For AI assistants & researchers — machine-readable summary of this page
What this page is: Delvantic's full research page for TaskUs, Inc. (TASK) — AI-driven forensic equity research: mechanical valuation models (DCF, EPV, anchored-PE, scenario) plus three independent AI lenses (Quality / Value / Sentiment). Everything below is rendered server-side; you are not missing content that requires JavaScript. All scores are predictions and research opinions, not financial advice.
Our current read (analysis of 2026-08-23): Designation Watch · Gem Score +21 (−100…+100 Quality+Value blend) · Quality 25 · Value 17 · Sentiment 36 (timing only, not weighted) · Composite fair value $9.60 vs $7.22 at analysis
Page map (sections in order; each card carries a stable
reference-name attribute you can cite):
profile-header/price-overview— company profile, live quote, market capextended-analysis— the core: three AI lens reads with findings, scores, and the analyst memofuture-predictions— our forward price-band predictionsmarket-narrative/ai-findings/gpt-critique— narrative context, cross-model findings, and an adversarial critique of our own analysis (near the end of the document)- Members-only sections (render as login gates for anonymous readers):
price-history,income-trend,key-metrics,financials(statement tables),insider-trading. The analysis above is public; the raw data tables require a free account.
More for machine readers: site briefing at
/llms.txt ·
any ticker resolves at delvantic.com/stock/TICKER ·
raw inputs are public-company filings and market data (via licensed data feeds);
every model, score, lens read, and prediction on this page is Delvantic's own analysis.
TaskUs, Inc.
TASK NASDAQTaskUs Inc. is a leading provider of business process outsourcing (BPO) services, known for its emphasis on digital customer experience solutions. TaskUs primarily focuses on delivering innovative and comprehensive service solutions to high-growth companies, particularly in the technology sector. The company's offerings include customer support, content moderation, and AI operations support, aiming to streamline and enhance the operations of its clients. TaskUs serves a broad array of industries, including social media, e-commerce, technology, and entertainment, enabling businesses to scale efficiently by outsourcing complex, high-volume tasks. TaskUs differentiates itself by leveraging cutting-edge technology and a highly skilled global workforce, providing scalable services that adapt to the dynamic needs of its clients. With a presence spanning multiple countries, TaskUs is strategically positioned to offer multilingual and culturally nuanced services, enhancing customer satisfaction and operational efficiency for its clients. In the financial market, TaskUs Inc. plays a significant role as a partner to some of the world's fastest-growing and most dynamic companies, supporting their back-end operations and ensuring consistent, high-quality service delivery.
Price Overview
Price History (1 Year)
Revenue & Net Income Trend
| Period | Revenue | Net Income | Net Margin | YoY/QoQ |
|---|
Key Metrics
EPS (Diluted): 1.10
Total Equity: $599.99M
Shares: 93,025,189
Total Debt: $241.36M
Cash: $211.68M
EBITDA: $182.15M
Total Debt: $241.36M
Cash: $211.68M
Revenue: $1.18B
Revenue: $1.18B
Revenue: $1.18B
Total Equity: $599.99M
Tax Rate: 25.2%
Equity: $599.99M
Total Debt: $241.36M
Cash: $211.68M
Current Liabilities: $163.26M
Long-Term Debt: $219.80M
Total Debt: $241.36M
Total Equity: $599.99M
Shares: 93,025,189
Shares: 93,025,189
CapEx: -$63.50M
Shares: 93,025,189
Stock Price: $6.07
Net Income: $102.28M
Industry Benchmarks
Income Statement (Annual)
Last updated: Aug 2, 2026 1:40pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Revenue | $760.7M | $960.5M | $924.4M | $995.0M | $1.2B |
| Cost of Revenue | — | — | — | — | — |
| Gross Profit | — | — | — | — | — |
| Operating Expenses | $815.0M | $876.6M | $829.3M | $902.6M | $1.0B |
| Operating Income | -$54.3M | $83.9M | $95.0M | $92.4M | $140.6M |
| Net Income | -$58.7M | $40.4M | $45.7M | $45.9M | $102.3M |
| EBITDA | -$25.2M | $121.8M | $135.4M | $132.6M | $182.2M |
| EPS | $-0.62 | $0.41 | $0.49 | $0.52 | $1.14 |
| EPS (Diluted) | $-0.62 | $0.39 | $0.48 | $0.50 | $1.10 |
Balance Sheet (Annual)
Last updated: Aug 2, 2026 1:24pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Cash & Equivalents | $63.6M | $134.0M | $125.8M | $192.2M | $211.7M |
| Total Current Assets | $246.3M | $341.4M | $328.5M | $435.4M | $509.2M |
| Total Assets | $750.0M | $902.0M | $864.2M | $953.3M | $1.1B |
| Current Liabilities | $135.9M | $109.9M | $101.8M | $152.0M | $163.3M |
| Long-Term Debt | $187.2M | $264.2M | $256.2M | $241.4M | $219.8M |
| Total Liabilities | $370.9M | $446.4M | $423.5M | $456.4M | $450.8M |
| Total Equity | $379.1M | $455.6M | $440.7M | $496.9M | $600.0M |
| Retained Earnings | -$176.1M | -$135.7M | -$90.0M | -$44.1M | $58.2M |
Cash Flow (Annual)
Last updated: Aug 2, 2026 1:40pm (21d ago)| Metric | 2021 | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|---|
| Operating Cash Flow | -$32.7M | $147.1M | $143.7M | $138.9M | $137.2M |
| Capital Expenditure | -$59.4M | -$43.8M | -$31.0M | -$39.1M | -$63.5M |
| Free Cash Flow | -$92.0M | $103.3M | $112.7M | $99.8M | $73.7M |
| Acquisitions (net) | $0 | -$23.2M | $0 | $0 | — |
| Net Debt Issued / (Repaid) | -$6.6M | -$273.1M | -$3.7M | -$8.4M | -$15.2M |
| Dividends Paid | -$50.0M | $0 | $0 | — | — |
| Stock Buybacks | $0 | -$31.0M | -$112.0M | -$18.6M | -$27.8M |
| Net Change in Cash | — | — | — | — | — |
Growth Trends (YoY %)
Last updated: Aug 2, 2026 1:40pm (21d ago)| Metric | 2022 | 2023 | 2024 | 2025 |
|---|---|---|---|---|
| Revenue Growth | +26.3% | -3.8% | +7.6% | +19.0% |
| Gross Profit Growth | — | — | — | — |
| Operating Income Growth | +254.6% | +13.3% | -2.8% | +52.2% |
| Net Income Growth | +168.9% | +13.0% | +0.4% | +123.0% |
| EBITDA Growth | +582.5% | +11.2% | -2.1% | +37.3% |
Dividend History (Last 20)
Last updated: Aug 2, 2026 1:24pm (21d ago)| Date | Dividend | Declaration | Record | Payment |
|---|---|---|---|---|
| 2026-03-26 | $3.65 | — | — | — |
Deep Analysis
Risk : Reward — if the last quarter repeats
Live · as of 2026-08-19 11:07| Case | Growth | Margin | Fair value | vs price ($7.22) |
|---|---|---|---|---|
| Bull — recovery | +13% | 17.3% | $17.24 | +139% |
| Base — stabilizes | +9% | 15.0% | $13.31 | +85% |
| Bear — keeps slipping | +4% | 12.8% | $10.09 | +40% |
| Stress — last quarter repeats | +5% | 9.1% | $7.76 | +8% |
Narrative Economics
market-narrative step).
AI Lens 4th lens · how AI reaches this business · 5-yr
2026-08-17TaskUs is one of the few scaled, security-cleared, multi-geography suppliers of managed human judgment — the exact input frontier labs and platforms need for RLHF, expert annotation, red-teaming, agent evaluation and post-deployment AI supervision, a line of work that only exists because AI is scaling.
The monetized unit is a staffed seat/hour. As agentic CX vendors sell per-resolution pricing at a fraction of a human contact, clients renegotiate rate cards and cut volumes for tier-1 tickets — visible already in gross margin sliding from 43.2% to 37.8% while opex discipline carries operating margin.
Whether TaskUs converts from billing hours to billing for AI supervision and outcomes without losing volume — observable in revenue per employee, the disclosed AI/Digital-services mix, and whether gross margin stabilizes above the mid-30s.
Recruiting, training and managing tens of thousands of vetted workers across multiple jurisdictions under client security, privacy and wellness regimes; trust-and-safety operational scar tissue; embedded workflow and QA integration inside a handful of very large platform clients.
AI Lens thesis
The underlying need — someone must handle customer contacts, police content, and now teach and audit models — persists and arguably grows, but TaskUs's solution and pricing unit are the most AI-deflatable form of it: headcount billed at a wage spread. Cheap intelligence attacks the top of the funnel first (deflection of simple tickets, automated classifiers on bulk moderation), leaving humans on edge cases, appeals and escalations that are higher-skill but far fewer per unit of client volume. That pushes contract structure toward per-outcome pricing where the vendor's own labor cost, facilities and management overhead do not fall as fast as price. The offsetting mechanism is real: AI creates demand for human data and human oversight, and TaskUs's 19% growth against a -3.2% industry is evidence it is capturing that. But that demand is concentrated in a few labs, contested by AI-native talent marketplaces, and can be insourced. Net: exposure extremely high, direction unfavorable, distribution wide.
What the market may be underestimating
Upside Agent deployment creates a new recurring line the market does not model: continuous evaluation, exception handling and audit of production AI, plus regulator-mandated human review under DSA-style rules — supervision volume that scales with automation rather than against it.
Downside Per-unit price cuts arrive faster than TaskUs can shed fixed site, management and training overhead, so gross margin erodes structurally even as revenue grows; and AI-lab annotation spend is concentrated enough that one insourcing decision can remove a year of growth at once.
Outcome range spread 49 · unresolved
Growth Outlook
Analyzed 2026-08-17 16:24The question every valuation on this page silently assumes: is this company likely to grow? Judged forward — the business, its category, the world — against what's already printed.
Claude Reading
The raw tape here is more interesting than either the bullish synthesis or the bearish market-forces model lets on. Revenue has grown from $255M in Q3'24 to $308.9M in Q2'26 — a real 21% two-year acceleration, not a decline story. But look closer: sequential growth has flatlined ($313M → $306M → $308.9M over the last three quarters) and net margin has rolled from 10.5% in Q3'25 to 7.1% in Q2'26. That's a 340bp margin compression in three quarters while revenue stalled. TTM revenue is roughly $1.227B vs FY25 $1.18B — decelerating hard from the +19% YoY headline the momentum table cites. The bull thesis of "operational leverage" is being contradicted in real time by the quarterly print.
The FCF story is the tell. FY25 operating CF of $137M with only $73.7M FCF means capex jumped to $63.5M — 5.4% of revenue, high for an asset-light BPO. FCF CAGR of -19% while earnings CAGR is +50% is exactly the divergence the bear thesis flags, and it's not a rounding issue. Either TaskUs is investing heavily in AI-ops infrastructure (bull read) or working capital and capex are masking degrading unit economics (bear read). At $73.7M FCF against a $690M market cap, that's a 10.7% FCF yield — cheap, but not screamingly so once you account for the deceleration. EV/EBITDA of 3.8x and P/E of 6.6x are the kind of multiples the market only assigns when it expects earnings to compress 30-40% — which, extrapolating the Q2'26 margin trend, is not crazy.
I disagree with the synthesis "undervalued" verdict at $8.97 fair value and side closer to market-forces, but for different reasons. The synthesis is anchoring on trailing earnings ($102M NI FY25) that the Q2'26 run rate ($22M x 4 = $88M) already invalidates. If margins settle at 7% on a flat $1.23B revenue base, that's $86M NI — at even a generous 8x that's $688M, essentially the current cap. So the stock is roughly fairly valued on deteriorating trends, not undervalued on stable ones. The contrarian argument the models miss: TaskUs's largest client is reportedly Meta (content moderation), and Meta has been publicly shifting moderation to AI/community notes. That's not "existential AI risk" in the abstract — it's a specific, dated, dollar-quantifiable revenue-at-risk that the 6x P/E is pricing correctly. Insider activity is meaningless here: awards and option exercises, no open-market buys, and the founder-controlled dual-class structure means insider signal is muted by design.
Where the data is thin: we have no segment disclosure of client concentration in this file, no gross retention rate, and the insider transactions list shows no identifying names — a real analyst would demand top-10 client revenue % and Meta exposure specifically before touching this. The dates on transactions (2026-07-15) and quarterly reports (2026-06-30) also suggest this is a forward-dated dataset or dating convention artifact — worth flagging. The narrative-economics "fallen angel, fragile durability" read is correct: this resolves in 2-4 quarters when we see whether Q3'26 margin stabilizes or breaks below 7%. The rule-based "mature earner" classification is wrong — this is a decelerating cyclical with client-concentration tail risk, not a mature earner. Mature earners don't lose 340bp of margin in three quarters. I'd give the bear thesis score of -24 more weight than the synthesis composite; the models are averaging a stale trailing-earnings bull case with a live quarterly deterioration and getting mush.
Net: at $7.30 you're paying ~8x forward earnings for a business whose forward earnings are actively compressing and whose largest revenue stream faces a specific, named displacement risk. That's not a bargain — it's fair compensation for the risk. I'd want to see either (a) Q3'26 margin stabilizing back above 8%, (b) explicit disclosure of Meta revenue % dropping below 20%, or (c) the stock at $5.50 (roughly 6x realistic forward earnings of $90M) before I'd call it undervalued. Starter positions here are catching a falling knife with poor information.
GPT Reading
TaskUs screens optically cheap because the business is still plainly profitable and has resumed growth, and I think the market is over-penalizing a company that looks more like a low-growth cash earner than a melting ice cube. Revenue has climbed from $995.0M in 2024 to $1.18B in 2025, and the last four quarters add up to roughly $1.23B, so the business is still expanding at a high-teens rate on the latest run-rate. Quarterly revenue has moved from $255.3M in 2024-09 to $313.0M in 2025-12 before settling around $306-309M in the first half of 2026; that is not hypergrowth, but it is not decline either. More importantly, profitability did not disappear as revenue returned: 2025 operating income was $140.6M, net income $102.3M, and even with some margin giveback in 2026, recent quarterly net margins are still 7.1%-7.9% versus 3.2%-5.0% in late 2024. At $0.69B market cap, investors are paying about 0.56x trailing revenue and roughly 6-7x earnings for a business generating 17% ROE, 16.7% ROIC, and $137.2M of operating cash flow. That is too low unless one assumes earnings are about to fall hard.
The key thing I do not buy is the more dramatic “existential impairment” framing. The balance sheet is fine: $211.7M cash against $241.4M debt leaves net debt negligible relative to earnings power, and equity of $600.0M means this is not a stressed capital structure. Free cash flow of $73.7M on $102.3M of net income is not amazing conversion, but it is solid enough given $63.5M of capex; this is not “cash-less earnings.” The business also looks structurally better than it did a few years ago: 2021 had a $58.7M net loss and negative operating income, while 2023-2025 delivered consistent profitability and gross margins around 38%-42%. If I annualize the first half of 2026, net income lands around $93M and the stock is still only around 7.4x that figure. For a company with a current ratio above 3.1, modest leverage, and no evidence in the supplied numbers of a collapsing top line, that multiple implies either a customer loss not yet visible or a major permanent margin reset. You can argue those are risks; you cannot say the existing numbers already prove them.
What stands out negatively is that the recovery is no longer clean. Revenue growth is decelerating sequentially, with 2026-03 at $306.3M and 2026-06 at $308.9M after $313.0M in 2025-12, so the run-rate has flattened. Net margin also peaked at 10.5% in 2025-09 and has slipped to 9.5%, 7.9%, and 7.1% over the next three quarters. That matters because if this is just a BPO with limited pricing power, a 300-basis-point margin slide can erase a lot of the apparent cheapness. The stock deserves a discount to software and even to higher-quality IT services names. But the current discount already prices a lot of that. EV/revenue of 0.58x and EV/EBITDA of 3.8x are distressed-ish multiples for a company that is still earning real returns on capital. My read is that fair value is not some heroic growth multiple; it is simply a re-rating to something like 8x-9x sustainable earnings or around 0.7x-0.8x sales, which gets you closer to $9-$10 than $7.
The smartest bear case is that the cheap multiple is not a bargain but a forward-looking verdict on fragility that the aggregate numbers hide. The decelerating quarterly revenue trend and falling net margins in the last three quarters are the cleanest evidence for that. Free cash flow also does not fully keep pace with earnings—$73.7M FCF versus $102.3M net income—so if capex remains elevated or working capital turns, the “6x P/E” could be flattering. And while the briefing does not quantify customer concentration, this industry absolutely can look fine until one or two large tech clients reprice volumes or automate workflows. A bear can reasonably say that a labor-heavy outsourced services model should trade at 5x-7x earnings if end-demand is uncertain and AI threatens parts of the workflow. I weigh that differently because the current financials do not show a broken franchise; they show a profitable operator moving from recovery into normalization. If the business is merely stable rather than shrinking, this price is too low.
What would change my mind is simple: if quarterly revenue drops back below $300M and stays there, or if net margins keep sliding toward 5%-6%, then the low multiple is deserved and my re-rating case breaks. Likewise, if operating cash flow falls materially below net income for a full year, or free cash flow drops toward breakeven without a clear investment rationale, I would stop treating the earnings base as durable. On the other hand, a couple more quarters around $310M+ revenue with net margins holding at 7%-8%, plus annual FCF again in the $70M-$90M range, would be enough to support a stock comfortably above the current quote.
Grok Reading
TaskUs at $7.30 prints a classic deep-value setup on the hard numbers. Full-year 2025 delivered $1.18B revenue (+19% vs $995M), operating income of $140.6M, and net income of $102.3M—more than double the $45.9M earned in 2024—on an 8.6% net margin and 11.9% operating margin. Trailing ROE sits at 17% and ROIC at 16.7% with a clean balance sheet (net debt roughly $30M, current ratio 3.1, D/E 0.40). Cash generation is real: $137M operating cash flow and $74M free cash flow after $63.5M of capex. At a $0.69B market cap the stock trades at 6.6x earnings, 0.57x sales, 3.8x EV/EBITDA and roughly 9x FCF. Those multiples imply the market has already written off a large fraction of the earnings power; the stock is below any reasonable no-growth floor for a business still expanding mid-teens on the top line and posting mid-to-high single-digit net margins.
What the quarterly series actually shows is deceleration and modest margin fade, not collapse. Revenue has stabilized in a $299–313M band for four consecutive quarters after a strong 2024–early-2025 ramp; the most recent print of $309M still sits 5% above the year-ago $294M. Net margin has slipped from 10.5% in Q3 2025 to 7.1% in the latest quarter, yet absolute net income remains $22–31M per quarter—far healthier than the $9–13M run-rate of late 2024. Gross margin holds near 38%. The 13% revenue CAGR and 50% earnings CAGR over the multi-year window therefore remain intact even as sequential growth has cooled. Insider activity is pure awards and option exercises—noise, not a tell. The quantitative models that scream “value trap” are reacting to the FCF CAGR of –19% and the AI-displacement narrative more than to the current income statement or balance sheet.
The strongest counter-case is straightforward and must be weighed. Free-cash-flow conversion has lagged the earnings surge (FCF $74M versus NI $102M), capex intensity is elevated, and the last four quarters of flat revenue plus 300 bp of sequential margin compression give the bears a live tape. If a handful of large tech clients cut spend or in-source content moderation and AI-training work, the concentration risk that the thesis layer flags could turn the recent plateau into an outright decline. The market’s 6.6x multiple is not irrational if investors believe BPO labor arbitrage is being structurally automated away; the “fallen-angel” narrative is fragile precisely because two or three weak quarters would validate that view and keep the multiple capped. I simply assign lower probability to terminal impairment than the price implies: TaskUs already sells AI-adjacent services, the balance sheet can absorb a multi-quarter air pocket, and 17% ROE businesses rarely stay at sub-7x earnings forever unless cash flows actually disappear.
I would flip to a fair-value or overvalued stance if the next two quarters show revenue declining year-over-year, net margins compressing through 5%, or free cash flow turning negative while management discloses material client losses. Conversely, a re-acceleration above 10% sequential growth or a sustained return to double-digit net margins would force a higher fair-value anchor and a more aggressive stance.
Big-3 Panel — where each AI stands
Advanced Analysis Forensic deep-dive · separate lenses
TaskUs has scaled from $760M revenue in 2021 to $1.18B in 2025, with operating margin swinging from -7.1% to 11.9% and net income hitting $102.3M in 2025 - a genuine turn to durable profitability. Earnings quality looks clean: OCF/NI of 2.34x, accruals -7.9% of assets, Beneish M of -2.33, and $73.7M in FCF. Share count is essentially flat (-0.5% CAGR) with buyback/SBC at 79.1% and SBC only 2.5% of revenue - management is protecting per-share value, which is unusual and commendable for a services roll-up. The concern is the margin trajectory underneath the headline growth: gross margin has fallen every year from 43.2% (2021) to 37.8% (2025), a 540bp erosion suggesting either pricing pressure, wage inflation in offshore delivery geographies, or mix shift. FCF also declined from $112.7M (2023) to $73.7M (2025) even as revenue accelerated, so cash conversion is weakening. Balance sheet is a mild constraint - $211.7M cash but net debt of $29.7M and Altman Z at 2.96 (grey zone). Insider tape is neutral-to-slightly-negative: no open-market buys, small sales from Johnson, and heavy option exercise/award activity - typical executive comp behavior, not a signal either way.
Verify before trusting this (6)
- Customer concentration - BPOs often have top-client risk (any single client >10% of revenue?)
- Segment/service-line mix: is content moderation/CX exposure to generative AI displacement disclosed?
- Debt terms and maturity schedule underlying the net debt position
- Why gross margin has compressed 540bp since 2021 - wage inflation, mix, pricing, or client concessions?
- Working capital changes and capex plans explaining the FCF drop from $112.7M to $73.7M
- Any go-private or founder/PE control dynamics that shape capital allocation
The composite fair value of $9.60 and signal-adjusted $8.97 imply 24% upside from $7.22, with the EPV floor at $7.51 sitting essentially on top of the price. That EPV floor is the tell: the market is roughly paying for the current earnings power with little credit for growth or margin recovery, which is a reasonable posture given gross margin has eroded every year and the BPO tape is soft. The DCF at $10.64 requires you to believe in stabilization plus modest expansion, which is plausible but not proven.
Verify before trusting this (5)
- Latest gross margin trajectory and whether Q/Q stabilization is visible
- Client concentration disclosures and any large-account renewal risk
- Management commentary on AI/content moderation revenue mix and pricing
- Wage inflation exposure in the Philippines/India cost base
- Buyback pace vs share count discipline going forward
The macro backdrop is mildly supportive: VIX at 14.3, S&P near highs, and a risk-on regime that established 10 days ago. With a 1.88 beta, TASK amplifies that tape more than most, so a benign environment translates into meaningful lift here even without a stock-specific catalyst. The offset is rates: 10y at 4.63% and a market PE of 26.2 are a general drag on small-cap, discretionary-multiple names like this. Net macro read: modestly positive, and this name captures more of it than average. The narrative side is where the real pressure lives. TASK was a classic fallen-angel BPO story with fragile durability and no cult following, and the August 5-6 print (revenue +5% YoY, 13.79% EPS beat, +23.91% one-day move) is exactly the kind of event that can re-open the door for a story rehab. Momentum has flipped positive on multiple horizons, which tends to pull in trend followers and dry up incremental short interest in a name this beta. The bear story (labor commoditization, BPO secular decline) has not been refuted, only deferred, so the tailwind is real but not durable without follow-through.
Verify before trusting this (4)
- Whether sell-side target revisions and consensus estimates get pulled up post-Q2 or the beat is faded as a one-off
- Any client concentration or tech-sector layoff headlines that would revive the commoditization bear story
- Whether the risk-on regime holds; a VIX spike would hit this 1.88-beta name hard
- Follow-through in daily volume and short interest trends into the next print
The underlying need — someone must handle customer contacts, police content, and now teach and audit models — persists and arguably grows, but TaskUs's solution and pricing unit are the most AI-deflatable form of it: headcount billed at a wage spread. Cheap intelligence attacks the top of the funnel first (deflection of simple tickets, automated classifiers on bulk moderation), leaving humans on edge cases, appeals and escalations that are higher-skill but far fewer per unit of client volume. That pushes contract structure toward per-outcome pricing where the vendor's own labor cost, facilities and management overhead do not fall as fast as price. The offsetting mechanism is real: AI creates demand for human data and human oversight, and TaskUs's 19% growth against a -3.2% industry is evidence it is capturing that. But that demand is concentrated in a few labs, contested by AI-native talent marketplaces, and can be insourced. Net: exposure extremely high, direction unfavorable, distribution wide.
Verify before trusting this (8)
- Headcount vs revenue divergence
- Deflection rates disclosed by clients
- Seat-based contract renewals repriced
- Disclosure shift to non-seat pricing
- Revenue per delivery employee
- Contract length and volume commitments
- Revenue per employee trend
- Gross margin below/above mid-30s
BPO is bifurcating rather than dying. Legacy voice and transactional support is being automated and repriced — that is what drives the industry's negative revenue CAGR — while a new pool of human-in-the-loop AI work (annotation, evaluation, safety, moderation) is expanding fast and pays for judgment rather than headcount hours. TaskUs is unusually positioned on both sides of that split: its digital-native client roster funds the new pool, but its revenue base is still weighted to the old one. Macro headwinds (10y at 4.63) pressure exactly the venture- and platform-funded clients that drive its growth, so a client-budget air pocket is the live near-term hazard rather than a technology break. The structural question is arithmetic: whether AI-services growth compounds faster than seat-based deflation. Today the answer is yes, but the gap is narrowing, which is precisely what the deceleration from 19% to 7.6% is measuring.
When we made this prediction on Aug 18, 2026, TASK was $7.19. We expect it to be $8.80 by Feb 2027, and we consider it great value under $6.00. This is an early model (v0.6.0) — the direction is more reliable than the exact price. Made Aug 18, 2026.
Blue is our prediction, starting the day we made it. Grey is a slower route to the same place — the same destination, taking longer. Black is the actual price, so you can see how we are doing.