11.11.26

How to Decide Between Contract and Permanent Hires for a Data or AI Transformation Programme

Most transformation programmes hit the same wall: you need a Databricks migration unblocked next month, and you need a permanent data engineering function that still exists in eighteen months. Treat those as one hiring decision and you'll get the wrong answer for at least one of them. Treat them as two parallel tracks with different logic, and you can solve for speed and durability at the same time. This guide walks through how to make that call properly — not as a formality handed to whichever recruiter answers the phone fastest, but as a decision that shapes how your programme actually delivers.

Prerequisites

Before you start splitting a hiring plan into contract and permanent tracks, get clear on:

• The delivery deadline that's driving urgency (a migration date, a go-live, a board commitment)

• Which parts of the work need to exist as in-house capability once the programme ends, and which are genuinely one-off

• Your current IR35 status determination process, or who owns it

• Budget mechanics — is contractor spend opex against a project budget, and is permanent headcount a separate approval cycle

• Whether you have a technical evaluator on your side who can sanity-check a specialist's Databricks, Snowflake, LLM, or Agentic AI experience, or whether you're relying on your recruitment partner for that

Step 1: Separate the "unblock now" problem from the "build capability" problem

Start by writing down, role by role, whether the need is about removing a delivery blocker or establishing a function. A live pipeline migration that's stalled because you don't have a Databricks-certified engineer this month is a delivery-blocker problem. A Head of Data role that will still exist after the migration ships is a capability-build problem. These call for different hiring instruments — d.Contract for the former, d.Perm or d.Engage for the latter — and conflating them is the single most common reason transformation hiring plans stall.

Don't default to "we'll just hire permanent for everything, it's cheaper long-term." A permanent search for a senior technical role can take weeks to source and vet properly, particularly for scarcer specialisms like LLM or Agentic AI engineering, and a live programme rarely has that runway sitting idle.

Step 2: Map your budget cycle against your delivery deadline

Contractor and permanent spend usually sit in different parts of your finance structure, and mismatching the hire to the budget line creates approval delays that defeat the whole point of moving fast. d.Contract engagements are typically framed against a defined assignment or interim capacity need, with contractor administration, performance tracking, and payroll support handled as part of the service, which suits project-budget-line spend. d.Perm and d.Engage searches are framed around fee, replacement provisions, and a hiring timetable agreed in the proposal — closer to a standard headcount approval.

Once you know which budget line each role sits in, you can start both tracks without waiting for one approval to clear before starting the other.

Step 3: Run IR35 as a design input, not an afterthought

IR35 status shapes what a contract role can actually look like — day rate, working pattern, and whether the engagement is genuinely outside IR35 — and it needs to be decided before you brief a specialist recruiter, not after an offer is verbally agreed. Dcoded's d.Contract service advertises IR35 compliance support as part of contractor recruitment, alongside sourcing, vetting, and payroll administration. Where the requirement calls for a longer commitment tied to genuine control and substitution rights, that's a signal to test whether permanent is actually the better structural fit, rather than stretching a contract to cover a role that behaves like employment.

Treat any advertised rapid-mobilisation capability — including 24-hour start claims — as conditional on the specific brief, candidate availability, and completed checks, not as a guarantee you can build a plan around.

Step 4: Decide where retained search earns its keep

Not every permanent hire needs a retained search, but some do — and knowing which is a strategic call, not a cost-cutting one. d.Engage is built for business-critical appointments where the client needs a committed search partner: dedicated time, access to passive candidates who aren't actively job-hunting, and regular updates through a long-term search. That's the right instrument for a Head of AI, a VP Engineering, or a CTO-level hire where the market of genuinely qualified people is small and mostly not looking. A more conventional d.Perm engagement — covering requirement definition, sourcing, interview management, offer negotiation and onboarding — fits roles where the talent pool is wider even if the bar is still high.

Case evidence shows what a well-run permanent search under pressure looks like: a retailer's stalled direct search for a Head of Data role was filled within ten weeks once Dcoded engaged, with four shortlisted candidates surfaced within five days of starting. That's a useful benchmark for what "urgent but permanent" can realistically achieve — faster than a stalled internal process, but still measured in weeks, not days.

Step 5: Build the contract track for genuine speed, and hold it to that standard

The contract track exists to answer "who can start this week," and the evidence bar for that claim should be case-specific outcomes, not a blanket promise. When a financial services firm had a contractor candidate fail checks partway through a Lead Data Engineer search, Dcoded reports sourcing two replacement profiles, running two interviews, and reaching an accepted offer within 36 hours, with compliance checks completed on top of that. On an Azure SME requirement, Dcoded reports placement confirmation within 24 hours of the initial client conversation, using an established specialist relationship rather than a cold search. On a CEO-level urgent search for a venture-backed fintech, five profiles were surfaced within three hours, first-stage interviews ran within 24 hours, and the offer was accepted inside 48 hours, with a five-day total from brief to start.

The pattern across these cases is that speed comes from an existing, vetted relationship with the specialist, not from compressing the vetting itself. When you brief a contract role, ask directly whether your recruitment partner already has a relationship with someone who fits, or whether they're starting a cold search — the honest answer changes what timeline you should plan around.

Step 6: Run both tracks in parallel, with one person accountable for the join

The mistake most transformation programmes make isn't picking the wrong hiring model — it's running contract and permanent sequentially, treating the contractor as a stopgap until "the real hire" is sorted. That approach means your delivery risk sits exposed for longer than it needs to, and your capability-build track only starts once the urgent problem is already solved. Instead, brief both tracks in the same conversation: what needs solving in the next two to four weeks, and what needs to exist in the org chart in six months. A consultancy facing a live bid used exactly this logic at scale — Dcoded reports submitting eight Golang engineer profiles across two shortlisting rounds, with four offers made and accepted, directly supporting both bid credibility and ongoing delivery capacity in parallel.

Nominate one person on your side — a Head of Engineering or programme lead — to own the join between the two tracks, so that when the permanent hire lands, the handover from the contractor isn't an afterthought.

Step 7: Brief your recruitment partner as an advisor, not an order-taker

The quality of the contract-vs-permanent decision depends heavily on whether your recruitment partner pushes back on the brief or simply executes it. A brief that says "get me a contractor" when the underlying need is a permanent capability gap should prompt a conversation, not a shortlist. Dcoded structures its services around this distinction explicitly — d.Contract for individual specialists or larger temporary teams, d.Perm for lasting in-house capability, and d.Engage for business-critical searches requiring sustained commitment — precisely so the engagement model matches the actual need rather than the first instinct.

If your recruiter never asks whether a role should be permanent instead of contract, or vice versa, that's worth noticing.

Tips & Best Practices

• Brief both tracks in one conversation. Splitting the contract and permanent conversations across different meetings, or different stakeholders, is how the two tracks end up running sequentially instead of in parallel.

• Confirm IR35 status before advertising a rate. Getting this sequence backwards creates rework and delay exactly when you're trying to move fast.

• Ask for the specific milestone, not a general promise. "24-hour start" and "five days brief to placement" are case-specific outcomes tied to particular engagements — ask your recruitment partner what's realistic for your specific brief, not what's achievable in the best case.

• Use retained search sparingly and deliberately. d.Engage is built for genuinely business-critical, hard-to-fill roles — reserve it for searches where passive-candidate access is the actual constraint.

• Keep vetting technical, not just procedural. A brief for a Databricks or Agentic AI specialist needs someone on the recruiting side who can actually interrogate that experience, not just match keywords against a job spec.

Troubleshooting

"Our contractor track is fast, but the permanent track keeps stalling." Check whether the permanent search is retained (d.Engage) or standard (d.Perm) — a business-critical, senior, or scarce role often needs the dedicated commitment of a retained search rather than a standard permanent process.

"We keep converting contractors to permanent instead of running two searches." That's a sign the two tracks were never separated at the brief stage. Go back to Step 1 and re-classify each open role as delivery-blocker or capability-build.

"Compliance checks are slowing down our contract hires." Compliance and IR35 checks are a necessary part of the process, not a bolt-on — as the Lead Data Engineer case shows, even a rapid 36-hour offer acceptance still included completed compliance checks before the process was considered done.

Summary

Contract and permanent hiring aren't two versions of the same decision — they're two different tools solving two different problems, and a transformation programme usually needs both at once. Separate the delivery-blocker roles from the capability-build roles, match each to the right budget cycle and engagement model, get IR35 status settled before you brief anyone, and run both tracks in parallel under one accountable owner. We built d.Contract, d.Perm, and d.Engage as distinct services precisely because the strategy question — not just the execution — is where most hiring plans go wrong. Bring us the whole programme brief, not just the next vacancy, and we'll help you work out which track each role actually belongs on.