IHP — IntegraFin Holdings plc

IHP — IntegraFin Holdings plc (Transact) · deep-value analysis

Analysis date: 08/24/2026 · Exchange: LSE (Main Market, FTSE 250) · ISIN: GB00BD45SH49 Yahoo symbol: IHP.L · Reference price: … GBp (08/24/2026, yfinance) · Market cap: ~… Shares: 331.3 mil issued / 330.4 mil basic weighted average · Fiscal year: ends September 30 Currencies: reporting GBP (millions), price GBp (pence) — … = 100 GBp, any yield relative to market cap requires a ×100 conversion

Identification note (symbol collision). The bare symbol “IHP” doesn’t resolve on Yahoo to the tracker’s company: today’s first run of deep_data_pack.py produced a completely empty file, named “141752.” The historical NYSE ticker “IHP” was IHOP Corp (today Dine Brands, DIN). The company in the tracker (Sumar r309, Exchange=LSE, ISIN GB00BD45SH49) is IntegraFin Holdings plc, the operator of the investment platform for financial advisers Transact, listed under IHP.L. I added the collision to _BVB_ALIASES in deep_data_pack.py and regenerated the data pack. It’s NOT an SEC issuer — no CIK, no 10-K/20-F filings; the primary filings are RNS releases and annual reports from integrafin.co.uk, which I downloaded and read directly (Annual Report 2025, 144 pages; Annual Report 2024; Annual Report 2023; Half-year report at 03/31/2026; FY26 Q3 Trading Update from 07/14/2026). For this reason there’s no SEC-Filings\IHP\ folder and step 2 of the procedure doesn’t apply.

Executive summary

The thesis. IntegraFin operates Transact, the UK’s largest investment platform for financial advisers, built entirely on proprietary technology. The economic model has a rare simplicity: a monthly ad valorem fee applied to assets under administration (funds under direction, FUD), of which 99% of platform revenue is recurring (AR2025, p45). At 06/30/2026 FUD was …, … y/y, and the quarter’s average daily FUD — the real revenue base — was …, … (Q3 Trading Update, 07/14/2026). The company has zero financial debt, equity of …, ROE ~29%, and capex that in normal years is under …. Since December 2025 an efficiency program has been running that caps administrative expense growth at ~3%/year in FY26 and FY27, versus … in FY25 — and in the half ended 03/31/2026, growth was already capped at 4%, with PBT margin above 51%.

What’s wrong with the headline numbers. The consolidated balance sheet shows … of assets and … of liabilities, and the cash flow statement shows CFO of … on a net profit of …. Both are accounting artifacts: the life-insurance subsidiaries (ILUK, ILInt) write exclusively unit-linked policies, and IFRS requires the policies’ assets and liabilities to appear gross on the balance sheet, in the income statement AND in the cash flow statement, with a net effect of exactly zero (AR2025, p46). Of the … of CFO, 272.2 mil is the movement of client money — visible without doubt in the fact that total group cash grew by exactly …, and cash held for policies grew by exactly … (1,895.0 − 1,622.8), while corporate cash fell by … (243.9 vs 244.1). Consequently CFO − capex = … isn’t the flow to the shareholder, it’s 5.8× larger than it. The real flow, reconstructed from the shareholder’s liquid-asset roll-forward, is … in FY25 (see the valuation chapter). This is the largest FCF-bridge divergence encountered across the deep-report series — four times larger than the AFYA reference case (73%).

Estimated value. Five triangulated models give a wide range, centered very close to the current price: base DCF FCFE … (… versus …), bear DCF … (…), bull DCF … (…), Greenwald EPV … (…), 5-year historical multiples … (…). Model median: …. The Monte Carlo simulation across 20,000 scenarios (same central assumptions, dispersion on owner earnings ±25%, g1 σ=4.5pp, r σ=100bp, gt σ=50bp) gives a median intrinsic value of … and a probability of undervaluation of …% — practically a coin flip. The broker consensus target (428–…, 9–11 analysts) is 8… above the price, so the market doesn’t see a big discount either.

The verdict. A high-quality business — a real moat through switching costs for advisory firms, 99% recurring revenue, zero debt, ROE 29%, growing share of net market flows (≈25% in H1 FY26), and a pricing policy that has cut fees 18 times in 17 years, which widens the moat rather than filling it in. But the price leaves no margin of safety. At … the stock is at the top of its 52-week range (306–… on monthly closes), at 20.8x TTM profit and 5.6x book value, with FUD at an all-time high reached through a +… market move in a single quarter. Nearly half of the next two years’ profit growth is already in consensus and in the price. Verdict: WATCH, don’t buy at …. Threshold of interest: below ~… (FY26E P/E ~16x, base-case DCF MOS ≈ …), where the Monte Carlo range would move the probability of undervaluation above 60%.