Overview

This update revisits the original July 2026 analysis and brings it current for September 2026. The question remains the same: which systematic screens—momentum, value, quality or a hybrid—give the best tradeoff of upside capture, drawdown protection and implementability for mid‑cap software stocks in today’s market? Since July 2026, three practical shifts matter for investors: wider dispersion within the mid‑cap cohort as AI adoption becomes more differentiated; higher trading volatility and episodic liquidity in several names; and increased regulatory and customer scrutiny of AI features that affect revenue durability. This article explains what changed, what to watch, and concrete screening and execution rules to use right now.

Background: What led to the mid‑cap bifurcation

From 2020 through 2025, investor capital and narrative focus flowed heavily into companies that either built AI platforms or quickly integrated generative AI features. That created a clear bifurcation: a small set of platform winners aggregated multiples and distribution, while a wide middle of mid‑cap software firms experienced divergent outcomes—some monetized AI and improved retention, others saw momentum evaporate when feature launches didn’t translate to ARR growth.

Through 2026 the re‑rating matured into selective re‑ratings: investors now prize demonstrable AI revenue, healthy unit economics and stable retention over feature announcements alone. At the same time, macro volatility—higher-for-longer rates earlier in the decade and episodic risk‑off days—has compressed multiples and amplified the importance of liquidity and execution controls.

Data and evidence: What’s different in Sept 2026

Key empirical observations for September 2026 that should shape screens:

  • Dispersion increased among mid‑cap software margins and retention. Cross‑section volatility of gross margins and net revenue retention (NRR) widened relative to 2021–2023, meaning quality metrics better separate durable winners from headline‑driven names.
  • Liquidity is patchy. Several previously liquid mid‑caps show intermittent drops in average daily volume during drawdowns—practical implication: raise minimum ADV thresholds and model slippage conservatively.
  • AI revenue disclosure matters. Firms that disclose AI‑related revenue or ARR growth tied to AI product lines tend to have more stable post‑announcement price behavior than firms that issue vague “AI roadmap” statements; investors increasingly require transparent ARRs or AI‑revenue percentages to trust the narrative.
  • M&A and consolidation picked up modestly. Larger platform companies have been acquiring mid‑cap capabilities to fill product gaps; acquisition premiums have, in some cases, validated quality screens that emphasized margin expansion and customer retention.

These patterns are supported by industry surveys and practitioner reports through mid‑2026 that emphasize customer renewal as the most reliable predictor of AI monetization success (see recent consulting firm and data‑vendor briefings). For investors, the signal is clear: emphasize hard, repeatable operating metrics rather than product roadmaps alone.

Multiple perspectives: how managers view the post‑AI opportunity

Fund managers and quant teams have adjusted:

  • Quant/Factor teams: Many have layered quality and liquidity constraints onto momentum screens. Momentum alone produced strong short‑term gains in 2024–2025, but 2026 drawdowns exposed single‑name risk. Quant managers are lowering turnover targets and increasing minimum ADV filters.
  • Fundamental growth investors: They focus on AI monetization evidence—ARR uplift, feature‑linked pricing, cross‑sell uplift—and are willing to pay higher multiples for clear economics, not for roadmap statements.
  • Value/contrarian investors: Some see more opportunities after multiple compression in 2025–26, but caution that “cheap” mid‑caps can remain inexpensive if churn or product‑market fit problems persist.

Across the board, practitioners emphasize execution: precise data normalization for ARR/NRR, scenario testing for slippage, and tighter position sizing in illiquid names.

Updated screening philosophies and practical rules (Sept 2026)

The three core philosophies—momentum, value, quality—remain useful, but their implementations should change to reflect 2026 realities.

1. Momentum (updated)

  • Why use it: Momentum still captures re‑ratings driven by tangible customer adoption or strong guidance revisions.
  • Updated rules: 6‑month total return in top 35% of universe; positive 3‑month EPS or ARR revision trend; minimum 3‑month ADV ≥ $7.5M (raise for larger position sizes).
  • Execution: Reduce turnover by combining momentum entry with a quality gate to avoid one‑day reversal risk.

2. Value (updated)

  • Why use it: Multiple compression since 2024 made value filters more productive, but value traps persist.
  • Updated rules: EV/Revenue below historical peer median adjusted for growth deceleration; require explicit path to positive free cash flow within 12–24 months or demonstrated FCF improvement quarter‑over‑quarter; add a minimum NRR threshold (≥100%) to avoid durable revenue decline traps.

3. Quality (updated)

  • Why use it: With AI features proliferating, quality metrics better separate durable businesses.
  • Updated rules: NRR ≥ 105% preferred (≥100% minimum), gross margin ≥ 55% (allow 55% threshold for vertical SaaS), FCF conversion > 0 over trailing 12 months or improving trend, evidence of AI monetization (explicit AI‑revenue disclosure or feature‑linked pricing).

Hybrid approach: updated example screen for Sept 2026

  1. Universe: US‑listed software companies, market cap $2B–$30B, 3‑month ADV ≥ $7.5M; exclude names with material audit or regulatory flags.
  2. Momentum: 6‑month total return top 35% and positive 3‑month ARR/EPS revisions.
  3. Quality: NRR ≥ 105% or gross margin ≥ 55% and FCF conversion improving quarter‑over‑quarter.
  4. AI disclosure: company must disclose AI‑related revenue or ARR guidance explicitly tied to AI features, or show measurable upsell to existing customers attributable to AI (as discussed on earnings calls).
  5. Valuation guardrail: EV/Revenue below 75th percentile of the filtered set; prefer diluted exposure to names trading at >10x EV/Rev unless quality metrics are exceptional.
  6. Position sizing & liquidity: max 2.5% per name, portfolio 15–25 names; simulate 2–3% slippage on entry for names near the ADV threshold; rebalance monthly but review around earnings.

Execution mechanics and updated pitfalls

Practical adjustments investors must make in 2026:

  • Data hygiene: Normalize ARR and NRR definitions across vendors; require reconciliations between GAAP revenue and company ARR disclosures where possible.
  • Model slippage: Increase slippage assumptions in backtests to account for episodic illiquidity; use historical intraday VWAP data for realistic execution cost estimates.
  • Regulatory & disclosure risk: Watch for evolving AI disclosure requirements; companies that can’t substantiate AI‑revenue claims faced sharper short‑term drawdowns in 2026 when auditors or analysts pressed for clarity.

Implications for investors

For retail investors and advisors: the hybrid approach remains the most practical. But in 2026 you should tighten liquidity and quality filters, require demonstrable AI monetization signals, and reduce single‑name exposures. For quant teams: momentum still matters, but low‑turnover hybrid strategies have outperformed pure high‑turnover momentum once transaction costs and episodic slippage are modeled.

Outlook: what to watch over the next 6–12 months

  • AI revenue disclosure standards: regulators and exchanges may press for clearer AI revenue reporting; clearer disclosures will re‑rate some names and penalize vague claims.
  • M&A activity: watch for targeted acquisitions from platforms—these can reprice mid‑caps quickly and are a key idiosyncratic risk/opportunity.
  • Customer renewals and pricing power: durable NRR and pricing expansion will likely remain the clearest signals of successful AI monetization.
  • Macro sensitivity: if rates fall materially, growth multiples could expand again—screens that capture optionality (momentum + quality) will be advantaged.

Practical checklist to implement today

  • Start with the hybrid screen above and run a month of paper trades to validate execution assumptions.
  • Raise minimum ADV to ≥ $7.5M (or higher for larger accounts) and simulate slippage using intraday data.
  • Require at least one hard operating proof point for AI monetization (ARR disclosure, AI‑linked pricing, or measurable upsell tied to AI).
  • Cap single‑name exposure at 2–2.5% and maintain 15–25 positions for a balance of concentration and diversification.
  • Rebalance monthly, but perform event‑driven reviews around earnings and material regulatory developments.

Bottom line

Since the July 2026 publication, the post‑AI re‑rating has shifted from a broad narrative to a narrower, evidence‑driven re‑pricing. Momentum still captures repricings, but only a layer of quality and clear AI monetization signals makes those gains durable in 2026’s higher‑volatility environment. Updated screens should raise liquidity thresholds, insist on transparent AI revenue evidence, and tighten position sizing. With careful execution and disciplined data hygiene, a hybrid momentum‑plus‑quality approach remains the most practical way for investors to gain exposure to mid‑cap software while managing single‑name and sentiment risk.

Frequently asked questions

How should I treat AI‑related revenue in screens?

Prefer explicit AI‑revenue or ARR disclosures. If a company does not break out AI revenue, look for quantifiable proxies: feature‑linked pricing, documented upsell to existing customers attributable to AI, or analyst‑verified ARR uplift. Treat vague roadmap language as insufficient for valuation uplift.

What liquidity threshold is appropriate now?

For retail accounts, raise the 3‑month ADV minimum to at least $7.5M. For larger accounts or when targeting position sizes >2%, prefer $10M+ ADV. Always simulate slippage using intraday VWAP to model realistic execution costs.

Should I chase momentum names that announce AI partnerships?

Not automatically. Use a momentum entry combined with quality gates (NRR, gross margins, FCF trend). Partnership announcements without attributable revenue or customer adoption often see short‑lived spikes followed by mean reversion.

How frequently should I rebalance these screens?

Monthly rebalancing is a practical default. Monitor around earnings and material AI‑related announcements; allow tactical adjustments but avoid excessive turnover that erodes returns through transaction costs.

Can a value‑only approach work in this market?

It can, but it requires deeper fundamental work to avoid value traps. In mid‑cap software, cheap multiples often reflect churn or product issues. Combine value filters with at least one quality metric (NRR or FCF conversion) to improve hit rate.