Most operators looking at AI see the headline pricing and budget for that. What they miss is the way AI tools actually charge, which is by token consumption. Long chatty inputs and detailed outputs both consume tokens, and the bill arrives at the end of the month based on what you used.

The fix is not about using less AI. The fix is matching the model you are using to the task you are trying to accomplish.

What you'll hear:

- Why AI bills surprise operators and what's actually being measured (tokens, not seats)

- The three model tiers - premium, workhorse, budget - and how they price differently

- A real comparison where the same task cost $1,000 on premium and $42 on budget

- Three questions for matching the model to the task (complexity, volume, judgment)

- How caching and batching compound the savings

- A real example of how SOS Signals uses all three tiers in a coordinated workflow

The framework is a compressed version of Decision Four in The High-Stakes AI Playbook. The book covers the full build versus buy decision tree and the architecture choices that determine whether AI is a cost line or a profit lever.

The manuscript is open for early access reads. Reply to the newsletter or DM me on LinkedIn for a copy.

Find more at outboundsos.com

Subscribe to the #NoVendors newsletter at jeffswan18.substack.com

Big Tech is selling owners a story about AI that says cut staff, deploy AI, and watch the stock surge. That playbook works for Big Tech. It does not work for businesses where people are on the job site, in the field, in the clinic, or in the truck.

This episode walks through three real AI deployments from high-stakes businesses and ends with the three-question test I run before any deployment touches a client company.

What you'll hear:

- Why the Big Tech AI playbook fails in businesses with hands-on operations

- How a live flood intake agent amplifies field assessments without replacing the human walking the property

- The mining ops manager's four-question test for any new technology (easier, faster, safer, better)

- Why AI-generated cold emails make every seller sound the same, and how to use AI for sales without losing your voice

- The three-question test I run before any AI deployment at a client company

- Why the People Protection Principle is the rule that holds all of this together

The three-question test:

1. Can it deliver without staff reduction in 12 months?

2. Are the people affected the ones carrying institutional knowledge?

3. Have you measured team experience as carefully as financials?

Three yes answers means the deployment respects the People Protection Principle. Any no is a signal to slow down.

The People Protection Principle is one of seven critical decisions in The High-Stakes AI Playbook, launching Q3 2026. The manuscript is open for early access reads. Reply to the newsletter or DM me on LinkedIn for a copy.

Find more at outboundsos.com

Subscribe to the #NoVendors newsletter at jeffswan18.substack.com

This episode opens with three patterns from a year of building AI agents alongside hands-on owners and CEOs of high-stakes companies.

- AI moves fast (and breaks things every model update).

- Specificity hurts AI performance (which is why knowing when to deploy generative versus deterministic software matters).

- Hallucinations make blind trust dangerous (which is why every Signal we build runs through human verification).

Then the announcement. I have written The High-Stakes AI Playbook, covering seven critical decisions hands-on operators need to make before deploying AI. Targeting Q3 2026 launch.

The book is for owners and CEOs of high-stakes businesses with real choice about how AI fits into their operations. It is not for companies in declining industries making survival decisions, or public-company CEOs operating under short-term fiduciary mandates.

The manuscript is going out to a thousand operators, analysts, and commentators. Honest reviews from real operators may end up in the foreword.

What you will take away:

- Three patterns that surface in every AI build for high-stakes operators

- A framework for when to deploy generative AI versus deterministic software

- The case for keeping humans accountable in the loop

- How to get a copy of the manuscript

Your move: If you are running a high-stakes company and want a copy of the manuscript, watch your inbox, reach out by DM, or email. Honest feedback is what I am asking for.

Find more at outboundsos.com

Newsletter at jeffswan18.substack.com

I break down the three categories of government activity creating buying triggers in 2026: tax incentives, direct funding programs, and policy mandates. Real examples include IRS Section 179 (where one auto dealer used the 100% same-year write-off to 5X commercial van sales), Canadian healthcare innovation funding tied to the top five priorities across health authorities, and the broader pattern of policy mandates from EV rebates to sunsetting unsafe technology.

The episode closes with an 8-step play for finding the programs relevant to your industry, building a targeted campaign around one offer, and executing with the authority of someone who showed up with the answer already mapped.

What you'll take away:

- The three categories of government-led buying triggers worth tracking

- A real case study (Section 179, luxury auto dealer, 5X sales lift)

- An 8-step play you can run this week to find programs relevant to your segment

- Why sellers who align with policy will outperform sellers blaming the economy in 2026

One reminder: I am not a tax accountant. All tax examples are illustrative. Consult your professional.

Resources mentioned:

- IRS Section 179

- National Research Council Canada IRAP

- Federal and provincial healthcare priorities (Canada)

Your challenge:

Pick one program in your industry, find it before the end of the week, and reply with what you found. I read every response.

Find more at outboundsos.com

Newsletter at jeffswan18.substack.com

In this episode, I walk through three places I've deployed AI in real businesses. SOS Signals for B2G research, a voice onboarding agent for a colleague's fast-growing services company, and a healthcare innovation that's in early pilot. Three different problems, three different agents, one principle underneath all of them.

In this episode

The Sam Altman moment that sparked this conversation, and why AI founder commentary keeps making working people uneasy

How SOS Signals went from a book research project on B2G markets to an agentic product that saves senior sellers hundreds of hours per proposal

The six months of structural framework-building required to get AI to deliver consistent results across different industries, clients, and regions

Why a voice onboarding agent loaded with founder knowledge solves attrition problems that growing companies can't fix by hiring more trainers

How the same systems thinking applied to healthcare procurement is helping close the gap between government-funded patient care initiatives and the vendors who can deliver on them

Why AI lies even on simple factual questions, and what that means for any operator trying to use these tools without verification

The principle that connects all three case studies and explains why some AI deployments succeed while others produce noise

Key takeaway

AI works as a complement and amplifier to great people. The operators who learn to direct these tools come out of this transition stronger than they went in.

Be the Human Boss of the technology.

Mentioned in this episode

SOS Signals, the agentic research product for sales teams selling into government and government-adjacent markets

The #NoVendors book project, original research that became the foundation for SOS Signals

The voice onboarding agent built for a colleague's growing services company to solve attrition during rapid growth

Healthcare procurement pilot, early-stage work to close the gap between funded patient flow initiatives and qualified vendors

Watch the video version

YouTube: https://youtu.be/b_wsKQ6im-E

Subscribe to The Signal

The Signal is a weekly podcast on what it takes to win in high-stakes government-shaped markets, where AI deployment, operator judgment, and human amplification meet.

Subscribe on Substack to get every episode delivered to your inbox along with weekly observations on building AI for real businesses, or follow/subscribe wherever you get your podcasts.

I told him to wait.

What happened next is the story behind this episode. Four months of working the formula. Six meetings per rep per month, climbing to fourteen. A plateau that taught me something I've carried into every founder conversation since: optimization works fast, and even working systems have limits, and headcount won't override either of those.

This episode is about what a sales formula actually contains, why founder motion is real but unscalable, and the three questions every founder in a long-cycle B2G market should ask before opening their next sales req.

If you've ever sat across from a CEO who wanted to hire faster than the formula could support, this episode is for you. If you are that CEO right now, this is the one to listen to twice.

—

In this episode:

(00:00) Why hiring at the wrong time is the most expensive decision

(00:30) The Mediterranean unicorn story

(02:20) The four-month optimization that changed everything

(03:03) Two lessons from scaling to twenty-five reps in six continents

(03:34) Founder motion vs. documented formula

(04:55) When the founder is still the closer

(06:14) What a sales formula actually contains

(07:43) Why this matters more in long-cycle B2G

(09:15) The three questions to ask before your next hire

(11:13) Why patience is the real work

(12:39) Headcount comes after the formula, not before

—

Read the companion article on LinkedIn: Why Hiring More Reps Can Make Your GTM Worse

https://www.linkedin.com/pulse/why-hiring-more-reps-can-make-your-gtm-worse-jeff-swan-jmq3c

Watch the full episode on YouTube:

https://youtu.be/7QaWGlRfPfs

—

The Signal is a weekly podcast about what it takes to win in high-stakes, government-shaped markets. New episodes every Tuesday. If something in this episode helped, the best thing you can do is share it with one founder who's about to make the hiring decision.

Have a question for a future episode? Reply to this email or leave a comment below.

In this episode, Jeff walks through what happened when he tried to solve this problem for a flood mitigation company in the years after Hurricane Sandy. The RFP tools failed. The Prime networking worked sometimes. What actually produced a 2x revenue jump in five months was something else entirely.

He also connects the play to the marine work SOS Signals is running now, and explains why sub-contractors across transit, infrastructure, and public works can adapt the same approach.

What's covered:

- Why subscribing to ConstructConnect accelerated pain instead of growth

- The limits of Prime networking as a sole go-to-market strategy

- The moment city councils in Philadelphia and New Jersey started asking questions

- How 4,000+ global project case studies earned a shortlist seat before any RFP existed

- The paper trail that signals a marine RFP 6 to 12 months before it publishes

- The question every subcontractor should be able to answer about their buyers

Referenced in this episode:

ConstructConnect. Hurricane Sandy. 15-foot flood barrier case studies. The Signal podcast previous episodes on pre-RFP timing.

Ask from the episode:

What's your client's Hurricane Sandy? Send Jeff a note and tell him.

Last Thursday, I brought it back. I sent 20 emails through one channel, and every message was built from verified Signal data. I got the best reply rates of my career.

The rules are different now.

The Problem

The original Pipeline Day was built on volume, and for years it delivered. The energy was real, the concentration was real, and the team showed up together. I scaled this model across multiple clients and later into a community of over a thousand sellers through RevLeague, where we turned it into a daily power hour with leaderboards, competitions, and some of the most fun I have ever had in sales.

What I saw over time was that volume without intelligence eventually hits a ceiling. Automated outreach flooded buyer inboxes, and reply rates dropped across the board. The lists were not good enough to justify the effort we were putting into them, and the results became random even when the energy stayed high. I was watching talented people work harder on accounts that were never going to convert because nobody had mapped whether those accounts were actually in a buying window.

The Proof

Last Thursday, our internal team ran Pipeline Day using SOS Signals for the first time. I had 20 accounts. One colleague had 10, and another had 15. Every account was selected because the signals confirmed a buying window was open or about to open.

We chose email as the only channel. Every message was built from Signal Strategic Notes, which provided Tier-1 verified evidence of each account's buying window, incorporating specific mandates, funding changes, and program indicators. No templates and no automated sequences. Every outreach led with something the buyer did not expect us to know, and the tone was designed to feel like a partner sharing intelligence, not a salesperson asking for a meeting.

Intelligence applied to concentrated effort produced the highest reply rates I have seen in a single session of doing this work.

The Risk

I am seeing too many companies still running pipeline generation on lists they are not confident in. They are burning the motivation of their best people on accounts that may not have a buying window open for another year, and the results feel random because they are random.

The companies that figure out how to combine the energy of a concentrated Pipeline Day with verified Signal intelligence are the ones that will stop guessing and start showing up to conversations their buyers actually want to have. Volume is the grave of profit, and that includes the volume of effort aimed at the wrong accounts.

The Save

Pipeline Day is back, and the rules are different. The original version was a three-hour session built on volume. The Signal Edition took one hour, used 45 accounts across three people, and produced better results than any volume session I have ever run. That is a 66% reduction in time with a higher conversion rate, because the intelligence did the heavy lifting before we ever opened our inboxes.

You do not need 125 calls. You need five accounts with a mapped buying window, one hour of concentrated outreach, and a message that presents you as a partner who understands their business, not a vendor looking for a meeting. Lead with insights, earn the right to ask.

Over the next few episodes, I will be sharing the progress we are making with Pipeline Day, including what is working, what is not, and what we are learning about timing outreach to the buying window. If you want to try it yourself, start with the Buying Window Mapper from last week and use step four to build your Pipeline Day list.

In this episode

- The origin of Pipeline Day and how it started with my colleague Ian Selbie at a startup we were both consulting with

- How we scaled it through RevLeague into a daily power hour reaching over a thousand sellers, complete with leaderboards and GIF celebrations

- Why volume-based Pipeline Day stopped producing results as automated outreach flooded buyer inboxes

- How our team ran Pipeline Day: Signal Edition with 45 accounts across three people, one channel, and verified signal data

- Why concentrated effort on signal-backed accounts outperforms volume every time

- The systematic lie that volume plus automation equals pipeline

Resources

Download: The Buying Window Mapper (4-step worksheet from E05) [link]

Referenced: Ian Selbie and RevLeague (RevGenius community)

Previous Episode: E05, When On-Time Is Too Late (listen)

Listen and Subscribe

Substack: jeffswan18.substack.com

Spotify: [link]

Apple Podcasts: [link]

YouTube: Outbound SOS

Want to challenge for Pipeline Day? DM me on LinkedIn or drop a comment. I will set up the rules, and we will run it together.

The Signal is a weekly podcast by Jeff Swan, founder of Outbound SOS and creator of SOS Signals. Each episode covers what it takes to win in high-stakes, government-shaped markets.

Within an hour I had found a live negotiated RFP in Nova Scotia, a provincial mandate in New Brunswick, a $5.2 million contract in Ontario, and an $11 million contract in Newfoundland and Labrador. All of them aimed at solving the exact problem we were sitting inside of.

A broken toe turned into five buying windows. That is what happens when you know where to look.

The Problem

Every market has a buying window. A specific stretch of time when your buyer is actively looking for help and open to new ideas. Show up too early and the project does not exist yet. Show up too late and the shortlist is locked.

Most B2G teams have never mapped theirs. They spend six figures a year on pipeline activity and cannot name the month their top five accounts enter a buying window. They are responding to RFPs that were shaped by competitors who showed up months or years earlier.

The Proof

A mechanical contractor in the Pacific Northwest sells HVAC systems to mid-rise developers. Their buying window is 30 to 60 days between permit clearance and shortlist lock. Miss it on either side and the project is gone.

In the marine industry, coastal ferry operators run vessels on five-year dry dock cycles. Miss the planning window by six weeks and the next chance is half a decade away.

A fixture manufacturer was told "come back in 18 months." They stayed close, shared insights, and helped the buyer shape the procurement criteria. When the RFP launched, the criteria reflected their strengths. They were a lock for the shortlist.

The Risk

If you are responding to RFPs without having mapped your buying window, you are starting a race that was designed around someone else's capabilities.

In healthcare alone, the stakes go beyond lost revenue. In British Columbia, 58.9% of all violence-related workplace claims originate in healthcare and social assistance. When patients are left in the dark, the frustration lands on staff. Fixing information flow is now a safety mandate, not a service upgrade. The companies that win consistently are the ones that showed up during the window and earned their seat before procurement launched.

The Save

I walk through a four-step process for mapping your buying window. Reverse engineer your buying cycle. Identify where the influence window sits. Figure out what your buyer actually needs at that stage. Align your sales motions around those specific windows.

The Buying Window Mapper worksheet is linked below. Try it for your top five accounts and see what you learn.

What I Cover

- Why a broken toe in a hospital waiting room surfaced five active buying windows

- The 30-to-60-day window that decides everything in real estate development

- Why missing a marine dry dock window costs you five years

- The four-step process for mapping your buying window

- Why "on time" in B2G means you already lost

Resources

- Download: The Buying Window Mapper (4-step worksheet) [link]

- Sources: MERX (Nova Scotia negotiated procurement), New Brunswick Dept of Health 2024-2025 Mandate, Ontario Better Access Alliance release, Newfoundland and Labrador virtual care contract release, Health Canada bilateral agreements, WorkSafeBC Statistics 2023 [link]

- Referenced: Josh Braun on "commission breath" [webinar]

Listen & Subscribe

- Substack: thesignal.substack.com

- Spotify: [link]

- Apple Podcasts: [link]

- YouTube: Outbound SOS

Next Episode: We go deeper on how to identify the specific signals that tell you a buying window is about to open, and why the companies that track these signals consistently are almost never the ones responding to public RFPs.

---

The Signal is a weekly podcast by Jeff Swan, founder of Outbound SOS and creator of SOS Signals. Each episode covers what it takes to win in high-stakes, government-shaped markets.

I used to be in those lines two or three times a month. Vancouver to Croatia, Germany, and everywhere between, living out of a suitcase because I had no other way to get in front of my clients.

Here is the question nobody is asking: what happens to your pipeline when the person responsible for all of your key relationships is stuck in one of those lines?

The Problem

Most B2G teams have that one-star seller. They travel to four cities a week, attend every trade show, shake every hand. They are absolutely loved. They are also physically unable to be in two places at once.

While they are on a West Coast swing, deals pop up on the East Coast that they might not see in time. At roughly $1,200 to $1,400 per domestic business trip fully loaded, that cost compounds fast when it is aimed at accounts that are not in a buying window.

The Proof (It Works When Timed Right)

A global fixture manufacturer client of ours got the classic response from a buyer: "We love what you do. Come back in 18 months."

Most teams set a CRM reminder and walk away. This team did the opposite. They gathered facts, figures, and insights their buyer could use to shape the procurement criteria, and they shared them with zero stipulations. No "we will give you this if you include us in the RFP." They led with value.

By the time the RFP launched, they were a lock for the shortlist. They won because they knew which relationship was worth staying in and what that relationship needed from them before procurement started.

The Risk

I heard a story recently about a woman crossing the Canada-US border with commercial materials for a trade show. She declared everything, did what she thought was right. Turns out, bringing commercial goods through the pre-check lane violates the program rules. They denied her entry and revoked her pre-check status entirely, two weeks before her trade show.

If she were the only person positioned for that event, her company's entire GTM presence just evaporated over a compliance error.

The Save

At Outbound SOS, we track digital footprints that mimic the same level of interaction you would get from a physical trip. We map out who you would meet with, what you would talk about, and the Signals that tell you an account is moving toward a buying window.

Travel builds trust. Intelligence tells you where to invest that trust. Separate those two jobs, and you stop wasting $1,300 trips on accounts that are eighteen months from buying.

Sources referenced in this episode:

- GBTA/CWT Business Travel Forecast (2025/2026): NORAM average ticket price $831, hotel ADR $188/night, fully loaded domestic trip range $1,200 to $1,400 (gbta.org / mycwt.com)

- GBTA Business Travel Index Outlook (2025): Average trip spending $1,128 USD (gbta.org)

- Fixture manufacturer engagement: first-party client example

- Border crossing story: personal account

- Previous episodes referenced: Episode 3, the three-step AI account research framework.

Subscribe to The Signal on Substack, Spotify, Apple Podcasts, or YouTube.

Want to see what a signal-led approach looks like for your team? DM me on LinkedIn or reply to this post.

I kept looking at the three slides I built and thought,

"This doesn’t look like a year’s worth of work.”

The framework looks simple because I worked hard to make it simple, and there’s a meaningful difference between those two things.

What I can’t compress into three slides is the pattern recognition that tells you a Texas Hospital Association doesn’t belong on a production company list before the filter even runs. The framework gives you the container. The experience fills it.

I’ll let the episode do the rest of the talking.

👇 Full episode audio here 👇

What I cover:

Step 1, Purpose. Any process that runs without a clear mission statement is a failed process. I walk through how to define one that provides AI with a precise enough basis to actually work from, tied directly to a measurable business goal.

Step 2, Process. Before you touch a tool, map how your team finds this information manually. I show exactly what that looks like using a real workflow audit, including the eight steps most teams take today and the five roadblocks that make every one of them expensive.

Step 3, Validate. This is the step most people skip, and the most costly one to get wrong. I’ve personally paid for the data fix project that follows when you leave this gate open. Quality assurance between AI output and your CRM is not optional. It’s the whole point.

The three statements you should use every time you run a search:

Statement 1, Mission and Purpose.

“Our business goal is to grow revenue 10% year over year. To support that goal, this research task will identify Tier 1 government agencies in [geography/sector] that are within 18 months of a contract renewal in our solution category, with a budget above [threshold] and a decision-maker change in the last 24 months. Success means 20 qualified accounts per month entering the pipeline ready for seller engagement.”

Statement 2, Process.

“To find this information manually, we would: search USASpending.gov and SAM.gov for expiring contracts in our category, cross-reference agency websites to confirm mission alignment, search LinkedIn for leadership changes in the last 24 months, verify budget tier against publicly available appropriations data, and confirm no active incumbent relationship exists in our CRM. The information lives across four sources. The roadblocks are outdated contact data, inconsistent contract record formats, and the time cost of cross-referencing manually.”

Statement 3, Validate.

“This process runs on a defined weekly cadence. AI handles the initial search and cross-referencing across the four identified sources. A senior team member with B2G experience validates every account before it enters the CRM, confirming mission fit, timing signal, and contact accuracy. No account touches a seller without a source-verified reason to reach out and a clear opening based on what the agency is dealing with right now.”

The question I’m sitting with:

Every team I work with that tries to use AI for account research hits the same wall. The tool isn’t the problem. The brief is. And the brief is only as good as the person writing it.

The real question is:

“Do you have someone with enough B2G experience to write the brief, run the process, and catch what AI misses before it touches your CRM?”

Episode 4 goes there. Coming soon.

Resources shown in this episode:

AI Process Map, Workflow Audit: Google Sheet with all research steps mapped, timed, and flagged for automation or human review. Download link below.

Sources:

* The Bridge Group 2025 Sales Development Report: bridgegroupinc.com

* Strategic Proposals Benchmarker PDF: strategicproposals.com

* ERI Economic Research Institute Salary Benchmarks: erieri.com

* Construct Connect bid success rate data: constructconnect.com

* USASpending.gov: usaspending.gov

* SAM.gov: sam.gov

* First-party client engagement data: outboundsos.com

#NoVendors by Jeff Swan is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

In this episode, I walk through five approaches I've used personally, or watched clients use, and why each one eventually broke down. I'm not sharing theory here. I'm sharing the math I lived through, including one client story that stopped me cold when I ran the numbers.

Five people. Two weeks. 250 accounts. $16,400.

That's what it cost one team, in salary alone, to build a research list the hard way. $65.60 per account. 48 minutes per account, per person, across five of the most expensive people on their team.

I'm going to walk you through why that number is more common than anyone wants to admit, and why every "solution" most teams reach for makes the same core mistake in a different wrapper.

What I cover in this episode:

Buying software. I've subscribed to ZoomInfo, Apollo, GovSpend, Construct Connect, and more. The challenge is that by the time a public RFP surfaces in any of these platforms, your win rate as a cold responder sits between 10 and 20 percent. Pre-positioned incumbents close at 60 to 80 percent. The data is that clean. (Sources: Construct Connect, Strategic Proposals Benchmarker)

Outsourced research teams. Offshore, nearshore, every model. Every time I've run this, the management overhead and data quality issues ate the cost savings. I frame this one exactly as I experienced it.

The intern model. One to three months to ramp. Four to six month internship. Do the math on productive output before they move on. I built programs specifically to solve this and watched the same pattern repeat.

Senior reps and executives doing it themselves. This is the $90-per-hour floor I talk about on camera. ERI pegs a fully burdened Senior Director of Sales at $65 to $84 per hour. A VP runs $128 to $150 plus. Using that talent for account research is the most expensive mistake on this list.

The SDR and BDR model. The Bridge Group's 2025 Sales Development Report puts annual SDR turnover at 34 to 40 percent under standard conditions. In my experience across 80-plus GTM teams, the real number is closer to 50 to 100 percent when you account for probation washout. 53 percent of SDRs fail to meet expectations or survive initial ramp. Productivity plateau typically hits at 15 months. The math on productive output per dollar invested breaks the same way the intern model does, just slower and at higher cost.

The question I'm sitting with at the end of this episode:

Every approach I've described has the same structural flaw. The function of finding the right account, at the right moment, before the RFP drops, that work has to live somewhere. The question is whether it lives on your team, and what that actually costs when you're honest about the numbers.

I'll get into that in Episode 3.

Sources referenced in this episode:

The Bridge Group 2025 Sales Development Report: bridgegroupinc.com

Strategic Proposals Benchmarker PDF: strategicproposals.com

ERI Economic Research Institute VP and Director of Sales Salary Benchmarks: erieri.com

Loopio Annual RFP Response Report: loopio.com

Construct Connect bid success rate data: constructconnect.com

Bacon and Pugh, Powerful Proposals, AMACOM, referenced via APMP practitioner research

Episode 3 - coming soon:

Next episode, I'm walking through what actually works. Full process, real numbers, and a question at the end that every team I talk to has to answer for themselves before they decide what to do about it.

Key topics in this episode:

- Why government funding is cyclical by design and how sellers can anticipate the shifts before they happen

- How a $50M Canadian grant program grew to $530M across multiple iterations, proving that funding doesn't disappear, it transforms

- The practical impact of the Biden-to-Trump infrastructure policy shift and what it means for B2G sellers right now

- Jeff's personal framework for navigating change: Be the Water, Move with the Operators, and Realign. Don't Retreat.

Subscribe to #NoVendors for weekly Signals, strategies, and evidence-based insights for selling to government: https://jeffswan18.substack.com/subscribe