Quick Summary
- 1The expensive part of picking wrong isn't the invoice — it's losing 6–9 months and restarting with someone else.
- 2How a vendor handles discovery tells you more than their portfolio. Fast quotes without deep questions almost always mean rework later.
- 3Insist on owning the source code, the cloud accounts, and the database — in writing, before kickoff.
- 4Talk to two reference clients on the phone. Ask what went wrong, not what went well — that's where the truth lives.
I've sat on both sides of this conversation — pitching as a vendor, and helping founders pick between three shortlisted agencies. The pattern that decides whether the project ships well has very little to do with price and almost everything to do with how the team thinks before they write code.
What follows is the checklist we genuinely use when a client asks "should we hire these people?" — not a marketing version of it.
Where most selection processes go wrong
The usual shortlist gets built from three signals: a low quote, a fast timeline, and a glossy portfolio. None of those are predictive. We've inherited rebuilds from companies whose previous vendor delivered on time, on budget, and produced code so tangled the only realistic option was starting over.
The signals that actually correlate with project success are quieter: how the vendor handles ambiguity, who is in the room during scoping, and what they say "no" to.
The 10 questions that actually matter
1. Are they trying to understand the business, or just the feature list?
A weak partner asks "what features do you want?" A useful one asks "who is this for, what does success look like in six months, and what's the cost of doing nothing?" If nobody on their side asks about your customers, your unit economics, or your team's appetite for change, the build is already drifting.
2. Have they actually shipped in your domain?
Fintech projects fail without RBI/PCI awareness. Healthcare needs HIPAA-style PHI handling and audit trails. Logistics needs offline-first thinking. Domain familiarity isn't optional — it shows up in the questions they ask in week one and the bugs you don't hit in month four. Ask for case studies in your specific vertical, not "we've done apps."
3. Do they treat discovery as paid, structured work?
A serious team will spend 1–2 weeks on a paid discovery sprint — stakeholder interviews, workflow mapping, risk register, prioritised scope, and a budget range with confidence levels. If you get a fixed-price quote with a fixed timeline after a 30-minute call, you're being sold optimism. The first invoice from that engagement is almost always a change request.
4. What does communication look like in week six?
Anyone can do good comms in week one. Ask specifically: who runs the standup, who you can WhatsApp on a Sunday if production is down, what tooling they use (Linear/Jira/Notion), and what the weekly demo cadence looks like. If the answer is "our PM will keep you posted," translate that as "you'll find out about delays at the demo."
5. How do they handle code quality under deadline pressure?
Ask three concrete things: branching model, test coverage targets, and what happens when a sprint slips. Teams that say "we ship Friday no matter what" are the ones whose code you'll be paying to fix in 18 months. Look for CI/CD pipelines, PR review culture, and at minimum integration tests on critical flows. Storybook for the design system is a nice signal for frontend-heavy work.
6. Who owns the code, the cloud account, and the data?
This goes in the contract before kickoff, not "we'll figure it out at handover." You should own: the GitHub/GitLab org, the AWS/GCP/Azure root account, the domain, the database, the CI secrets, and every third-party service signed up using your email. We've seen vendors hold deployment access hostage during payment disputes — handle this on day zero.
7. Pricing model — and what it incentivises
Pure fixed-price encourages cutting scope quietly. Pure time-and-materials encourages slow weeks. The model that works for most mid-sized projects is fixed-cost per milestone with a clear change-request process, and a small T&M buffer (10–15%) for the edge cases nobody can scope upfront. Be wary of vendors who refuse to share blended hourly rates — it usually means a wide gap between quoted seniority and assigned seniority.
8. Who actually shows up after the contract is signed?
The classic bait-and-switch: the senior architect runs the sales call, then a 1-year-experience developer runs the project. Ask for named CVs, GitHub profiles, and a written commitment on minimum seniority for the lead engineer. Ask about their attrition rate over the last 12 months — anything above 25% means your project lead might leave mid-build.
9. What does "secure by default" actually mean to them?
You're looking for specifics: secrets in a vault (not .env files in git), least-privilege IAM roles, encrypted backups with tested restore drills, dependency scanning in CI, and an answer to "how would you respond to a credential leak at 2am?" If they wave a "we follow best practices" answer, they don't. Our cloud solutions write-up covers the baseline we set up on day one.
10. Reference calls — done properly
Don't accept written testimonials. Get on a 20-minute call with two past clients and ask: "What did they get wrong? What would you do differently? Did the team you started with finish the project?" The answers will be more candid than anything on a case-study page.
Planning a Website? Don't Overpay or Underbuild
Most businesses overspend on features they don't need — or underspend and rebuild within a year. We help you scope it right from day one.
The "fast quote trap" — and how to spot it
If a vendor sends a detailed quote within 48 hours of your first call, they aren't planning, they're pattern-matching to a previous project. That works fine if your project happens to look like the previous one. It rarely does. Pay for a discovery phase, get a real estimate, and treat the early invoice as the cheapest insurance you'll buy on the project.
What working with a good partner actually feels like
- They push back on requirements that won't move the metric you actually care about.
- They tell you when an off-the-shelf tool would do the job for 1/10th the cost — and lose the sale honestly.
- They produce architecture decision records (ADRs) so the next team doesn't have to reverse-engineer choices.
- They proactively flag scope creep instead of silently absorbing it and missing the deadline.
You're hiring a thinking partner, not a code-typing service. The clearest sign you've chosen well: by month three, they're disagreeing with you on something important — politely, with evidence, and usually correctly.
Related reading
Pair this with our India website cost guide for budgeting context, and the custom vs SaaS framework if you're still deciding whether to build at all. The portfolio has a few of the projects referenced above.
For businesses based in Gujarat, we've also published a local guide covering the top software development companies in Ahmedabad and what to look for when hiring locally.
How we work at ZANISS SOFTWARES
Our default engagement starts with a paid 1-week discovery sprint, a written scope with prioritised cuts, and a named senior engineer who stays on the project end-to-end. contact us if you want a second opinion on a quote you've already received — usually we can spot the rough edges in 30 minutes.
Pro Insight
Ready to Build a Website That Generates Leads?
At ZANISS SOFTWARES, we don't just build websites — we build growth systems.
- ✓SEO-first architecture
- ✓Conversion-focused design
- ✓High-speed performance
- ✓Scalable, future-proof code
📩 Response within 24 hours
