Where the data in your network apps comes from, and what's inside.
Where the data comes from
Five places. The database pulls from each automatically, every day.
π€
Sonar
Our internal network intelligence platform. The main source β people, companies, who knows whom, and email/meeting totals.
π§
Microsoft 365
Every email you send and receive, every calendar meeting you attend. Flows into Sonar first, then into here.
π
Harmonic
Outside data on people (job history, LinkedIn) and companies (funding, headcount). Keeps "current company" current.
π
Granola
For each call in the Network Calls folder, we pull in the meeting title, attendees, and Granola's AI summary.
π€
Claude (AI)
Reads each Granola summary and pulls out the companies and people mentioned in the conversation.
How fresh it is
Sources update all dayβWe refresh every 30 minutesβYour apps see current data
An email you send or meeting you attend should show up in the database within about an hour β we sync every 30 minutes from the upstream systems.
What's inside the database
Six areas of information. Each one is refreshed automatically.
π
Connections
People tied to our team across LinkedIn, Microsoft 365, and Thrive. See the depth breakdown below β the total isn't one number, it's layers of increasingly strong ties.
132,211total
πΌ
Work history
Every job someone's held β current and past. Useful for "who used to work at Gusto?" or "who's been a CFO before?" Covers 38,101 of our 40,978 CRM people (93%).
π
Sonar notes
Notes the network team leaves after meetings or calls. Lucy, Hannah, and Brooke wrote 83% of these.
1,102notes
ποΈ
Network calls
For each call we store the Granola summary, attendees, and an AI summary of the conversation created by Claude β the companies and people that came up, with why. Soon: each Granola folder (Network Calls, Tech Council, Advisorsβ¦) will get its own tailored summary style.
76calls
π§βπ€βπ§
Stripes team
Who's on the team (76 people), and which squad they're in. The Network squad is Lucy, Hannah, and Brooke.
76team members
How deep our network goes
"Our network" isn't one number β it's layers of increasingly strong ties. An app can ask for any layer depending on the question.
1
LinkedIn-connected (direct)
A Stripes team member is directly LinkedIn-connected to this person β we clicked "Connect" at some point. When we've also emailed or met with them through Microsoft 365, those counts get layered on the same row. The strongest single source of "people we actually know."
27,173people
2
Emailed or met with (and in the Sonar CRM)
People we've actually exchanged emails with or met through Microsoft 365 β and who are also tracked in the Sonar network CRM. Updated live. Same number as "all-time" in the recency table below.
23,497people
3
In Thrive (with email)
People formally tracked in the Thrive CRM with an email address β candidates we're evaluating for searches, prospects from referrals, and anyone logged against a company deal. Being in Thrive doesn't mean we've met them; most are prospects someone added to a search.
5,638people
4
Inferred network ties
External people we're tied to structurally, but haven't emailed or met. Two flavors β both inferred from shared company history, not contact:
β’ Career overlap β a Stripes team member and this person worked at the same company at the same time (from LinkedIn work history).
β’ Investor network β a Stripes team member and this person are both tied to the same portfolio company through investing (cap-table membership, board seats, led a round).
Useful for warm-adjacent intros when there's no direct path.
10,417people
5
Historical email-only
The long tail: people a Stripes mailbox has ever exchanged an email with, who aren't in the Sonar CRM. Mostly cold one-off correspondence, conference intros, reply-to-alls. Big volume, low signal per-person β useful for "have we ever talked to this person?" but not actively refreshed today.
101,733people
βͺ
Union β distinct people across all layers
Every unique person across the five layers above, deduped by email. The broadest honest answer to "how many people are in our network?" Any given person usually sits in multiple layers at once.
132,211people
Each number is counted independently; layers overlap significantly (someone we've emailed is usually also LinkedIn-connected), which is why the union (131k) is much smaller than the sum.
How recent is the activity
People we've emailed or met with who are also in the Sonar network CRM, bucketed by recency. Live numbers from Microsoft 365, refreshed every 30 minutes. Numbers match Tier 2 above β same population, just split by how recently we were in touch.
Last 30 days
2,195
Last 90 days
3,692
Last 6 months
6,030
Last 1 year
9,457
Last 2 years
12,910
Last 5 years
18,633
All-time
23,497
Three paths to reach a target company
When we want to reach execs at a target company, the data gives us three different kinds of warm paths. Each comes from a different source and answers a slightly different question.
π―
1. Best Stripes contact
A Stripes team member already directly emails or meets someone at the target. Comes from Microsoft 365. Strongest path when it exists β already a warm relationship.
π°
2. Investors who can intro
An investor we know sits on the target's cap table. We know the investor through a Stripes team member, so the path is: Stripes team member β investor β target exec.
π
3. Former colleagues at the target
Someone in our network used to work at the target, or with someone currently there. Comes from LinkedIn work history. Widest of the three β often the only path when we have no direct contact.
Live example: Investors who can intro us to Base Power
A real slice of what the "investors who can intro" path returns β for Base Power. Each row is an investor on the cap table plus the Stripes team member who knows them best. The same rows that the legacy Sonar widget shows, now available as a single RPC call.
Investors who can connect us to Base Power execs
Ken Fox knows Lee FixelΒ· Addition β led Base Power's Series C
3 meetings133 emails
Bryan Gertzog knows Sebastian DuesterhoeftΒ· Lightspeed Venture Partners β led Base Power's Series C
2 meetings23 emails
John Diorio knows Eric TarczynskiΒ· Contrary β led Base Power's Series C
2 meetings14 emails
Elizabeth Kalaris Scheeren knows Jackson MosesΒ· Terra Industries β independent investor in Base Power's Series C
2 meetings10 emails
Ron Shah knows Lee FixelΒ· Addition
2 meetings39 emails
Matches the legacy Sonar widget row-for-row. Refreshed every 30 minutes from Microsoft 365 and cap-table sources.
What you can ask
Apps built on this database can answer questions like:
Who at Gusto do we know, and who on our team knows them best?
Which people did I meet in the last month that I still need to follow up with?
Show me every network call that mentioned OpenAI.
Who used to work at Stripe and is now in a CTO role?
How many times has Hannah exchanged emails with Jane at Acme?
Which investors on Ramp's cap table could intro us, and who on our team knows them best?
Pull together everything we know about this person before the meeting tomorrow.