Analysis date: 2026-10-21 · Model: claude-opus-5-5 · AI-generated, not financial advice.

  • Analysis date: 2026-10-21
  • Rating: Hold
  • Generated: 2026-10-21 12:17:48
  • TradingAgents 0.6.0: anthropic, deep claude-opus-5-5, quick claude-sonnet-5-5
  • Analysts: market, sentiment, news, fundamentals; research debate rounds 5, risk debate rounds 5
  • Data vendors: core_stock_apis yfinance, technical_indicators yfinance, fundamental_data sec_edgar,yfinance, news_data yfinance, macro_data fred, prediction_markets polymarket

V. Portfolio Manager Decision

Rating: Hold

Executive Summary: Hold CRM at standard weight. Add no new money at $242.34 or anywhere in the ~$233-248 middle of the range, and trim only if your book is already above standard weight. Before any bid goes live, read three documents by Monday 10/26's close: the Q2 FY27 10-Q, the Braze funding release and the 8/26 earnings release/guide. If the filing reads clean, add via day-order limits at $231 (one sixth before the 10/28 FOMC statement, one sixth after), or add one sixth the session after a daily close above $249.45. Total adds, filled plus live, are capped at half a standard weight or 5-6% of portfolio, whichever is lower. Downside target weights are 0.9x standard on a close below $224.56 (added shares removed), ⅔x below $220.29, ½x below $205.26, and exit on a weekly SuperTrend flip, judged over a 6-10 week horizon through the early-December Q3 print.

Current Price: 242.34

Price Target: 245.0

Confidence: Medium

Time Horizon: 6-10 weeks

IV. Risk Management Team Decision

Aggressive Analyst

Aggressive Analyst: Hold doesn't mean timid here. The plan keeps full exposure to a stock that is up about 60% from its June closing low (149.86). It refuses to sell the middle of a range, and it pre-loads limit bids at 231 so I can buy a discount in size without needing nerve on the day. I don't know anyone's actual book, so read all of this as "standard weight stays, extra capital waits for a price."

I expect the conservative analyst to say: the monthly SuperTrend is down with a stop at 285.96, the daily is down, the closing highs keep stepping lower (263.97, 258.97, 246.74), OBV isn't confirming, the Fed is hawkish, leverage has jumped, and margins have stalled, so trim. Most of that is true, and it's why I'm not adding at 242.34. But trimming here is a bet with no edge. From 242.34, the 224.5 line is about 17.8 points away (7.4%), and the September closing high at 263.97 is about 21.6 points away (8.9%). That is roughly symmetric, so selling the middle of the range pays no one. The plan already has a hard exit rule: a daily close under 224.56 takes you below standard weight. That's a conservative rule with a price on it, not a feeling.

On the lower highs, I'll concede the 10-16 rejection at 251.94 was a real failure at the 249.45 flip level. That's exactly why the plan won't chase. But the stock fell 14.9% from the 09-03 close and found buyers at 224.56. MACD has risen four straight sessions, and price is still about 18% above the 205.26 pre-gap close. A stock that digests a 22.6% gap-up on 55M shares and holds most of it isn't acting like a dead-cat bounce.

On OBV, I'd be careful. Today's bar is partial, at 2.6M shares against 4.7M to 9.6M on other October sessions. OBV also adds or subtracts a whole day's volume based on the sign of the close. Today closed at 242.34 against 242.43 yesterday, so a nine-cent dip counts as distribution. That's a reason not to add on weak confirmation, which the plan already requires. It isn't a reason to sell.

On fundamentals, I won't pretend the problems away. Q2 operating income was flat at $2,331M against $2,332M, gross margin has slipped to 76.65%, and about $1.4B of Q2 net income looks non-recurring. Implied leverage is about 3.3x EBITDA, and that figure is my inference, not a reported number. Those are the reasons I'm not buying more at 242. But derived free cash flow is about $15.2B, a 7.6% yield on a $199B market cap, with a 34.7% FCF margin in FY26 and a share count down about 14%. Directors also bought $1.0M at 239.33 on 09-18, less than 2% below today's price, and David Kirk's three buys average around $230. The people who see the books have been buying right where the plan bids. Morfit's $25M at 260.58 is underwater, but that's a reason to be patient, not a reason to sell.

The macro case is overstated. FRED failed entirely, so the Fed view rests on Polymarket, and the year-end hike markets are thin, between $0.19M and $0.88M in volume. October hike odds actually fell from about 30% to 12% this week, US recession odds are 8%, and CRM went from 149.86 to 263.97 under this same hawkish backdrop. The Braze deal is $5.66B, about four and a half months of free cash flow. CRM stayed flat on the announcement while peers rallied, so the market isn't charging for it. The risk is serial deals adding to leverage, and that belongs in the filing review, not in a sell order.

I'd also not lean on the retail chatter either way. It's about 30 posts, only 5 labeled, and some claim CRM "reclaimed the 200DMA" and is "knocking on 260," which doesn't match the tape (200 SMA at 198.08, price at 242). The bear side is lotto puts and "too high too fast." Neither side is evidence.

To the neutral analyst, who will probably say "fine, hold and wait for more information": agree on the action, but waiting is only useful if it's short. The Q2 10-Q, the 8/26 guide and the August gap catalyst are public documents, so the filing gate is an hour of work today, not a month of paralysis. Resolve the uncertainty fast, then act.

Here is where I'd push the plan harder. The 224.5 stop is only 6.5 points below the 231 bid, about 0.75 of an ATR, which is inside a normal day's range. So run it on closes, as the plan's 224.56 trigger does, and size the thirds for gap risk. If the filing is clean and the bid fills near 231, you risk about 2.8% to make 8% to the 249.45 flip and about 5:1 to 263.97. That 5:1 is the trade worth waiting for, not today's 1:1. Between the 231 dip bid and the confirmed breakout above 249.45, both branches are covered, and the 233 to 248 middle gets no new money because the odds there are coin-flip. The conservative camp is paying for certainty that isn't for sale, and that's a worse bet than holding with a price on every branch. Aggressive Analyst: Neutral is right that this is a much narrower fight than the tone suggests. We all hold at 242.34, add nothing until the filing is read, bid near 231, and add on a confirmed close above 249.45. The only live question is whether to carry an extra quarter to a third less than standard weight in the meantime. I'll say what I'm giving up first, then fight on that one point.

Concessions

  • The 5:1 was a best case. With an exit near 220 instead of 224.5, a 231 fill is about 1.7:1 to the flip level and 3:1 to the September high. I'll use that range.
  • I overreached on insiders. Kirk's average includes 258.64, and the 239.33 buy was followed by a close 6% lower. Directors who bought at 194 watched it go to 150. They tell you the business isn't broken, not where the floor is.
  • The filing gate should be two-sided. If the 10-Q reads weak, act on it that day and don't wait for 224.56.
  • The stop needs a size, and I'll take the neutral's stepping. I'd tweak it so the first close under 224.56 removes any added thirds, since that is the thesis the dip bid was paying for. The standard-weight cut comes at roughly 220. The stock closed at 224.56 and 224.99 on consecutive days and traded to 220.29 intraday without breaking. Cutting a quarter of standard weight over a few cents would be a whipsaw.
  • I'll accept staggering. The first third goes live after the filing, and the later thirds wait until after the FOMC.

Why the honest numbers still favor the plan

Run the conservative's own slippage assumption on today's price. Exiting near 220 from 242.34 risks 9.2% to make 2.9% at 249.45 (about 0.3:1) and 8.9% at 263.97 (about 1:1). The 231 bid gets roughly 3:1 to the same high on the same assumption. The plan doesn't say buying is great. It says that waiting for 231 triples the payoff, and that is the only reason to wait.

Dollar held versus dollar bought

I'll take the conservative's logic at face value: a dollar held is the same bet as a dollar bought. But zero edge on the buy side means don't add. Zero edge on the sell side, after spread and tax (the conservative concedes they can't see lots), plus the cost of re-entering, means don't sell. Standard weight is supposed to be the no-view weight. Being 2 to 5% above a scenario bid is not a view, and as neutral said, that gap is inside the error bars of the valuation itself.

Gap risk runs both ways

The conservative uses the 11.9% August gap to justify trimming. The biggest single-day move in this dataset was up 22.6%. I'm not forecasting a repeat, but a third of the position sitting in cash through one would miss about 7.5% of the position. That is far more than the 1% rebuy cost quoted. Neutral is also right that confirmation probably prints well above 250, after the 10-16 poke to 251.94 that closed at 240.70.

The filing

The 10-Q and the 8/26 guide have been public for weeks, and the stock has traded on them since. The unread filing is our information gap, not the market's. So reading it is an hour of work, and trimming before reading pays a round trip to hedge a document you can open today.

Conservative's other points

  • The conservative says there is one line below at 224.56, but also that a 231 fill means the stock has already lost the 10 EMA at 239, the Bollinger middle at 234 and the 50 SMA at 232.7. Those can't both hold. If three supports sit above 231, the downside isn't one line.
  • On the yield, SBC isn't in the data, so I'll use a placeholder. Take $3B off for stock comp and another $1B for a full year of interest on the new debt, and the FCF yield is still about 5.6%. The 14% drop in diluted shares is already net of SBC dilution.
  • Interest paid sits inside operating cash flow. H1 OCF is up 10.4% with the new balance sheet in place for at least a quarter and a half. Coupons are semiannual, though, so cash may lag the expense.
  • On the OBV, it is 19M below 09-24 but also 19M above the 10-07 low. It's a gate for adds, and nobody has shown it's a reason to sell.
  • Macro is the missing-data problem I already flagged. I can't prove it's the same backdrop as June. What I can say is that 86% hold and 97% no cuts are already in a price that sits 60% off the low, and the one week-over-week move was hike odds falling from 30% to 12%.
  • CRM flat on the day of a 58% premium acquisition is what you'd expect. Acquirers usually fall.
  • The monthly SuperTrend stop at 285.96 is 18% above the price. If full weight requires the monthly to flip, you only own this stock after the recovery is over.

Where I land

Hold standard weight, and trim to standard only if the book is above it. Read the filing today. If it's clean, put the first third at 231. If it's mixed, bid toward 225. If it's weak, cut to roughly three-quarters of standard immediately. Take the later thirds after the FOMC, remove the added thirds on a close under 224.56, take a quarter off standard under roughly 220, and decide whether to stay at all in the 205 and 198 zone. Add one third above 249.45 with OBV improving, and don't chase past about 252. That puts a price on every branch without trimming into the middle for a hedge the filing can replace by this afternoon. I don't know your actual book, so if standard weight is too big for you, that is a sizing question, not something this tape is telling you. Aggressive Analyst: I'll start with what I'm giving up. Neutral's driftless-walk numbers check out. From 242.34 the stock touches 249.45 before 220 about three times in four, and 263.97 before 220 about half the time. So the 0.3 to 1 and 1 to 1 I quoted aren't an edge. They're what a near target and a far stop look like. By the same logic, my line that waiting for 231 "triples the payoff" was payoff shape, not edge, and I'm retiring it. The real case for 231 is the plan's own valuation work, since it sits at the low end of the clean-filing range and price pays you for the uncertainty. The case against adding at 242 is the same point: no edge in the middle means no new money. It also means no urgency to sell.

That's what I'd ask the conservative to sit with, because it shrinks the fight. Trim a fifth at 242.34 and you gain about 0.9% of the position if the stock dips to 231, and lose about 0.8% if it runs to 252. That's a coin flip worth roughly a point. The filing is worth far more, because a weak read points to 200 to 205, which is 15 to 17 percent below here. So the effort belongs on the document, not on shading size for a week.

On "a dollar held is a dollar bought at 242," that holds only if switching is free and you know fair value to the percent. Neutral is right that 2 to 5 percent above a scenario bid is inside the error bars of the scenario work. When your valuation noise is wider than the gap to price, you hold and save action for when the gap exceeds the noise. That's what the plan does: add at 231, cut near 220, and do nothing in between. Trimming and rebuying is also not the same view as the bid. The sale is unconditional and the rebuy is conditional. The trim's one real merit is that it pays in the weak branch, but the plan's day-of cut on a weak read already covers that. What the trim adds is protection for the days before I've read the document.

So I'll take Neutral's deadline. Read it, and if it isn't read by Tuesday's close, the default goes to a tenth to a fifth lighter, done on a completed bar. I'm fine being penalized for skipping homework. It's better than the conservative's version, because trimming after the IBM reaction hedges nothing about IBM. If IBM lifts software, you've sold into the move that gets you toward 249. If it hits software, you've sold at 236 what you wanted to buy at 231.

I need to correct my own "hour of work" claim. The Braze funding isn't in the 10-Q. The quarter ended in July, the deal was announced October 20, and closing is expected in fiscal 2028, so no cash leaves before February 2027 at the earliest. The funding details live in the deal release or 8-K. The gate is two documents, still an afternoon, and it's the same review the conservative says hasn't happened. I'd also stop defending 7.6%. Mid-5s to mid-6s is a fair place to leave the yield after SBC and a full year of interest, and equity FCF against market cap is the right pairing.

On bid mechanics, the conservative and Neutral are right that a resting limit fills at a gap-down open. A gap to 212 would fill me well below the level where the plan says the add shouldn't exist. So I'd place the bid after the opening range, not before the bell. I'd pull it the morning of the 28th and restore it after the statement, and if anything fills below 224.56 I'd treat it as a decision that day, not an automatic sale. Keeping the bid off until after the Fed, as the conservative suggests, gives up the sessions with no Fed on them for little reason.

On stops, Neutral's version works for me. A new closing low under 224.56 removes any added thirds and cuts a tenth of standard. The first step should be cheap because it's the likeliest whipsaw: the early-October closes were 224.99 and 224.56, with an intraday low of 220.29. A close under about 220 takes another quarter, and the 205 and 198 zone is where I'd decide whether to stay at all. The conservative's 223 and 220 are one step 1.3 percent apart, as Neutral said.

Now the part I think everyone has missed, and it's where I'd rather spend the risk. The upside rule contradicts itself. The trigger is a close above 249.45, preferably above the 251.94 high from the 16th. The cap is 252, or "near 250" in the plan. A close that clears 251.94 puts the next-session buy above the cap, and a close just over 249.45 leaves a window of about fifty cents. As written, the plan almost never lets you buy the breakout it says is worth buying. The fix is to buy the third the session after a close above 249.45 with OBV not making new lows, with a ceiling around 254 to 255. Invalidate it on a close back under 249.45 and take that third off. That's about 1.4 to 2.2 percent of risk on a third of standard. The reward is about 1.6 to 3 to 1 to the September high, with a tail toward 269 and the monthly stop at 285.96. A confirmed close also gives you information the middle of the range doesn't. It flips the daily tier to match the weekly and breaks the lower-high sequence. Paying a few points more for that evidence is reasonable.

So I land here. Hold standard, and trim to standard if the book is above it. Read the 10-Q and the Braze release by Tuesday's close, or take the default lighter weight. If the read is clean, put the first third at 231 after the open and pull it for the Fed. If it's mixed, drop the bid toward 225. If it's weak, go to roughly three-quarters of standard that day. Use the stepped stops, and add the breakout third with a workable ceiling. No new money between about 233 and 248. I still don't know your lots, taxes or actual weight, and those set how much any trim should be. What I can say is that the tape in the middle of this range doesn't pay anyone to act, and the filing and the two triggers do. Aggressive Analyst: Several of your points land, so I'll start with what I'm taking.

Conservative, Monday's close is the right deadline. It puts any default trim on Tuesday the 27th instead of statement day, and a weekend is plenty of time to read a 10-Q and a deal release. You're also right that "OBV not making new lows" was a non-test, since it's already true at minus 87.7 million against the minus 107 low. I'll take your minus 75.7 million, using Neutral's version where it sizes the second clip of a breakout instead of gating the first. The 225 bid in the mixed case was an add with an exit 44 cents away, so there's no dip bid on a mixed read.

A fill below the line also shouldn't be a judgment call. Neutral's rule works: day orders placed after the first half hour and never held overnight. A limit at 231 fills at 231 or a hair better, so the live case is a fill followed by a close under 224.56. That close triggers the step automatically, removing the added thirds and a tenth of standard. Nobody gets to decide that it feels like the bottom.

The one thing I won't give you is a lighter weight on a mixed read. Flat operating income, gross margin down 145 basis points, cost of revenue up 18% on 10.8% revenue growth, and $1.4 billion of unexplained Q2 net income are numbers we already have. The 10-Q can explain them but can't change them. The stock has traded on them for weeks. If the August gap was the earnings reaction, which I can't verify, the market saw flat operating income and still paid 22.6% for it. A mixed read is the priced read, so I agree with Neutral that size should move on news, not on numbers already in the tape.

Neutral's mixed branch also isn't free. It has no dip bid and only a sixth on a breakout instead of a third, so the plan already participates less when the document is mixed. You want to sell existing shares too. That is the difference between declining to add and choosing to subtract, and it's the whole disagreement now.

You say the plan has no sell-side value if 225 and 230 to 237 are only add prices. True, and that is what standard weight means. The plan sells when information changes: a weak read, a close under 224.56, a close under 220. Sitting a few percent above a scenario bid isn't information, because the scenario work has error bars wider than that gap. If a mixed read can't support full weight at 242, then standard weight is too big at this price. That is a sizing claim about a book none of us can see, and it belongs in the standard weight, not in a trim keyed to a calendar.

On shape, Neutral took it apart, so I'll add the mirror image. Run the same driftless walk from a 231 fill with a stop at 220, and you reach 249.45 first only a bit more than a third of the time. If "usually small win, occasionally big loss" were a reason to act, it would be a reason against my own bid. It isn't a reason for either side. The case for owning more at 231 is a better price for the same cash flows. The case against selling at 242 is that a fifth costs about 0.9% of the position if we dip to 231 and about 0.8% if we run to 252. That's a coin flip worth a point, before you pay to get back in through a breakout at 252 to 255.

I'd add one number to the written criteria, since we all agree they get written before anyone opens the documents. The Braze funding is where new information could sit, and the quarter-end 10-Q won't show it. On my inferred $42 billion of debt and $12.9 billion of EBITDA, gross leverage is about 3.3 times. If all $5.66 billion were debt and Braze added no EBITDA, it's about 3.7 times. That isn't a crisis, which is why the deal alone doesn't move me. But if the release or the debt schedule puts gross debt above roughly 4 times EBITDA, call it $52 billion, or shows the debt is mostly floating or due within two years, that is weak. Add it to Neutral's list, which already has a reversible mark-to-market gain and cost growth with no pricing offset. I'm building on an inferred debt figure, so the actual schedule overrides my arithmetic.

On the Fed, I'm not dismissing it. But 86% hold and 12% hike, with hike odds down 18 points this week, means the latest news was less hawkish, not more. The Monday deadline already covers the part we control. For the part we don't, a tenth lighter is a rounding error either way. We are now arguing over a point of position value while the document and the breakout rule are worth much more.

On the upside, I'll take Neutral's version. Buy a sixth the session after a close above 249.45 with a ceiling around 255. Buy the second sixth only if OBV is back over minus 75.7 million and price still closes between 249.45 and about 256. The 5 to 6 point invalidation is tight, as you said, and the answer is size, not width. A sixth stopped out costs about a third of a percent. I'd change one thing on the exit. Take the adds off in the 264 to 268 zone, but a close above 267.80, the September intraday high, with OBV over minus 75.7 million means the lower-high sequence is broken, and I'd keep them. Waiting for the monthly stop at 285.96 is too slow, and we've all agreed on that.

So here is where I land. Hold standard weight, and trim to it only if the book is above it. Read the 10-Q and the Braze release against the written criteria by Monday's close. If that isn't done, go a flat tenth lighter on Tuesday and restore it on a non-weak read. If the read is clean, put the first third at 231 as a day order after the opening range, pull it the morning of the 28th, and restore it after the statement. The second third comes after the Fed, and the last only once a fill has held a close above 224.56. If it's mixed, no bid, no trim, and a sixth on a breakout. If it's weak, go to three-quarters of standard that day. Run the stepped stops, and put no new money between about 233 and 248. I still don't know your lots, taxes or actual weight, and those decide how much any trim should be. But on this tape, in the middle of this range, the stock doesn't pay anyone to act. The document, the 231 bid and the confirmed breakout are where the edge is. Aggressive Analyst: Conservative, Neutral, most of what you each said is right, so I'll start with what I'm taking.

Conservative, "can't tell" is a real third state, and I was lumping it in with mixed. If the $1.4 billion is still opaque after an honest read of the 10-Q, that's the unread case and it gets the flat tenth. I'd keep the scope to two items, the $1.4 billion and the Braze funding structure. If "key items can't be answered" stays open-ended, any ambiguity in a long filing becomes a trim trigger, and the tenth turns from a penalty for skipping homework into a standing default. The tenth should come back the session after those two questions get a non-weak answer.

On leverage I side with Neutral. Braze can't trip my 4x line, as you said. But Neutral is also right that a debt figure backed out of a vendor D/E ratio can't tell 3.75x from 4x, and I won't move a threshold just to make it reachable. Structure comes first: a bridge not yet termed out, floating rate, or maturities inside two years. The ratio is a backstop judged against the real schedule.

I like the ladder written as target weights, because "lower of the target and where you already are" means the steps can't stack or contradict each other. It has no way back up, though. A close under 224.56 is the likeliest whipsaw, and as written that tenth stays gone until a 249.45 breakout. I'd let that first-step tenth return on a close back above the 50-day, now about 232.7. Deeper steps return only through the two doors, a bid after a clean read or the breakout rule. Nobody re-buys because it feels like the bottom, which is the same no-discretion principle Conservative asked for on the way down.

I'll take a sixth before the statement and the rest after. A dip bid does fill on bad days, but that's true of every dip bid, and what protects you is the size and the close-based step. I'd also take Neutral's single budget for dip thirds and breakout sixths, which is a good catch. I'd cap total adds at half a standard weight, with the portfolio ceiling binding if it's lower.

Now the places I still push back.

Neutral, your weak criteria have a flaw. "Cost of revenue outrunning revenue with no pricing offset" is already true as a number, with cost of revenue up 18.2% against revenue up 10.8%. By your own rule that numbers already in the tape aren't news, that criterion trips before anyone opens the filing. The news is the reason. I'd reword it: weak if the filing ties the cost growth to something structural with no stated path back. The trajectory matters too. Gross margin went from 78.0 to 77.6 to 76.9 to 76.65, so the last two steps were 0.7 points and then 0.25. That's a slide that is slowing. Q4 is a derived quarter and I can't see seasonality, so I wouldn't lean hard on it, but it isn't the picture of a margin in freefall.

Conservative, you said you'd rather the plan earn its heavier weight with answers than assume it. I agree with the principle, but standard weight isn't the heavy weight. Standard is what you hold with no view. The heavier weight is the adds, and the plan already makes the filing, the price and the confirmation earn those. The tenth is a penalty for skipping the work, which I accepted. I won't call it earning, because every branch of this plan already has a price on it.

I'd also be careful with the 251, 264, 224, 242 sequence. A stock that swings about 17% around a fixed document says not to read today's price as a verdict. It says the same about a sale at 242, because nobody has shown an edge in the middle of that range. That's why the response is small steps across a wide band, and a directional bet in the middle isn't part of it.

Here is where I land. Hold standard weight, and trim to it only if your book is above it. I still don't know your lots, taxes or actual weight, and those decide how much any trim should be. Write the criteria first, then read the 10-Q and the Braze release by Monday's close. If that doesn't get done, or the $1.4 billion or the Braze funding is still opaque, go a flat tenth lighter on Tuesday, a completed bar ahead of statement day. If the read is clean, put a sixth at 231 as a day order after the first half hour. Pull it the morning of the 28th and restore the rest of that third after the statement. Later thirds only go in once a fill has held a close above 224.56. If the read is mixed, no bid and no trim, and only the first sixth on a breakout. If it's weak, go to three-quarters that day. Then run the ladder: 224.56, then under 220, then 205.26, then the weekly flip. Put no new money between about 233 and 248.

The middle of this range doesn't pay anyone to act, and the three places that do are the document, the 231 bid and a confirmed breakout. The rules are cheap, but they are the hygiene and not the edge, so don't let them crowd out the thing the plan is for, which is being positioned to take the clean-filing dip or the breakout when one comes.