Free
$0every month
2 monthly calls
refreshes monthly
Enough to turn your next couple of discovery calls into cited client deliverables.
- No credit card
- Same scope brief, email, and issue-stub outputs
- Calls refresh monthly
Paste a call. Get the scope, the email, and the issues.
Scopebase turns a discovery-call transcript into a scope brief, a follow-up email, and Linear-ready issue stubs — every line cited back to the call. Built for whoever owns the 60–90 minutes of writing after every client call.
source / transcriptpaste-only / stays in your browser
Sign up free to run it — 2 calls every month, no card.
See each one on a sample call:
Nothing runs until you sign up. Your paste stays in this browser tab — we carry it into your first call, and it’s never uploaded until you start the run.
2 calls free every month · No credit card · Nothing runs until you sign up · See it on a sample call
01 / the wedgerecall is not delivery
Meeting notes answer “What happened?” Scopebase answers “What do we owe the client?”
After every call comes 60–90 minutes of unbillable writing: the scope the client can sign, the email that goes out today, the tickets your team picks up tomorrow. Your note-taker’s summary stops at recall. Scopebase collapses the writing into a paste and a review pass.
The lead file matches the call. Discovery and scoping calls open with the scope brief; working sessions open with a decision log. You get the email and the tickets either way.
one paste returns
3 files per call- scope-brief.md
Scope briefdiscovery
The scope of work your client signs — decisions, assumptions, and open questions kept separate, never blended into a summary blob.
- decision-log.md
Decision logworking session
What was decided, who committed to what, and what stayed open: the internal record that keeps working sessions from evaporating.
- follow-up-email.md
Follow-up email
The cover note you send the same afternoon: what was decided, what happens next, who owes what.
- issue-stubs.md
Issue stubs
Linear-ready tickets with acceptance criteria, scoped so your team can pick them up tomorrow.
every line cited back to the transcript
02 / proofruns in your browser
See a finished call packet
Below are fictional sample calls, already processed. Each packet leads with the file that matches the call — a scope brief on discovery calls, a decision log on working sessions — plus the email and the issue stubs. Every claim links back to the transcript line it came from.
pick a call scenario
runs in your browser — nothing is uploadedFictional sample transcripts, shown so you can see what each document looks like and how citations behave. Loaded: Customer portal rebuild — discovery call — A focused discovery call with clear decisions, requirements, and next steps. The best-case input.
Extracted:4 open questions5 requirements3 assumptions4 decisions4 risks2 implementation items3 follow-ups
Scope risks & unresolved items
11 flagged — settle these before you commit to a price- Open questionWhat does the portal need to do on day one for this to count as a win?
- Open questionHow many active customer accounts are we talking about?
- AssumptionAround 1,200 active accounts, and maybe 80 concurrent users on a busy morning.
- RiskOne concern from our end: the legacy authentication service is a single point of failure, and the documentation for it is thin.
- RiskThat's the riskiest integration in this project.
- RiskThe team that built it is gone, so there are unknowns there for us too.
- AssumptionI'm assuming the API can return shipment events in under two seconds.
- AssumptionThat assumption is probably safe for status lookups, but bulk invoice queries are slow today.
- Open questionCan we get read access to a staging copy of the API before kickoff?
- Open questionWhat's the timeline pressure on your side?
- RiskLeadership wants something customer-facing before the October board meeting, so the timeline is tight.
Source transcript
Customer portal rebuild — discovery call- 1
Sam: Thanks for making time, Priya. I want to leave today with enough detail to write a scope brief your leadership can react to.
- 2
Priya: Sounds good. The short version is that our customer portal is ten years old and our ops team spends half their day answering questions the portal should answer.
- 3
Sam: What does the portal need to do on day one for this to count as a win?
- 4
Priya: Customers need to see live shipment status without calling us. That is the single biggest driver of support tickets.
- 5
Priya: We also need self-serve invoice downloads. Finance has to stop emailing PDFs by hand.
- 6
Sam: How many active customer accounts are we talking about?
- 7
Priya: Around 1,200 active accounts, and maybe 80 concurrent users on a busy morning.
- 8
Sam: Okay. We're going to use your existing REST API as the system of record and build the new portal as a separate front end. We agreed on that with your IT lead last week.
- 9
Priya: Yes, that decision is already approved on our side.
- 10
Sam: One concern from our end: the legacy authentication service is a single point of failure, and the documentation for it is thin. That's the riskiest integration in this project.
- 11
Priya: Understood. The team that built it is gone, so there are unknowns there for us too.
- 12
Sam: I'm assuming the API can return shipment events in under two seconds. If that doesn't hold we'll need a caching layer in front of it.
- 13
Priya: That assumption is probably safe for status lookups, but bulk invoice queries are slow today.
- 14
Sam: We should also build a small usage dashboard for your ops team so they can see which customers actually log in.
- 15
Priya: That would be useful. One hard requirement: the portal must support single sign-on through our existing Okta setup. Security will not approve anything else.
- 16
Sam: Can we get read access to a staging copy of the API before kickoff?
- 17
Priya: I'll check with IT this week, but I expect the answer is yes.
- 18
Sam: What's the timeline pressure on your side?
- 19
Priya: Leadership wants something customer-facing before the October board meeting, so the timeline is tight.
- 20
Sam: Then I'd treat the bulk invoice export work as a phase two item so phase one stays focused on shipment visibility and single sign-on.
- 21
Priya: Agreed, let's keep bulk export out of phase one.
- 22
Sam: I'll send over a scope brief and a draft project plan by Friday, and I'll follow up with your IT lead about staging access.
- 23
Priya: Perfect. Loop in Marcus from our side when you send it.
Stakeholder Follow-up Email
Hi Priya,
Thanks for making the time today — that was exactly the conversation I was hoping to have. Here's where I think we landed, so you can flag anything I got wrong before it hardens into a plan.
Where we landed: The portal's job on day one is to take support out of the loop — customers checking shipment status themselves instead of calling , and pulling their own invoices instead of waiting on finance to email PDFs . Single sign-on through your existing Okta setup is a hard requirement; security won't approve anything else . We're sizing for roughly 1,200 active accounts and about 80 concurrent users on a busy morning .
What we agreed: We'll build the new portal as a separate front end on your existing REST API, which stays the system of record — already approved on your side . Bulk invoice export moves to phase two so phase one stays focused on shipment visibility and single sign-on before the October board meeting .
Two things I want to keep visible rather than bury: the legacy auth service is the riskiest integration in this project — thin documentation, and the team that built it is gone . And my two-second assumption on shipment lookups is probably safe, but bulk invoice queries are slow today, so I'll validate it early .
Next steps: I'll send the scope brief and a draft project plan by Friday, and I'll follow up with your IT lead about read access to a staging copy of the API . You mentioned looping in Marcus — I'll include him when it goes out .
Talk soon,
Sam
[n] markers cite transcript lines; click one to see the source. Copy strips citations for pasting; Download .md keeps them.
This demo is a simplified local version, shown so you can judge the output shape — the product uses LLM-backed extraction on your real transcripts. Run it on a real call.
03 / built to be checked
Built to be checked
Anything that goes to a client under your name has to be right. Every document Scopebase drafts shows its work, so the review pass takes minutes, not the evening.
- Deliverables, not summaries
- A scope brief, a follow-up email, and Linear-ready issue stubs in one pass — documents you can send, not a recap you have to mine.
- Every claim cited
- Each decision, requirement, and risk links to the transcript line that produced it. No orphan claims, no “the AI said so.”
- Edit and send
- Edit anything inline — citations stay intact. Export Markdown for email, Linear, or Notion.
origin
Scopebase comes from House of Giants, a product engineering studio. We built it because we needed it ourselves — our own delivery pipeline runs on it.
04 / pricingmetered by monthly calls
One call write-up pays for the month.
Converting one call into a scope, an email, and tickets takes 60–90 minutes by hand — less than one billable hour covers a month of Starter. Every plan creates the full cited call packet; choose by monthly call volume.
Starter
$49per month
10 monthly calls
refreshes monthly
For solo consultants and fractional CTOs scoping a couple of calls a week.
- All three deliverables, fully cited
- $492/year annual option — save $96
- Upgrade, downgrade, or cancel anytime
Studio
$99per month
30 monthly calls
refreshes monthly
For agencies and dev shops where every discovery call becomes client-ready scope.
- Premium monthly call volume
- $984/year annual option — save $204
- Upgrade, downgrade, or cancel anytime
Paid billing runs through Clerk. Monthly and annual include the same call volume each month; annual is available at checkout.
05 / questions8 on file
The questions operators actually ask
Do I need to install anything?
No. Paste a transcript into the web app — no bot joins your calls, no browser extension, no calendar access.
Which transcripts work?
Any text transcript: Fathom, Zoom, Otter, Fireflies, Granola, or raw notes. Paste-only today; direct imports are on the roadmap.
What about calls that aren't discovery?
Pick the call type when you paste. Working sessions generate a Decision Log instead of a scope brief: what was decided, who committed to what, and what stayed open. You get the follow-up email and issue stubs either way.
How is this different from Fathom or Granola?
Those tools answer “What happened?” — a summary and action items. Scopebase produces the scope brief you'll send the client, the follow-up email that goes with it, and the tickets your team picks up tomorrow. Pair them: your note-taker records the call, Scopebase does the write-up.
What if the AI gets something wrong?
Every claim cites the transcript line that produced it, so checking takes seconds — not a re-listen. Edit anything inline before you send; citations stay intact.
What happens to my transcripts?
They're your workspace content, processed only to draft your documents. We don't use your transcripts to train models.
Does it replace my PM?
No. It replaces the 60–90 minutes after a call where someone turns a transcript into the documents you'd write anyway. Scope judgment, client relationships, and technical decisions stay yours.
Can I cancel anytime?
Yes. Free is 2 calls every month with no card. Upgrade, downgrade, or cancel anytime — no contract.
06 / next step
Stop writing scope at 11pm.
Paste your next discovery transcript. Walk out with the brief, the email, and the issues.