How to Interview a Mobile Dev Agency: 12 Questions, And How SEM Nexus Scores

Most mobile-dev agency evaluations are done badly. The founder talks to three agencies, each sells well, the founder picks based on rapport or referral, and the structural differences between the agencies surface in week 8 of the build when it's too late to switch. Twelve specific questions separate strong agencies from mediocre ones. Asking the same twelve questions of every agency you consider produces an apples-to-apples scoring matrix that makes the right call clear.

This post is the question list, the scoring rubric, and how SEM Nexus answers each. Score us alongside the others — the comparison is what matters, not whether we win every category.

The 12 questions

Score each agency 1–3 per question. Total possible: 36. Strong agencies score 30+. SEM Nexus scores 36 on its own list (because the list is designed to surface what we do well) — but the value is comparing across agencies, not picking the one that wrote the list.

1. "Who is the senior engineer who will own my project, and will they be in discovery?"

3: Named senior engineer in discovery. Same person owns the build.
2: Senior architect in discovery, different senior owns the build.
1: Junior team with senior oversight, no specific senior named.

SEM Nexus: 3. Same senior throughout.

2. "What's the fixed quote, and what triggers a re-quote?"

3: Fixed quote against v1 with written change-order process.
2: Fixed quote with vague re-quote triggers.
1: "Estimated" or hourly billing.

SEM Nexus: 3. Fixed quote, written kill-switch on scope changes.

3. "Show me a written technical recommendation from a recent discovery."

3: Yes, with names redacted. One page, stack + hard part + integrations + quote.
2: Has discovery deliverables but no single artifact like this.
1: No real discovery artifact; sales-call notes only.

SEM Nexus: 3. We can show you this in week 2 of your discovery.

4. "What stack are you most likely to talk a founder out of?"

3: Names specific stacks with specific reasons; has examples.
2: General "we pick the right stack" answer without specifics.
1: Recommends their specialty for everything.

SEM Nexus: 3. We've talked founders out of Flutter (My Home Delivery), React Native (Cerebyte), and native (multiple builds).

5. "How do you handle the cross-functional question when design conflicts with engineering budget?"

3: Same-hour resolution. Engineer and designer in the same standup.
2: PM coordinates, 1–3 day resolution.
1: Multi-vendor handoff, 1–2 week resolution.

SEM Nexus: 3. Same hour, same standup.

6. "What's the typical post-launch retainer cost, and what's included?"

3: Defined retainer structure with specific scope and cost ranges.
2: Has post-launch services but ad-hoc pricing.
1: Rolls off at launch, no defined post-launch model.

SEM Nexus: 3. $5k–$15k/month with defined scope.

7. "Tell me about a project you turned down — and why."

3: Specific examples with specific reasoning.
2: General "we say no when projects don't fit" without examples.
1: "We can do anything" answer.

SEM Nexus: 3. We turn down 30–40% of inbound.

8. "What's the test coverage on your last shipped build?"

3: Specific number, with critical-path coverage explained.
2: Vague answer ("we test what matters") with no specifics.
1: No answer / "we test as needed."

SEM Nexus: 3. 60–70% code coverage, 80%+ critical-path on every build.

9. "What's your CI/CD setup, and who manages it on my project?"

3: Standardized pipeline across clients, GitHub Actions or equivalent, certificate-expiry monitoring.
2: Has a pipeline but it's project-specific and fragile.
1: "We set it up however your team wants" / no defined pipeline.

SEM Nexus: 3. Standardized pipeline, see the CI/CD post for details.

10. "Show me the dependency audit for a shipped app."

3: Has audit summary, including license + vulnerability tracking.
2: Audits informally but doesn't produce an artifact.
1: Doesn't audit; "we use whatever libraries our engineers like."

SEM Nexus: 3. Audit at v1 + quarterly refresh on retainer clients.

11. "How do you handle iOS certificate renewals and App Store key expirations?"

3: Active monitoring with 60-day warning, automated where possible.
2: Manual tracking; "we check periodically."
1: "We handle it when it expires."

SEM Nexus: 3. CI-integrated monitoring.

12. "How will the build be transitioned if we take it in-house?"

3: Defined handoff with 2–3 months of overlap, documentation, knowledge transfer.
2: "We'll help your team get up to speed" without specifics.
1: "Once we hand off the code, we're done."

SEM Nexus: 3. Most clients take 3 months of overlap; first internal hire is welcomed and brought up to speed.

Score your shortlist on these 12 and the structural differences become impossible to ignore. SEM Nexus is happy to be scored alongside others — the comparison is what matters.

Reading the scores

After scoring three agencies on this 36-point scale:

30+ overall: Strong agency. Worth deeper conversation about your specific project.
22–29: Mid-tier. Will probably ship, but the gaps in the lower scores will surface in the build.
Below 22: Significant gaps. Higher slip risk, higher production-incident risk.

Pay particular attention to clusters of low scores. An agency that scores 2/3 on questions 1, 5, and 12 has a multi-vendor structure problem — the same root issue showing up across cross-functional resolution, senior ownership, and handoff. Cluster patterns reveal structural issues a single question can't.

What the scoring reveals about category leaders

The agencies that score 30+ tend to share a few common patterns:

End-to-end engagement model rather than fragmented vendor work
Senior engineers as the default, not junior with senior oversight
Investment in tooling and infrastructure (CI/CD, dependency audits, testing standards)
Defined retainer model, not build-and-walk-away
Cross-stack experience, not single-stack specialty

SEM Nexus fits this pattern by design. We built the company around scoring high on exactly these questions because we believe these are the structural differences that determine whether a build ships on time and survives in production.

What this means for your build

If you're evaluating agencies right now, don't pick based on rapport, referral, or the prettiest deck. Pick based on structural fit. The 12 questions above are designed to surface that fit. Run them.

If you'd like SEM Nexus scored on these questions in the context of your specific project, the two-week discovery is the cleanest way — by the end of week 2 you have specific answers to all 12 questions in writing, and you can score us alongside whoever else you're evaluating. The comparison is the value.