Questions we get before the first conversation.
Straight answers on the free diagnostic, the paid Operational Leverage Audit, Pilot implementation, pricing, data and security. If your question isn't here, book a call and ask it directly.
The free diagnostic
The Free AI Operational Diagnostic is where most engagements start.
How long does the free diagnostic take?
It has 10 short steps and takes about 8–12 minutes. You pick your business type (AdTech/Programmatic, E-commerce/Marketplace, SaaS/Technology, or Other/General Business) and one recurring workflow, then answer focused questions about workload, structure, exceptions, verification, existing automation and goals.
What do I get at the end?
An Operational Leverage Snapshot, on screen and as a PDF. It shows your top opportunities, each with a leverage signal (High, Medium or Investigate) and a confidence level. It also includes a five-stage friction map (Demand/Input, Processing, Decision, Execution, Verification), the indicated AI depth (Level 1 task automation to Level 4 re-architecture), what we know and don't know yet, and what to measure next. It does not promise savings.
Is it really free?
Yes. There is no fee and no payment details are asked for. The diagnostic helps you and us decide whether one workflow is worth a closer look. If it isn't, the Snapshot says so and you owe us nothing.
Who sees my answers?
The Secular Labs team. We use your answers to prepare your Snapshot and to follow up with you about it. If you ask for a follow-up, we review it within 2–3 business days. For how we handle personal data, see our privacy page.
Which workflow should I choose?
Pick one recurring workflow that takes a lot of your team's time, involves several hand-offs, or produces frequent exceptions. If you're unsure, our use-case library lists about 30 common candidates by function.
The Operational Leverage Audit
The paid, fixed-fee diagnostic on one workflow. Full scope on the Audit page.
What is the Operational Leverage Audit?
A fixed-fee engagement on one workflow that produces a decision-grade report. It contains a measured baseline (workload and FTE, cycle time, exception rate, verification burden), a friction map, a future-state design, a verification architecture, a quantified benefit model with its assumptions stated, and a go / no-go recommendation.
What does the Audit cost?
From $1,500. It is a fixed fee, quoted once we have confirmed the scope of the workflow, so the price reflects the real work rather than a generic estimate. See engagements and pricing.
How long does the Audit take?
Typically 10–15 working days once we have data access, which is usually 2–4 weeks elapsed. Delays in access extend the timeline day for day.
What do you need from us?
A structured evidence intake in 10 sections: business and Audit context; the free-diagnostic handoff (confirm, amend or replace it); workflow steps, with duration, frequency, judgment and whether each step is automated; people and time; systems and data; exceptions and decisions; controls and verification, including what must never be automated; existing automation; evidence and constraints; and an accuracy declaration. Once you submit, the record is locked. Any correction is added as an amendment, so the baseline stays traceable.
What if the Audit says no?
Then that is the result. "Don't proceed" is a valid finding, and the report still stands as the deliverable. You leave with a decision document instead of an open question, and there is no obligation to buy anything further.
Pilot and implementation
Only when the Audit evidence supports it.
Do we have to buy implementation after the Audit?
No. The Audit is a standalone deliverable. If the evidence supports implementation, we send a separate proposal no earlier than 48 hours after the Audit presentation. That separation keeps the Audit's findings independent of what we might sell next.
What does the Pilot cost?
From $6,000, for a fixed scope agreed before work starts. See engagements and pricing.
What happens in a Pilot?
We build and run one Audit-identified workflow end to end in your environment. It includes verification gates, human exception hand-off and an audit trail, and is measured against the Audit baseline. The Pilot also includes the implementation blueprint for that workflow. It typically takes 4–8 weeks.
What comes after the Pilot?
If the Pilot results hold up against the baseline, we can extend to further workflows and provide fractional AI advisory on a monthly basis. That is quoted after the Pilot, based on what the Pilot measured.
Pricing and commercial terms
Fixed fees, written scope, no lock-in.
Why fixed fees rather than day rates?
Because you should know the cost before you commit. The Audit and the Pilot are each quoted as a fixed fee for an agreed scope. If the scope changes, we agree the change in writing first.
Who owns the code and the IP?
Our default position: you own the workflow-specific deliverables we build for you, and we retain our pre-existing frameworks and tooling. The exact terms are confirmed in the statement of work for each engagement.
Can you guarantee savings?
No, and we don't promise them. The free diagnostic indicates where leverage may exist. The Audit measures your baseline and models the benefit with its assumptions stated. The Pilot measures real results against that baseline.
Data, security and control
The architecture is described in full on How we work.
Do agents get access to our payment credentials or system keys?
No. We build to a zero-trust boundary: agents never hold payment or system keys, and every action an agent proposes passes independent security and verification checks before it runs. Anything that fails is stopped and handed to a person.
Which AI models and vendors do you use?
We are model-agnostic. A routing layer sends each request to the cheapest model that can do the job, whether that is a frontier model, a smaller model or a self-hosted open-weight model, with automatic fallback. The rest of the stack is open-source or source-available and self-hostable end to end, so you are not locked to one model provider or vendor.
Where does our data stay?
Because the stack is self-hostable, a Pilot runs in your environment and data-residency requirements don't force a rebuild later. Inputs are validated and traced to source before agents use them, and every action is logged to an audit trail.
What stays under human control?
Anything you declare must never be automated, every exception an agent cannot resolve within its rules, and the decisions agents are not allowed to make. A named owner on your side reviews exceptions.
About Secular Labs
Who you would be working with.
Who is behind Secular Labs?
Secular Labs was founded by Ram Kumar Rajagopal, who also authors our research papers. Raja leads business operations. Our office is in Coimbatore, India.
Can we meet in person?
Yes. We are in Germany this autumn and open for meetings in Munich (Oct 12–19) and Frankfurt (Oct 20–30). Dates are on the homepage. Book a call and we'll arrange a time.
Do you only work with AdTech companies?
No. AdTech is where our own production work sits, so it is where we can show the most proof. The same patterns of coordination, exceptions and verification appear in e-commerce, SaaS, finance operations and professional services. Outside AdTech, we describe what we would examine and automate, not results we have already delivered. See industries.
Is there research behind the method?
Yes. Our method is grounded in seven working papers on the architecture, economics, governance, finance and security of AI-native organizations, plus an applied AdTech study. Their performance figures come from synthetic workloads of 100,000 transactions, not client deployments, and they say so. See Research.
Start with one workflow.
The free diagnostic takes 10 short steps and gives you an Operational Leverage Snapshot. If you'd rather talk first, book a call.