Technology cases are software, platforms, and hardware product companies. They are not a telco ARPU case and not “should we put our ERP in the cloud.”
Last full pass: 24 August 2026.
A technology case interview in 2026 (McKinsey, BCG, Bain, Bain tech, BCG Platinion strategy rounds, or a product-strategy boutique) is usually a recurring-revenue problem: seats, usage, net revenue retention, and cloud gross margin. The client sells software or a platform. If the client is a cable operator, a streamer, or a mobile network, you want the TMT case interview. If the client is a bank replacing a core system, you want digital transformation. Those three prompts get mashed together in candidate brains; interviewers notice.
What you are deciding
- Packaging and price: seat vs usage vs platform fee; discounting to win a logo.
- Growth vs NRR: buying new ACV while existing customers shrink on price is not growth.
- Build vs buy vs ecosystem: rarely a science project; it is time-to-coverage and gross margin.
- Cloud vs on-prem mix: COGS and the sales motion both change.
Clarify ARR vs bookings vs billings vs collected cash. Clarify land-and-expand vs one-year contracts. Clarify who the buyer is (CIO vs line vs developer). Ask whether a price cut on new logos will be MFN’d to the installed base — in enterprise SaaS, sales leaks. Ask if COGS is hosting plus inference (an AI feature can turn an 80% gross-margin story into a usage-tax).
Open with: (1) ARR bridge and NRR/GRR, (2) packaging (seat vs usage), (3) sales efficiency (magic number / payback), (4) gross margin, (5) implementation risk (time-to-value, which drives churn). A CPG-style “cut list price, volume up” is how you shrink ARR on purpose.
Exhibits you should expect
| Exhibit | Meaning |
|---|---|
| ARR bridge (new, expansion, churn, downsell) | NRR lives here. Logo count can rise while ARR falls. |
| Magic-number / CAC payback | Sales efficiency. A 0.6 magic number is a factory that eats cash. |
| Gross margin (cloud COGS, support) | “Software is 80% margin” is false for usage-heavy AI wrappers. |
| Usage vs seats | If usage is the value metric, seat cuts can raise consumption. |
| Cohort retention | B2B: logo vs dollar retention. Do not mix them. |
Units that trip people
ARR vs bookings. Bookings can spike on a three-year prepaid deal; ARR is the annualized run-rate. Mixing them inflates “growth.”
NRR vs GRR. NRR includes expansion; GRR does not. A 118% NRR with 82% GRR is a company being saved by a few whales.
MAU on a B2B product is a product metric, not revenue, unless you are an ads platform — which is TMT-adjacent. For SaaS, talk paid seats and NRR.
Practice a technology-industry case
Run a scored SaaS case where ARR, NRR, and discounting cannot be hand-waved as “growth.”
Worked mini-case: discount the list, shrink the ARR
Prompt. Northgrid sells workflow SaaS. 4,000 customers, ACV $21,000, ARR $84.0m. Dollar NRR 108%, logo churn 9%. Sales wants a 12% list-price cut to win 400 extra logos this year. Existing customers will demand the same price at renewal (the CRO already leaked it on a call). Cloud COGS is 22% of ARR. Do you cut?
Math. If only new logos got $18,480 ACV: +$7.39m ARR, existing $84.0m → $91.4m. That is the slide sales will draw.
If the cut flows through the base at renewal (assume one-year contracts, so the whole book reprices this year): 4,400 × $18,480 = $81.3m ARR. You added logos and lost $2.7m of ARR. NRR on the old book becomes a downsell: 4,000 × 18,480 / 84,000,000 = 88% before any real churn. Gross profit dollars: 0.78 × 81.3 = $63.4m vs 0.78 × 84.0 = $65.5m.
Recommendation. No across-the-board cut. If a segment is price-sensitive (SMB, new geography), create a feature-gated SKU at $18k that does not reprice Enterprise, and put a 24-month price-lock in the MSA so Sales cannot leak it. Watch NRR, not logo count. Risk: a usage-based competitor; answer that with a usage SKU, not a list-price panic. Next exhibit: ACV and NRR by segment.
What a generic profitability tree misses here
- Volume is not seats if you just cut price on the installed base. ARR is price × seats × expansion.
- Churn is a revenue and CAC problem; replacing a lost logo is not free.
- COGS is cloud and support, not factories — but it is still variable with usage and AI inference.
- Multi-year bookings can mask a rotting NRR.
- Platform effects (developers, marketplaces) do not show up in a 3-box cost tree.
If you treat Northgrid like a supermarket, you will cut sticker price, celebrate 400 logos, and miss a shrinking ARR.
See where you stand on a technology case
Practice ARR bridges and NRR so-whats.
Related guides
- TMT case interview (media, telco, ads)
- Digital transformation case interview (change programs)
- Fintech case interview
CoachNed is independent and unaffiliated with McKinsey, BCG, Bain, or other firms named for interview-style practice. Cases on CoachNed are AI-simulated.