Sitelet https://www.civic.com/field-notes
Skip to main content
BRYNby Civic
AboutNewsTrustBrandLegalAlso from CivicCivic Auth opens in a new tabCivic MCP opens in a new tab
Log inGet Started
Field Notes

Field Notes

Writing from the team on building and running agents, and on the work of go-to-market.

Series
Title card: Your Rules, Readable. Subtitle: read it, change it, see what your change did. Below, a panel of three config lines in monospace type, one per part a growth team owns. Profile: icp.industries software, devtools, fintech. Weights: signal.careers_view minus 2. Rules: play pricing then comparison then repeat within 7 days, to Slack, Approve mode. Beside each line, three columns headed READ, CHANGE, and SEE, each marked with a filled dot. Illustrative config, not a benchmark.bryngtmscoring

Your Rules, Readable

Transparent scoring is a claim. Readable rules are a test: can you read each rule in plain words, change it without a ticket, and see what your change did? Here is where each answer lives in Bryn for the profile, the weights, and the Plays, including which profile fields count toward fit today and what happens to existing accounts when you edit.

Civic TeamHero image: a one-person org chart on a dark instrument panel. A top box labeled Head of Growth connects to six role boxes: Demand, Lifecycle, Sales assist, Ops and data, Reporting, and Tools. Every box, including the top one, reads YOU, and every box is lit, with a teal dot marking it as listed. Each role box names its parts in small type, such as site, pixel, and paid under Demand, and pipeline note and the board slide under Reporting. A strip beneath the chart shows the four operations of the series, with 01 Inventory lit and Subtract, Delegate, and Hand off dim. To the upper right, a job posting card lists one responsibility, own growth, above five blank lines marked nobody wrote it down. Title: The Job Nobody Wrote Down. The Growth Team of One, part 1 of 4.Growth Team of One, 1 of 4growth-team-of-onegtm

The Job Nobody Wrote Down

If you are the whole growth team, you do not have a job. You have a department with one name on it, and nobody wrote it down, including you. Before you decide what to stop, automate, or hire for, write the job down by recurrence, not by function, and tag every line. The longest column is the one nobody owns. Part 1 of 4 of The Growth Team of One.

Brad WebbGrowth stack: build vs. buy. Build the brain. Buy the infrastructure. Own the logic about who to target and what happens next. Buy the infrastructure that has to stay reliable and auditable. Stat callout: 32 percent decided against at least one software purchase or feature they could build in-house, McKinsey, State of AI 2026. Left card, build: your intelligence. Your ideal customer profile, your signal weights, your response rules. Your judgment about your own buyers. Keep it where you can read it. Right card, buy: the infrastructure. Visitor identification, connectors, approval step and audit log. Needs uptime and ongoing maintenance. Someone else will eventually need to trust the record. Bottom band, the sixth question: who else will need to trust what this system did?brynbuild-vs-buygrowth-stack

Build the Brain. Buy the Infrastructure.

Coding agents have made more of the growth stack practical to build in-house. McKinsey says 32 percent of organizations have already decided against at least one software purchase or feature because they could build it themselves. The implication is not to build everything. You should own the judgment about who to target, what signals matter, and what should happen next. But you should buy the infrastructure that has to stay reliable, connected, and auditable.

Chris HartTitle card: Onboarding, Run by an Agent. Subtitle: the part worth trusting is what the setup refuses to let an agent do. To the right, a two-position state glyph labelled DRAFT and LIVE, set to DRAFT, captioned live only at the last step. Below, one illustrative onboarding run for yourcompany.com as a six-step rail: workspace done, events current and marked refused, then icp, signals, outputs, and plays, with plays captioned go-live, owner or admin. Under the rail, three console lines: the agent asks to complete the events step; the step is refused because no pixel event has arrived yet, with the fix; then the agent checks events, sees the first pixel event from www.yourcompany.com, and the events step passes. Illustrative.brynonboardingmcp

Onboarding, Run by an Agent

Bryn's setup can now be run by an agent, either Bryn in the app or your own agent over the Bryn MCP server. Clicking through a wizard is the easy part. The part worth trusting is what the setup refuses to let an agent do: go live on a half-staged draft, tick a step that did not happen, fetch a tracking script nobody agreed to, or turn Plays on without an owner or admin. Here is one illustrative run, step by step, and three questions to ask any vendor before an agent sets their tool up for you.

Civic TeamHero image in the Bryn app's light register: warm paper ground with the headline Bryn's First Big Leap on the left, under a mono kicker reading Field Notes, product, and above three short lines: one screen per job, setup as a conversation, a readable record. On the right, a miniature of the rebuilt Bryn console: a sidebar with the Bryn mark and the screen list, a topbar, a Grid and Bands pill, an account card for a fictional account with Fit, Intent and Timing meters, their weights beside them, a written formula line and a composite of 78, and a Plays deck card with its active switch, trigger chips and delivery tiles. Byline: Brad Webb, Chief Growth Officer, Civic.brynproductchangelog

Bryn's First Big Leap

The Bryn app is rebuilt today around one rule: one screen per job. Scores read as meters with their weights, never a bare number. Setup can now be run as a conversation, by Bryn in the app or by your own agent over MCP, into a draft that goes nowhere until go-live. And the record got easier to read.

Brad WebbTitle card: The First Hour, Four Classes. To the right, a clock face marked for one hour, with the minute hand at twelve and sixty minute ticks around the rim. The face is divided into four quadrants labelled with Roman numerals I to IV in monospace: I owned, II rented, III filed, IV everyone rents. Beneath the clock, four short record lines, one per quadrant: I, acct_3c07, trial then second seat in 7 days, ran, posted and task opened. II, acct_7d19, pricing then repeat in 7 days, ran, tagged and pages posted. III, acct_2210, closed-lost then pricing return after 8 months, held, approved and ran. IV, acct_3c07, hiring RevOps role 12 days, modifier, fired nothing alone. Footer: the hour does not disappear, it moves to the record. Illustrative.bryngtmsignals

The First Hour, Four Classes

On Wednesday Chris described the hour with no owner: someone scans the overnight visitor list, checks the CRM, checks the trial, picks three or four, writes the follow-ups, and none of it is written down. Brad closed the field guide yesterday with the fourth class of signal. Hold the hour up against the legend and every check has a plate number. Written as Plays, the hour is four record lines, one per class, and the planning question Chris asked has an answer per class.

Civic TeamHero image: a naturalist's field-guide plate showing four GTM signal classes as specimen etchings. Plate four, ecosystem exhaust, is highlighted with a weather-chart etching: curved isobars across the plate, a barometer dial with its needle leaning toward change, and two small specimen cards under glass, one a job posting and one a funding round. The glass is stamped EVERYONE'S in coral and restamped JOINED in teal, with a teal join line running from the cards to a pricing page node. Title: The Class Everyone Rents. Part 4 of 4 of Signals: A Field Guide.Field Guide, 4 of 4bryngtm

The Class Everyone Rents

Ecosystem exhaust is the fourth class in the legend and the one everyone rents. Hiring, funding, tech installs, leadership changes: the company's weather, sold to you and to everyone who sells against you, and now callable by everyone's agent over MCP. Everyone's weather is nobody's edge. The only thing that makes it yours is the join. Part 4 of 4 of Signals: A Field Guide, and the legend, complete.

Brad Webb2027 planning season. Plan 2027 with the team you already have. Finance wants flat headcount. Sales wants more pipeline. Start by deciding which repetitive work should stop being done by hand. Stat callout: $640K versus $370K net new ARR per GTM FTE, high versus low AI adopters, ICONIQ, 150 plus B2B software GTM leaders. Left card, the 2027 constraint: flat headcount, more pipeline, the classic do more with less, but now with a twist. Center: move the repetitive work, with an arrow pointing right. Right card, the team's new job: 1 define the rules, 2 review the results, 3 improve the workflow. Bottom band: plan the work before the headcount. Move the repetitive, time-sensitive work first. Then decide where another person would add the most value.brynplanningheadcount

Plan 2027 With the Growth Team You Already Have

The good news is that high AI adopters report roughly twice the net new ARR per go-to-market employee as low adopters. But, for 2027 planning the critical question is which repetitive, time-sensitive work your existing team should stop doing by hand.

Chris HartTitle card: Approve Mode Is Not a Speed Bump. Below the title, a single audit-log record line in monospace type: play pricing-return-intro-v2, account acct_2286, signal pricing then return twice in 7 days, score 0.78, boundary pass, action drafted, with a small chip reading held for approval and the field approved_by owner 09:12. To the right, a two-position switch glyph labelled RUN and APPROVE, set to APPROVE. Illustrative.bryngtmplays

Approve Mode Is Not a Speed Bump

Run mode and Approve mode do the same work: watch, score, check the boundary, draft. The only difference is where the Play says the final click lives, in the rule or in a person's morning. Neither is more governed than the other, because the governance is the Play. Here are both modes by behavior, three cases for choosing between them, and two record lines from the same Play, one in each mode.

Civic TeamTitle card: Your vendors ship MCP servers now. Below the title, a shelf holding four small server cards labelled signals, company data, intent, and enrichment, each with a plug glyph and the word MCP; no vendor logos. One line runs from the shelf down into a single console frame labelled Bryn, the record. The console shows one record line in teal monospace: acct_5108, funding then Series B in 18 days, plus pricing then repeat in 7 days, held intro-note-v2 for approval. Under it, Bryn's line: I joined one vendor answer to one signal of yours, matched intro-note-v2, and held it for your approval. Logged. Footer: illustrative, not a benchmark; account anonymized; generic servers, no vendor logos.bryngtmagents

Your Vendors Ship MCP Servers Now

In the last six weeks the signal data layer became callable. PredictLeads exposes every dataset as a tool at mcp.predictleads.com, Coresignal shipped an MCP server with OAuth on August 4 and moved its company dataset to daily delivery on September 10, Demandbase lists native MCP support for enrichment and intent, and Clay puts its data providers and workflows behind an MCP server. Last Monday we wrote that Bryn is the agent your agents call. Now the vendors are too. Good news, and it changes less than it looks like it changes: an agent that can ask for a signal still has to decide, act inside a rule, and write down what it did. Callable is a data property. Run is an operating property.

Civic TeamTitle card: One Account, Four Classes, Six Tools. To the right, a four-by-six grid with signal classes down the side (product, site, CRM, ecosystem) and tools across the top (Website, Product, CRM, Slack, Outbound, Warehouse). Three cells are lit in teal: product behavior under Product, site intent under Website, CRM lifecycle under CRM. A bracket beneath the grid spans all six columns and is labelled one record. Illustrative.bryngtmsignals

One Account, Four Classes, Six Tools

Chris walked one account across six tools on Tuesday: pricing three times in a week, two people active in a trial, a CRM record with a stale owner, and a team expecting Slack, outbound, and the warehouse to do their jobs. Read the same account with Brad's Field Guide legend and it is three signal classes, not six tools. The other three tools are where actions land and records go. The account is three classes wide and six tools long, and it becomes one account in exactly one place: the record.

Civic TeamTitle card: Q4 Is Built in September. Below the title, a thin ruled calendar strip runs from September 23 to December 31: one tick per weekday, 72 in all, with the federal holidays marked in coral, three teal signal marks in the last week of September, and the final fortnight from December 21 hatched as the stretch when nobody signs. Illustrative, not a benchmark.bryngtmsignals

Q4 Is Built in September

Q4 is not a quarter. It is a deadline with a holiday inside it. The last hundred days of the year begin September 23: 72 weekdays, fewer after the holidays and the December fortnight when nobody signs. Count back from the last day anyone signs and you land in September, so the deals Q4 closes are firing signals now. Three Q4 specimens by their parts, the working-days math, and the one decision to make today: a person or a Play.

Civic TeamHero image: a naturalist's field-guide plate showing four GTM signal classes as specimen etchings. Plate three, CRM lifecycle, is highlighted with a filing-drawer etching: a cabinet with one drawer half open, three record cards standing inside, one card lifted out on a teal return arc back to a pricing page node. The drawer front is stamped FILED in coral and restamped OPEN in teal. Title: The Class You Filed. Part 3 of 4 of Signals: A Field Guide.Field Guide, 3 of 4bryngtm

The Class You Filed

CRM lifecycle is the class you own on paper and read least. Closed-lost is filed as an ending when it is a state, stalled evaluations sit in a stage built for counting, and renewal dates never met a behavior. A filed record alone is a filing. A filed record that comes back to your pricing page is a specimen. Part 3 of Signals: A Field Guide.

Brad WebbWhere the agent lives versus where the work happens. Your pipeline crosses six systems. Its controls need to cross them too. Left: an agent inside a platform, one platform with its own rules and log. Right: one approved workflow and one Bryn audit log across the stack, connected to six tools: Website, Product, CRM, Slack, Outbound, Warehouse. One account, six systems, before a seller responds. The question for every agent demo: who can act across the tools, under what rules, and where can you review what the agent decided and did?brynagentsgovernance

Dreamforce and UNBOUND Are Putting Agents Inside Their Platforms. Your Pipeline Runs Across Six Tools.

Dreamforce and UNBOUND are showing what agents can do inside Salesforce and HubSpot. Most growth teams I talk to do not work inside one platform: buyer activity moves across the website, product, CRM, Slack, outbound, and the data warehouse. The practical question is who is allowed to act across those systems, and where you can review the decisions and actions across the workflow.

Chris HartTitle card: The agent your agents call. Below the title, two console frames side by side. The left frame is labelled your agent (Claude, Cursor, in-house) and shows three plain-language questions: which fit accounts hit pricing this week, why did acct_2210 score 86, what ran on it and who approved. The right frame is labelled Bryn, MCP, and shows the incoming call get_account_brief for acct_2210 and Bryn's reply lines: scored 86 against threshold 70, Fit 88, Intent 91, Timing 76, last run demo.follow-up on September 9 at 14:02 UTC, approved by op_3, record run_7d2e19, Logged. A single call line between the frames carries get_account_brief. Under both frames, one reply line in teal monospace: acct_2210, 86, pricing then comparison then repeat in 7 days. Illustrative, not a benchmark.bryngtmagents

The Agent Your Agents Call

Dreamforce and UNBOUND open this week on the same idea: the agent lives inside the platform. Bryn is already the agent your agents call. Over MCP, a Claude, Cursor, or in-house agent can ask who is on your site from a fit account, why an account scored what it did, what ran on it and who approved, and hand Bryn a Play to install, with every answer drawn from the same record the UI shows. That is live today. Next, Bryn is learning to read a company the way a buyer does. Yours first.

Civic TeamTitle card: Show Me a Blocked Action. Below the title, a single audit-log record line in monospace type: play pricing-return-nudge-v1, account acct_4471, score 0.74, boundary fail, blocked: suppression_list highlighted in coral, action none. Illustrative.bryngtmgovernance

Show Me a Blocked Action

Chris gave buyers a vendor test on Wednesday: show me an account the agent was not allowed to touch, the rule that blocked it, and the record. Here is that test run against Bryn, in public: one Play, one signal that clears the score, one account on a suppression list, one blocked action, one record line another person can reconstruct from. Then the honest edge of what Bryn cannot show.

Civic TeamHero image: a naturalist's field-guide plate showing four GTM signal classes as specimen etchings. Plate two, anonymous site intent, is highlighted with a visit-path etching of three page nodes and a return arc in teal, stamped with a coral lease tag. Title: The Class You Rent. Part 2 of 4 of Signals: A Field Guide.Field Guide, 2 of 4bryngtm

The Class You Rent

Anonymous site intent is the class almost everyone rents, including us. Renting is not the sin. Renting badly is: paying for a match rate you never route, treating a resolved visit as a lead, and forgetting that a rented instrument reprices, degrades, and gets rented to your competitor. Part 2 of Signals: A Field Guide.

Brad WebbMore than 25 percent of Workday's new annual contract value came from AI. Its CEO said customers trust the agents because of the controls around them. Three controls: clear boundaries, what can the agent do. Recorded approval, who authorized it. Audit trail, what happened afterward.brynagentsgovernance

AI Drove More Than a Quarter of Workday's New Sales. Its CEO Credited the Guardrails.

Workday says the controls around its agents are part of why customers trust them. Gartner predicts governance failures will force many companies to pull autonomous agents back from production. The common issue: companies need to know what an agent can do, who authorized it, and what gets recorded afterward.

Chris HartTitle card: Tools Converge. Action Doesn't. Five shelf outlines labelled enrichment, sequencing, intent, visitor ID, and CRM slide toward one another on a shelf, while one slot at the end stays empty, outlined in coral and labelled action.bryngtmsignals

Tools Converge. Action Doesn't.

Clay made its Sequencer generally available on September 1: sourcing, enrichment, and sending in one tool. It lands weeks after 6sense put intent inside agents and as visitor-ID vendors sell engagement. Every category is absorbing its neighbors, and every launch stops at the same line: someone still has to decide the action, approve it, and own the record.

Civic TeamHero image: a punch-clock time card rendered as an instrument panel. Five rows, Monday to Friday, each an hour bar from 8am to 6pm split between work only you can do, in violet, and chores, in coral. A coral rubber stamp across the card reads follow-up, maybe, forgot, three days. Title: Follow-up has no owner.bryngtmfollow-up

Follow-Up Has No Owner

Labor Day is a decent excuse for a time-spent audit. The growth owner's real labor is not strategy, it is unowned follow-up: checking, cross-referencing, guessing, writing, forgetting. The pattern is follow-up, maybe, forgot (3d). Give it an owner: a person or a Play. Save Your Labor (Day) makes the first month of Bryn free through Sep 17.

Brad WebbTitle card: Ask About Tuesday. A weekly calendar strip with Tuesday highlighted.bryngtmagents

Ask About Tuesday

Chris's three agent-buying questions (setup, ongoing work, exit) collapse into one you can actually evaluate in a sales call: describe a normal Tuesday, three months after launch. Here is the script, and what good and bad answers sound like.

Civic TeamHero image: a naturalist's field-guide plate showing four GTM signal classes as specimen etchings. Plate one, product signals, is highlighted with calibration markings. Title: The Class You Own.Field Guide, 1 of 4bryngtm

The Class You Own

There are four classes of GTM signal and only one of them is yours. Product signals are the highest-fidelity class, the only class nobody can rent out from under you, and the class almost nobody runs. Part 1 of Signals: A Field Guide.

Brad WebbHero image: Big companies scaled agents this year. Everyone else stood still.bryngtmagents

Big Companies Scaled Agents This Year. Everyone Else Stood Still.

McKinsey's 2026 State of AI: 40 percent of billion-dollar companies are scaling agents, up from 27. Smaller companies sat flat at 22. The gap is implementation capacity, not software cost, and there are three questions to ask before you buy.

Chris HartTitle card: Intelligence Became an Ingredient. A row of labeled jars feeding one highlighted execution vessel.bryngtmsignals

Intelligence Became an Ingredient

6sense now pipes its buying-stage predictions and intent data into Claude, ChatGPT, Writer, and Agentforce via MCP. When the strongest see-layer vendor makes its intelligence callable from anyone's agent, intelligence has stopped being the product. What stays scarce is governed execution.

Civic TeamTitle card: Thirteen Agents, One Owner. Thirteen small agent marks, one highlighted owner mark.bryngtmagents

Thirteen Agents, One Owner

Salesforce's Agentic Enterprise Index puts the average agent fleet at 13, up from 5 a year and a half earlier. Building an agent got 53 percent faster. Governing one did not. Three verbs tell you if your fleet is run or just running.

Civic TeamHybrid Isn't a Compromise. Pipeline per seat per month: human-only about 187 thousand dollars, AI-only about 94 thousand dollars, hybrid about 278 thousand dollars. Source: Bridge Group and Apollo ZoomInfo data as aggregated by Digital Applied. Team Civic, Civic Field Notes.bryngtmoutbound

Hybrid Isn't a Compromise

Hybrid pods out-produce both pure configurations: roughly $278K pipeline per seat per month versus $187K human-only and $94K AI-only. The reason is not that AI is half-good. It is that the division of labor matters: the human decides the pattern, the machine executes inside it.

Civic TeamThe Meter Is a Permission Structure. Whatever the contract says, your team reads the meter as policy. Team Civic, Civic Field Notes.bryngtmpricing

The Meter Is a Permission Structure

Whatever the contract says, your team reads the meter as policy. An opaque unit teaches hesitation, an invisible count teaches asking permission, an unknown cap behavior teaches running the smaller job. The fix is a legible meter and a three-line internal memo.

Civic TeamOne moment on the measurement bench. The moment card, pricing then comparison then repeat within seven days for account 4471, sits on a bench with probes clipped to its six stages: signal fired 11 times, identity confidence 0.86, score 0.78 against a 0.70 threshold, Play pricing-follow-up-v2, bounds of two named channels and a cap of 40 runs, record one line per run. Below, the week's four survival numbers: 11 fired, 7 cleared, 6 landed in-window, 3 moved. All values illustrative, not a benchmark.Personalization, 3 of 3bryngtm

The Moment, Measured

Open rates lie to you about all six stages at once. Part 3 of 3 on personalization closes the series with the measurement: four numbers that survive contact (moments seen, moments cleared, touches landed in-window, accounts that moved), why time-to-touch is the master metric, and how the record turns measurement from a project into a query.

Brad WebbAI Pricing Broke This Year. Check what your vendors started counting. About four in five vendors recently introducing AI pricing chose capacity over consumption, with three checks: read the unit, audit the count, know the cap. Source: Bain and Company, August 10, 2026.bryngtmpricing

AI Pricing Broke This Year. Check What Your Vendors Started Counting.

About four in five vendors recently introducing AI pricing chose to meter pricing on capacity over consumption. Buyers need to know what gets counted, who carries the quality risk, and whether they can audit the bill.

Chris HartBounded Autonomy Is the Spec. Five checks: named channels, a cap, a suspend that suspends, per-run records, explainable scores. Team Civic, Civic Field Notes.bryngtmgovernance

Bounded Autonomy Is the Spec

The evaluation question for agentic GTM tools flipped this year: not what can the agent do, but what can it not do, and can you prove it. Five questions make the spec: named channels, a cap, a suspend that suspends, per-run records, explainable scores. Run them before the demo.

Civic TeamHero: Three Surfaces, One Architecture. Three stacked planes, the tool's UI, the agent inside the tool, and the agent outside over MCP, threaded by a single glowing spine labeled the record. Caption: two unrelated teams arrived at the same shape, that is architecture, not positioning.brynarchitectureagents

Three Surfaces, One Architecture

On a recent feedback call, a founder who has spent about twenty years building marketing software described our architecture back to us, unprompted, for his own product: software now lives on three surfaces at once, the tool's UI, the agent inside the tool, and the agent outside it reaching in over MCP, and the record is how you stay in control once you let an agent act. Two unrelated teams arriving at the same shape independently is not positioning. It is architecture.

Civic TeamHero: Deliverability Is a Commons. Five sender agents graze one shared band of inbox trust. Four draw on it lightly; one coral over-sender draws three lines into the same pasture, and the band reddens beneath it. A fence-line chip marks enforcement at 0.3 percent spam rate. Caption: every over-sender degrades the channel for everyone, including themselves.bryngtmdeliverability

Deliverability Is a Commons

Email deliverability is a commons: every over-sender degrades the channel for everyone, including themselves. Google and Yahoo enforce a 0.3% spam-rate ceiling, most fully autonomous AI SDR pilots get pulled inside 90 days, and warmup tricks do not fix a grazing problem. A usage cap protects the budget. An audience cap protects the channel.

Civic TeamHero: Anatomy of a Moment. One moment, the pattern pricing then comparison then repeat inside seven days, drawn as an exploded instrument diagram. On the left, the assembled moment card with its three timestamped visits. Dashed explode lines lead to six part panels: 01 signal, a pulse trace with a seven day window; 02 identity, account 4471 resolved at confidence 0.86, resolved not certain; 03 score, fit 0.74, intent 0.86, timing 0.70, weighted 0.78 against a 0.70 threshold, cleared; 04 Play, the approved card pricing-follow-up-v2 with trigger, qualify, action, destination; 05 bounds, channel email only, run 14 of 200 this month, suspend instant and global; 06 record, the first person audit line, I ran pricing-follow-up-v2 against acct_4471, what was seen, why it qualified, what happened. Every number illustrative, not a benchmark.Personalization, 2 of 3bryngtm

Anatomy of a Moment

Between a signal firing and a relevant touch landing there are six stages: signal, identity, score, Play, bounds, record. Each one is a place teams silently fail, and doing all six by hand, in minutes, is a staffing problem disguised as a discipline problem. Part 2 of 3 on personalization: the mechanics.

Brad WebbRamp Spent Four Years Rebuilding Outbound. You Shouldn't Have To. 30 percent of pipeline, shut down on purpose. Sending became a commodity. The advantage moved from a volume edge to a selection edge between 2021 and 2025.bryngtmagents

Ramp Spent Four Years Rebuilding Outbound. You Shouldn't Have To.

Ramp shut down the AI SDR program behind 30 percent of its pipeline, then spent four years rebuilding outbound around evidence, bounded action, and a record. The architecture they landed on is right. The timeline is the part almost nobody can copy.

Chris HartHero: The Reply Rate Isn't the Metric. Two dashboard cards from the same quarter. The left card shows raw reply rate falling from 4.7 percent to 2.9 percent, down 38 percent, and says outbound collapsed. The right card shows absolute replies rising from about 54 to about 215 a month on 6.4 times the volume and says outbound never worked better. Caption: both dashboards are lying. Measure the moment you acted on, not the mail you sent.bryngtmoutbound

The Reply Rate Isn't the Metric

AI-assisted volume pushed sends per rep from about 1,150 to about 7,400 a month while raw replies fell from 4.7% to 2.9%. One dashboard says outbound collapsed. The other says it never worked better. Both are lying. The metrics that survive volume inflation: positive reply rate, conversion to booked, channel health.

Civic TeamHero: The Four Handoffs. A rail of five funnel stages, lead, prospect, trial, customer, revenue, connected by four glowing gaps labeled where deals die. Caption: your funnel report shows the stages, deals die in the gaps.bryngtmconversion

The Four Handoffs

Conversion is four handoffs: lead to prospect, prospect to trial, trial to customer, customer to revenue. Your CRM reports the stages. Deals die in the gaps, and nobody owns the gaps. That is Bryn's place in conversion.

Civic TeamHero: The FirstName Fallacy. A split email composer. Left pane, a greyed template titled TOKENS with merge fields FirstName, Company, and Title, stamped DELETED UNREAD. Right pane, a lit draft titled THE MOMENT referencing a pricing page visit and an invited teammate, stamped REPLIED IN 4 MINUTES. Caption: one knew my name, one knew what I did.Personalization, 1 of 3bryngtm

The {FirstName} Fallacy

Token personalization is free now, which means it is worth nothing. Real personalization is responding to what a lead, a prospect, or a trial user just did, and the moment expires in minutes. Part 1 of 3 on personalization: the why.

Brad WebbHero: The EU AI Trust Deadline Moved. The Buyer Didn't. Two panels compare the regulator's clock (EU AI Act high-risk dates deferred to December 2027 and August 2028) with the buyer's clock (evidence, record, and exit due on this quarter's deals).bryngtmagents

The EU AI Trust Deadline Moved. The Buyer Didn't.

Europe pushed its central high-risk AI requirements to 2027 and 2028 while most of its transparency rules took effect on schedule. If your trust roadmap was timed to a regulator's calendar, you now know it was the wrong clock.

Chris HartHow Bryn Answers the Three Questions. Three panels labelled deliver, record, and exit sit over an audit-log strip. Bryn answers the buyer's three questions at the mechanism level.bryngtmattribution

How Bryn Answers the Three Questions

Chris named the three questions serious buyers now ask: what did it deliver, can you show me the record, how do I leave. This is the mechanism answer for Bryn. Every action traces to an account, a signal, a score, and the Play that governed it. The audit log is the operating model, not an accessory. Billing is monthly and the record is yours to export, including on the way out.

Civic TeamThe MQL is dead. The signal isn't. A crossed-out MQL record card beside a live-signal record. The MQL measured what a buyer declared. The signal measures what they did.gtmsignalssignal-based-selling

The MQL Is Dead. The Signal Isn't.

The MQL measured declared interest: a form fill, a lagging score, a routing schedule. That worked when buyers raised their hand early. They don't anymore, so the form fires late or never. The intent the MQL was trying to approximate is more alive than ever. It just moved into live signal, and the replacement is signal-based routing: name the pattern by its parts, score it transparently, run the Play while the signal is still warm.

Civic TeamProof of Claims. A status console reprices the burden of proof: Proof of Concept is struck out and greyed, Proof of Claims is lit. Under it stand three proof chips: attributable pipeline in the buyer's units, a checkable record tracing signal to score to Play to outcome, and a clean exit that exports the log.bryngtmagents

Proof of Claims

For a decade, PoC meant Proof of Concept. The mature market repriced it. The proof that closes now is a Proof of Claims: a claim in the buyer's units, a record they can check, and a clean exit. Why that is the burden GTM leaders should want to carry.

Brad WebbCivic Field Notes. The AI buyer grew up. Good. Delivered results, a verifiable record, and a clean exit. Three question chips and three 2026 survey stats: 21.7% direct P&L impact (Futurum), 75% AI strategy 'more for show' (Writer), 40%+ agentic projects headed for cancellation by 2027 (Gartner).bryngtmagents

The AI Buyer Grew Up. Good.

The 2026 buyer wants to know what your agent delivered, how you can prove it, and what happens if it fails. With only 29% of companies reporting significant ROI from generative AI, that scrutiny is healthy.

Chris HartBryn in Your Stack. Inputs on the left (product events, site, CRM, warehouse, intent) flow into Bryn in the center, which watches, scores, runs, and logs, then writes to outputs on the right (CRM, Slack, sequencer, tasks). Bryn sits as the governed execution layer across a stack you already run.bryngtmsignals

Bryn in Your Stack

Bryn is not a rip and replace and not another dashboard. It sits across the stack you already run, reads the signals your tools produce, and runs the Play you approved through the tools you already use.

Civic TeamWhere the other 83 percent goes. A single bar of the buying journey: a thin lit band, about 17 percent, marked sales-touch, and a large dark band, 83 percent, marked anonymous first-party signal. The visible part of the journey is the smallest part of it.gtmsignalssignal-based-selling

Where the Other 83% Goes

B2B buyers spend only about 17% of the buying journey with sales reps. The deal is mostly decided in the 83% you cannot see: anonymous product behavior, trial depth, pricing revisits, comparisons. That 83% is not empty. It is the richest first-party signal you own, going unrun.

Civic TeamWhen the buyer is an agent. A run log whose actor is agent reads docs, calls the API, and runs the CLI, then resolves to an account, glowing on a dark instrument ground, with a faint human cursor ghost fading behind the live agent trace.bryngtmsignals

When the Buyer Is an Agent

Your buyers are starting to send agents to evaluate you: reading your docs, calling your API, running your CLI before a human ever fills a form. That is a new class of first-party signal, and the teams that capture it and run it under governance will win the next few years.

Brad WebbCivic Field Notes. You already paid for this pipeline, why aren't you converting? A mid-market funnel narrows from 2,700 visitors to about 38 leads to 1 customer, next to three published benchmarks: mid-market stage benchmarks compound to one customer per 2,700 visitors; the top 10 percent convert visitors to leads at 8 to 15 percent against an average near 1.5 percent; acting on a signal within five minutes yields an 8x higher contact rate than waiting 24 hours. The other 98 percent got counted, aggregated, and forgotten.bryngtmsignals

You Already Paid for This Pipeline, Why Aren't You Converting?

Your site converts at ~2% of visitors. The other 98% aren't gone. They're sitting in tools you already pay for, waiting for someone to notice them.

Chris HartOne operator, a portfolio of Plays. A single operator on the left connects to a grid of six Plays running in parallel, each with its own status: pricing to repeat (running), de-anon visitor to Slack (watching), trial stall nudge (scored), usage to expansion (running), sign-up abandon (watching), G2 compare (logged). Tagline: you don't scale headcount, you scale Plays.bryngtmplays

One Operator, a Portfolio of Plays

The lean teams everyone is citing did not win by automating outreach. They win because one operator now covers what used to take a team, by running a portfolio of Plays. The unit of GTM leverage is shifting from headcount to Plays.

Brad WebbWhat the audit log saw. One week read back through the running work record: signals Bryn watched, scored, ran, or held for a human, each one logged.brynaudit-logplays

What the Audit Log Saw

One week at Civic, read back through the audit log. Not metrics theater, a few real patterns Bryn watched, scored, ran, or held for a human. The point lands quietly: the audit log is the running work record, not a compliance artifact.

Civic TeamAnatomy of a Play. A Play is a named pattern with four parts: a trigger, bounds, an action, and a log. Shown as the loop watch, score, run, learn, threaded through one Play: pricing to comparison to repeat within seven days.brynplaysgtm

Anatomy of a Play

People ask what a Play actually is. So here is one, taken apart. A Play is a named pattern with a trigger, a set of bounds, an action, and a log. An operator approves it once; Bryn runs each instance and logs every one.

Civic TeamThe signal you can't buy. Left, OWNED: your first-party product behavior flows as a single bright stream into your own lane, exclusive and fresh. Right, RENTED: a shared third-party intent pool feeds many teams that all converge on the same accounts. Tagline: the edge isn't the signal you buy, it's the one you own.bryngtmsignals

The Signal You Can't Buy

For a decade the play was to buy more signal. But a signal your competitor can also buy was never an edge. The case for owning the one signal you can't buy: how people use your own product.

Brad WebbCivic Field Notes. Seeing is consolidating, acting is still open. Four go-to-market signal acquisitions in eleven months on a timeline: Cursor acqui-hires Koala (Sept 2025), Apollo acquires Pocus (Mar 2026), HubSpot to acquire Warmly (Jun 30, 2026), Zoom to acquire Common Room (Jul 2, 2026), with Acting marked as the high ground still open.gtmsignalsconsolidation

Seeing Is Consolidating. Acting Is Still Open.

Four independent go-to-market signal companies disappeared into four larger platforms in eleven months. That is not noise. It is the market telling us where the value is moving: seeing is becoming infrastructure, and acting is the layer still worth building on.

Chris HartThe land motion flipped. In 2021 about a third of B2B SaaS ran a product-led motion; by 2026 it was the majority, roughly 60%. The dominant land motion crossed over, and the richest buying signal moved inside the product.gtmplgsignals

The Land Motion Flipped

For two decades most new B2B SaaS launched sales-led. In 2026 that flipped: more new companies now launch product-led. When the product is how you land, the product's behavior is your richest buying signal. The motion changed, and the signal layer changed with it.

Civic TeamThe axis collapsed. Buy and build converge into Can, which is now cheap. The new decision point is a Should gate: some jobs pass to run, others are not yet.gtmagentsstrategy

Buy vs Build Has Become Can vs Should

For twenty years the go-to-market tooling decision was buy versus build. Agents collapse that axis. Capability stops being the constraint, and the real question becomes should this run at all, and should it run on its own.

Brad WebbThe Edge Is the Minute After. Field Notes, Civic.gtmsignalssignal-based-selling

The Edge Is the Minute After

Detection is commoditized. Everyone can see intent now. The teams that win this year are the ones that can act in the minute right after a signal fires, and stand behind what they did. The constraint was never vision. It was hands.

Civic TeamCustomer Zero. A dark console panel labeled Bryn run #0001 glows over an instrument ground, logging five timestamped steps: watched anweledig pricing_view x3 at 10:02:11, scored icp.growth cleared 4 of 5 axes, approved by brad in 4 minutes, ran send.intro to owner, and logged run #0001 at 10:06:46, captioned the moment the thesis became a log line. Below, a long coral bar labeled days collapses through an arrow into a short teal bar labeled minutes.bryngtmsignals

We Are Customer Zero

Before Bryn ran for anyone else, it ran for us, and it still does. Why we stopped buying and started building, in Bryn's own voice.

Brad WebbTransport. Attention. Reactive. Field Notes, Civic.aiagentsgtm

Transport, Attention, Reactive: What We Think AI Is Actually For

The dominant story of AI is the answer machine: a box you query, a generator you prompt. We think the higher-leverage use is different, an agent that pays attention for you and carries the moment into action. That is what Bryn is.

Civic TeamWhat Bryn Actually Does. Field Notes, Civic.bryngtmsignals

What Bryn Actually Does

You heard the name this week. Here is the honest, mechanical version: what Bryn watches, how it scores, how it runs the Play you approved, and what it does and does not do at launch. No demo magic, just the loop.

Civic TeamBryn, by Civic. The Signal-Based GTM agent for Growth teams.bryngtmsignals

Announcing Bryn, by Civic

Today we're launching Bryn, by Civic: the Signal-Based GTM agent for Growth teams. Generally available, self-serve, live. Here's what it is, why we built it, and the eyes-versus-hands argument it settles.

Brad WebbMany faint lines converging through a single aperture into one focused purple beamstrategygtmvertical-ai

Why We Narrowed

We focused our platform that could do anything to a product that does one thing on its own. Here's why.

Chris HartA dark console log. A buying signal fires at 10:02, tagged live. It sits unworked through an 11:41 alert and a Wednesday standup. The first human reply finally leaves on Thursday, tagged cold. A caption reads: we counted the signals we caught, we never counted the 3 days in between.gtmsignalssignal-based-selling

We Tried to Buy Our Way Out of the Signal Gap

Every quarter we bought another tool that could see intent. None of them could act on it. What I learned trying to close the gap between a signal and a move, and why we stopped buying and started building.

Brad WebbA dark chart titled "The model was never the problem." A value-delivered curve peaks at a point labeled "the demo / budget approved" and falls to a low point labeled "week 6 / channel goes dark." A note reads: the gap was never intelligence, it was a job, a loop, and permission to act.gtmagentsproduction

The Model Was Never the Problem

Most agent pilots stall for reasons that have nothing to do with intelligence. A go-to-market view on why, and what actually gets an agent into production.

Chris HartA dark console listing five numbered habits for using AI: one job per session, refine don't restart, argue the other side first, propose don't solve, give it tools. A purple cursor blinks after the fifth line.ai-literacygtmprompting

Five AI Habits for GTM Operators

Most GTM teams got handed AI with no manual. Here are the five habits that separate the operators who get real work out of it from everyone still talking to a chat window.

Brad WebbA dark graph titled "Everyone has signals. Almost no one runs them." An intent-value curve starts at a high-intent signal marked 10:02 and decays toward a baseline labeled hours later. A bracket over the first stretch of the curve is labeled "the gap," a point partway down is labeled "action, finally," and a note reads "the edge moved from detection to the minute after it."gtmsignalssignal-based-selling

Everyone Has Signals. Almost No One Runs Them.

Signal-based selling became the consensus go-to-market strategy of 2026. The bottleneck is no longer detecting intent. It is what happens in the next minute.

Chris HartA dark illustration of a code diff panel showing a changed lead-scoring rule, one line removed and one line added, with the author field reading "unknown." A purple line leaves the panel and splits: one path loops back as a rollback, the other frays and fades out.gtmgtm-engineeringrevops

Version-Controlling Your Go-To-Market

Your ICP definition, scoring, routing, and sequences are source code with no version control. Four habits fix that, and none of them require a new tool.

Civic TeamA schematic styled as a TypeScript file named goto_market.ts. A buyer-state lifecycle runs from unknown to retained, a code panel defines a goToMarket function over an event stream, and a pipeline strip maps each stage to a function call ending in retained.netOf(churn).gtmplggtm-engineering

A Field Guide to Go-To-Market, for People Who Read Code

How to read the commercial system with engineering eyes.

Brad WebbWe Built an AI Booking Bot. Most of the Code Is Guardrails.aiguardrailsagents

We Built an AI Booking Bot. Most of the Code Is Guardrails.

The AI part of our booking bot took a day. The guardrails took a week. Here's why that ratio is exactly right.

Titus CapilneanBest Agentic Frameworks in 2026frameworksaicomparison

Best Agentic Frameworks in 2026

The agentic framework landscape has matured fast. Here's what each one is actually good at, and how to pick the right one for your build.

Civic TeamFrom Claws to Fingers: Evolving MCP for the Agent Revolutionaimcpsecurity

From Claws to Fingers: Evolving MCP for the Agent Revolution

As agents move beyond chat into real systems, the challenge is no longer model intelligence but the tooling layer beneath them. Here's how to make it production-ready.

Daniel KelleherAn OpenClaw agent and its creator share their experienceopenclawagentsbuilding

An OpenClaw agent and its creator share their experience

OpenClaw experiment: how Civic’s Brad Webb built Snowcrab, an agent with identity, live hockey monitoring, Civic Nexus integration, and real-world lessons.

Brad WebbAgentOps: The safety net for autonomous AIagentopssecurityai

AgentOps: The safety net for autonomous AI

As AI agents move from prediction to action, AgentOps emerges to manage risk, enforce governance, and ensure safe, accountable automation.

Civic TeamMCP for all: solving the auth problemmcpauthsecurity

MCP for all: solving the auth problem

Explore the challenges of MCP server authentication and how @civic/auth-mcp simplifies OAuth 2.0 for secure, multi-user AI agent access.

Daniel Kelleher