The Blog

Start with the unit that actually drives cost: the patient

Almost every trial budget is really a per-patient budget in disguise. Before you touch a spreadsheet, get three numbers straight: how many patients you expect to enroll (not screen), how many visits each patient completes across the full schedule, and what actually happens at each visit — the procedures, assessments, and staff time.

Once you have those, the per-patient cost is the sum of what each visit costs, multiplied across the visit schedule. Everything else is built on this foundation, so if the per-patient number is soft, the whole estimate is soft.

The cost categories you can't skip

• Per-patient (variable) costs — procedures, assessments, coordinator and investigator time, per-visit stipends. These scale directly with enrollment.
• Fixed startup costs — qualification, initiation, regulatory submissions, training, system setup. New sites routinely under-price this because it all happens before a single patient walks in.
• Pass-through costs — central lab kits, imaging, shipping, equipment, patient travel. Keeping these separate from your fee is what protects your margin later.
• Overhead — the institutional percentage applied on top. Forget it, and you've quietly agreed to absorb it.
• Screen failures — patients you assess but don't enroll still cost real time and procedures. A budget that only pays for enrolled patients leaves this on the floor.

Work an example

Say you're modeling 20 enrolled patients, 8 visits each, and roughly $600 of procedures and staff time per visit. That's about $4,800 per patient, or roughly $96,000 in variable cost. Add fixed startup — call it $25,000 for qualification, initiation, training, and setup — plus a pass-through line for labs and shipping. Layer overhead on the fee portion, add a screen-failure allowance, and the "roughly $96k" study is meaningfully larger — with every number defensible line by line.

That's the point: a budget isn't a single figure, it's a stack of assumptions. When someone challenges the total, you want to point at the specific assumption they're questioning instead of defending a black box.

Sponsor view vs. site view

One trial produces two legitimate budgets. A sponsor-level view rolls costs up across every site and the whole program. A site-level view is what a single site needs to negotiate its own contract. They rarely match — different overhead, different pass-through handling, different screen-failure assumptions — and that mismatch is usually where negotiations stall. Knowing which view you're building, and saying so explicitly, saves a lot of back-and-forth.

Pressure-test your assumptions

A good estimate comes with a range, not false precision. Before you send anything, ask: what happens if enrollment comes in 20% low? If the schedule adds an unscheduled visit per patient? If overhead is a few points higher? If small shifts blow up your number, you've found the assumptions worth nailing down first.

Get to a defensible number faster

You can build all of this by hand, and it's worth understanding every line. If you'd rather start from a structured estimate and adjust, our free Budget Estimator walks through these categories in both sponsor-level and site-level modes — the first estimate is free. It's an assumption-based planning tool, not a substitute for a formal sponsor budget or a CRO bid. To see exactly how it calculates, the methodology page lists every formula, multiplier, and sensitivity range.

What the questionnaire is really asking

Behind every question is the same worry: can this site actually enroll and run this study on time, cleanly? Sponsors have been burned by sites that over-promised on paper. So when they ask about your population, staff, or equipment, they're building a risk model of whether choosing you will cost them time and money. Answer with that in mind and your responses sharpen immediately.

Answer enrollment questions with evidence, not optimism

The most scrutinized section is patient access. "We see plenty of these patients" is exactly the answer that gets sites into trouble. Instead: give a realistic number rather than a best case — sponsors heavily discount inflated projections, and a site that consistently hits its numbers earns far more future work. Show where the patients come from (standing population, referral network, registry) so the number has a source. And note competing studies honestly; hiding them doesn't make them disappear, it just resurfaces later as an enrollment shortfall with your name on it.

Be specific about capability and capacity

Two sites can both tick "yes, we have the equipment." The one that wins describes it: what they have, whether it's calibrated and validated, who's trained on it, and what their current study load looks like. Capacity matters as much as capability — a site with the right equipment and no bandwidth is a risk. If you have genuine spare capacity, say so; it's a selling point.

Turn "experience" questions into proof

When asked about therapeutic-area experience, quantify it: number of similar studies, track record on timelines and data quality, relationships with the relevant patient community. Specificity reads as competence; vagueness reads as padding.

The mistakes that quietly cost sites studies

• Inflated enrollment projections — the fastest way to win a study once and never be invited back.
• Slow turnaround — feasibility often closes first-come; a great answer submitted late loses to a good answer submitted on time.
• Generic, copy-pasted responses that clearly weren't tailored to this protocol.
• Hiding weaknesses instead of framing how you'll manage them. Sponsors trust honest sites more than apparently flawless ones.

Build a reusable feasibility profile

Most sites re-answer the same 80% of questions from scratch every time. Keep a living site capability profile — population, staff, equipment, therapeutic experience, past performance — that you update once and reuse. Every questionnaire gets faster, more consistent, and higher quality, and you're never reconstructing your own numbers under a deadline.

That's the thinking behind our Site Network & feasibility work: a reusable, aggregate, non-PHI site profile you maintain once and present to sponsors and CROs, currently in early access. If you'd rather start from a document today, the SOP & template marketplace includes original feasibility and site-startup templates — and our free SIV checklist covers the visit that follows.

We launched the first version of AVD Clinical in January 2026 with three things: a resource library, a budget calculator, and a set of original SOPs. This is what's changed since then.

What shipped in Q2

• Medicine Reminders — a free, encrypted personal medication reminder tool. You enter your own medications; the data stays on your account, protected by row-level security, never shared. This is the first feature that is not a document download or a calculator — it's a light daily-use tool. No PHI accepted.
• Expanded SOP library — 100+ original SOPs, checklists and templates now available, including bundles for site startup, monitoring, QA, closeout, PHI handling, inspection readiness, validation, sponsor/CRO oversight, patient-facing workflows, and budget planning.
• Revised pricing — we simplified the Calculator to a monthly or annual unlimited plan and removed the confusing "lifetime" tier that was causing mismatched expectations. Current prices are always on the pricing page.
• New legal pages — we rewrote the Privacy Policy and Terms of Service to cover every product we actually offer (not just the library), added a proper Template License (EULA), a HIPAA & PHI Notice, a Refund Policy, and a Security page. The footer on every page now links to all of them.
• Platform Roadmap section — the homepage feature grid now shows a live/coming soon/enterprise roadmap label on every card, so you know exactly what you can use today versus what's planned.

What we changed based on feedback

A few things that looked fine in testing didn't hold up in practice. The calculator paywall was blocking the mode-switch tabs — that's fixed. Paid buttons weren't real links for crawlers — that's fixed. The homepage claimed features like AI eConsent and wearable integration as if they were live; they're now clearly labeled as Enterprise Roadmap items.

What's next

Role-based clinical accounts (investigator, sub-I, coordinator, pharmacist) with license verification are in design. AI-assisted SOP generation is in development. Certifications and on-demand courses are planned for Q4. We'll update here as things ship rather than on a forced monthly cadence.

Decentralized clinical trial (DCT) approaches — remote consent, home nursing visits, direct-to-patient drug shipment, wearable ePRO — went from novel to expected between 2020 and 2023. By 2025, most protocols at large sponsors included at least one DCT element. The question now is whether those elements are actually delivering on the original promise.

What's working

Remote consent and pre-screening have had the most consistent positive evidence. Sites using digital pre-screening funnels (not AI matching — just structured online eligibility forms) generally report fewer first-visit screen failures, because patients arrive already understanding the eligibility criteria. We have not measured this ourselves; treat it as a widely reported pattern, not a figure you can plan against.

Direct-to-patient drug shipment works well in low-complexity, oral medication trials. It reduces the burden on patients who would otherwise need to travel to site for drug pickup. The pharmacy logistics are manageable when the drug is stable and the patient monitoring requirements are light.

What isn't working as expected

Wearable ePRO sounds compelling in the protocol but creates real data quality headaches. Compliance rates drop sharply after week four. The data streams are large and require cleaning infrastructure most sites don't have. Regulatory acceptance of wearable-derived endpoints remains inconsistent across major regulators. Most sponsors who've run a wearable-heavy trial once are more cautious the second time.

Home nursing visits are expensive, logistically complex, and harder to quality-assure than anticipated. The cost per visit is often higher than bringing the patient to site, and the audit trail requirements for home visits are stricter than many vendors initially communicated.

Full decentralization — where there is no site at all — has not proven viable for most indications. Recent regulatory guidance on decentralized trials has been clear that removing the investigator from clinical oversight entirely creates problems that technology doesn't solve.

Where this lands for site teams

The most effective DCT implementations have been hybrid: keep the investigator relationship and the site as the anchor, add remote elements for visits that don't require in-person assessment. This is what a risk-based GCP framework actually supports — fit your monitoring intensity to the risk of the activity, not to an ideology about remote vs. in-person.

For coordinators and CRAs, the practical implication is that your SOPs and monitoring plans need to explicitly address which visits are eligible for remote conduct, what the audit trail requirements are for those visits, and how you handle a patient who loses connectivity or doesn't respond to remote prompts.

The current revision replaced the previous integrated addendum. Most coverage focused on the structural changes — the guidance is now organized differently with clearer principles vs. annexes — but the operational impact lands at three specific places in site work.

1. Quality by Design becomes a documented requirement

Under the older version, "quality by design" was an aspirational concept. Under the current revision, sites are expected to document how their procedures are designed to prevent the most likely errors before they happen, not just detect them after. For most sites this means revising your site SOP for protocol training to include an explicit risk discussion at the SIV — not just a topic checklist.

2. Risk-based monitoring is now expected, not optional

Sponsors operating under the current revision are expected to implement risk-based monitoring. This pushes pressure down to sites: full SDV across every data point is no longer the default. Sites need to be ready for monitoring visits that focus on critical data and processes rather than line-by-line review.

3. Electronic systems get sharper requirements

The current revision tightens the language around electronic data — audit trails, validated systems, controlled access. If your site uses a paper-electronic hybrid (still common in academic settings), you'll need to be explicit about which is the source of truth for each data point.

What to do today

• Review your site Quality Manual against the current quality-by-design language and update if needed
• Update protocol training documentation to capture risk discussion explicitly (not just topic checklists)
• Audit your hybrid paper-electronic workflows and document which is the source of truth per data element
• If you use validated electronic systems, confirm audit trail completeness against current expectations

Clinical research templates — protocol frameworks, study report skeletons, risk assessment tools, site initiation toolkits — are meant to be starting points, not finished documents. The teams that get the most value treat them as scaffolding they build on, not forms they fill in blindly.

The single most useful skill

Many industry templates are written for broad consensus, not first-time readers, so the actionable content is buried inside long preambles. Skip the preamble. Jump directly to the "Instructions for Use" or "User Guide" section — that's where the operational content lives.

How to use a template effectively

• Don't fill it in directly. Save a copy under your organization's name and modify there. The original stays as a clean reference point.
• Track your changes vs. the source. When auditors ask why a clause differs from the baseline, you need to know.
• Local regulatory adjustments come last. Complete the standard sections first, then add country/state-specific addenda at the end rather than modifying the body.
• Keep a change log. Record what you changed and when, so your team always knows which version is current.

Where AVD Clinical fits

AVD Clinical publishes its own original SOPs, checklists, and templates covering the trial lifecycle — site startup, monitoring, budgets, QA, safety, and closeout. Each one is a starting framework with placeholders you adapt to your protocol and route through your own QA review.

Clinical research has too many tools that promise everything and deliver nothing. We're trying not to be one of them. Here's where we actually are.

Live today

• Resource Library — curated regulatory templates with AVD commentary on each one. Free, with one-time sign-up.
• Budget Calculator — sponsor and site-level estimation modes. First calculation free.
• Original AVD SOPs — 25 starting frameworks covering site, CRO, and sponsor activities.

What's next

• More SOPs as common requests come in
• Calculator scenario comparison and PDF report exports
• Monthly newsletter on regulatory changes and platform progress

The platform vision (longer-term)

Basic email OTP sign-in is live today for Library downloads and Calculator access. The fuller platform vision — role-based clinical accounts, license verification, donation matching, regulated workflow dashboards, and audit logs — requires validated backend infrastructure and a development team. We are NOT going to fake those features while we build them.

What we're not building

• A clinical trial recruitment site — we don't enroll patients
• An EMR/EHR integration product — that's a Phase 5+ effort
• An AI clinical decision support tool — that's Software as a Medical Device territory

We'd rather have a small set of things that work than a big set of things that don't.