Jonathan Kvarfordt

Reality Check · 2026-09-19

Agentforce Pricing Explained: Credits, Licenses, and Total Cost

Agentforce is not one price. It is a stack of editions, entitlements, meters, and platform costs that only resolve into a number once you know which product you are buying. Here is how to work out which one you are looking at, and which question to ask next.

Reality Check · 2026-09-18

What Are Salesforce Core, Advanced, and Max Editions, and How Many Flex Credits Does Each Include?

On Sep 3, 2026 Salesforce published Core ($195), Advanced ($395), and Max ($550) per user/month for Agentforce Sales and Service, with org-level Flex Credit pools of 500,000 / 1 million / 2.75 million. Credits do not scale per seat. Legacy edition pricing stays for existing customers; Agentforce 1 can move to Max at no extra seat price.

Reality Check · 2026-09-18

Should We Sign an Agentic Enterprise License Agreement (AELA) or Stay on Flex Credits and Conversations?

Salesforce documents AELA as a fixed-price, fixed-term agreement with unlimited access to covered consumption services and no usage-based billing. Scope is agreement-specific. Compare the flat fee to your Flex Credit and Conversation forecast before you treat unlimited as settled.

Reality Check · 2026-09-17

What Is Salesforce Koa, and Is It GA for Agentforce?

Koa is Salesforce's first CRM reasoning model for Agentforce, post-trained from NVIDIA Nemotron 3 Super inside the Salesforce trust boundary. As of the Sep 15, 2026 press release it is available to select pilot customers only, with general availability expected winter 2026 in U.S. regions. It is a model-provider choice, not a new interface layer and not a Flex Credits usage type.

Reality Check · 2026-09-17

Does Agentforce Coworker Bill by Seats or Flex Credits, and What Makes a User Fall Back to Usage Metering?

Salesforce documents seat-based and usage-based Coworker pricing in the same org. Seat CRM and optional Slack search can be unmetered for eligible authenticated users. Missing the user-based AI permission flips seats into credit burn. Guest and portal users stay on usage. Data 360 sources add their own meters.

Reality Check · 2026-09-16

What Is AIforce, and How Does It Relate to Agentforce, Claudeforce, and Slackforce?

AIforce is Salesforce's live interface layer that brings CRM data, workflows, logic, and permissions to any AI interface. It sits on top of Data 360, Customer 360, and Agentforce. Claudeforce, Slackforce, and Agentforce Coworker are the first three surfaces. Open beta for Claudeforce is not GA; the product page is stale versus the Sep 15 announcement.

Reality Check · 2026-09-16

Which Agentforce Voice Meter Should We Put on the Order Form: Voice Actions or Voice Minutes?

Salesforce meters Agentforce Voice two ways: Voice Actions (30 Flex Credits) or Voice Minutes (60 per production minute). They are mutually exclusive on the Order Form. Break-even is action density versus talk time, not feature names. Telephony stays a separate line.

Reality Check · 2026-09-15

Do Agentforce Token Counts Equal Flex Credit Bills?

Salesforce's own August 18, 2026 post says the AI Usage Data Model provides token counts, not Flex Credits directly, and that Digital Wallet is the single source of truth for exact consumption, billing verification, and contractual overage calculations. Use the data model for near-real-time dashboards. Use Digital Wallet for the invoice.

Playbook · 2026-09-15

The Reversal Ledger: Counting the AI Decisions Your Team Had to Undo

Every revenue org tracks what AI did. Almost none track what humans had to reverse. That second number is the one that tells you whether your agents are earning trust or borrowing it.

Reality Check · 2026-09-14

When Does Agentforce Help Agent Bill 400 Flex Credits for a Resolution?

A billed Help Agent resolution costs 400 Flex Credits at Salesforce list ($2). Escalation, negative feedback, or walk-away is not billed. Sandbox resolutions are 0 on the rate card. Data 360 and Agentforce actions stay unmetered during the Help Agent session.

Reality Check · 2026-09-14

Which Agentforce Flex Credits Buying Model Should We Sign: PayGo, Pre-Commit, or Pre-Purchase?

Salesforce sells Flex Credits as PayGo, Pre-Commit, and Pre-Purchase. Cash timing, commitment risk, and unused-credit expiry decide the Order Form, not agent features. Start on PayGo until Digital Wallet shows a stable baseline; Pre-Commit needs a named true-up floor; Pre-Purchase only when discount, cash, and no-rollover are accepted on one worksheet.

Reality Check · 2026-09-11

Do Agentforce and Data 360 Share One Flex Credit Pool With No Upfront Allocation?

Flex Credits are fungible across Agentforce and Data 360. Salesforce says you do not allocate upfront. Data work can burn the same pool your Agentforce forecast assumed was yours.

Reality Check · 2026-09-11

How Many Flex Credits Does Salesforce Foundations Include, and When Do We Outgrow Them?

Foundations FAQ grants 200K Agentforce Flex credits to get started. Do not plan on the 450k marketing headline alone. One-time pilot tank, list math to about 10,000 standard actions, outgrow test before you treat Foundations as production budget.

Reality Check · 2026-09-10

Do Sandbox Agentforce Actions Burn Flex Credits, and at What Rate Versus Production?

Sandbox is not free. Salesforce's Flex Credits Rate Card applies a sandbox multiplier to pre-production environments, including scratch orgs. Standard and custom actions bill 20 credits in production and 16 in sandbox. Voice actions bill 30 and 24.

Reality Check · 2026-09-09

How Do We Chargeback Agentforce Flex Credits to Business Units When Digital Wallet Reports at the Org Level?

Digital Wallet is org-level billing visibility, not BU chargeback. Tag actions and cohorts by cost center before scale, allocate month-end burn by tagged share, and name Finance for the journal entry and RevOps for the allocation table.

Reality Check · 2026-09-09

At What Actions Per Conversation Does Flex Credits Beat the $2 Conversation SKU?

Run the list-price arithmetic before the AE call. At Salesforce list, a standard action is $0.10 and a conversation is $2. Break-even is 20 standard actions. Below 20, Flex Credits win. Above 20, Conversations win. Org cannot mix models.

Reality Check · 2026-09-08

Which Agentforce SKU Are We On: Flex Credits Only, or Einstein Requests Plus Flex Credits?

Ask the AE which meter your prompts hit before you build the forecast. A1E/A4X burns Flex Credits only. E1E/E4X and some Foundations paths still split Einstein Requests and Flex Credits. One SKU identity question under the forecast.

Reality Check · 2026-09-08

The Forecast Call After AI: What Breaks When Agents Touch Your Pipeline

AI now summarizes calls, updates stages, and nudges next steps. Great. Except your forecast call was built on the assumption that a human rep's judgment sat behind every field. Here is how to rebuild forecast integrity when the judgment is shared with a machine.

Playbook · 2026-09-08

Renewal and Expansion Planning When Agents Watch the Account

Risk detection can be automated. The renewal conversation cannot. Here is the split, the four signal classes worth acting on, and a 120-day renewal plan that says exactly which steps a machine owns.

Playbook · 2026-09-08

Will This AI Category Still Exist in 2027? A Durability Test for Revenue Buyers

Some AI categories become platform features within a year. Others harden into infrastructure. Five durability factors, a scoring method, and the contract terms that match each score.

Reality Check · 2026-09-07

What Rollback Plan Must We Require Before an AI Agent Touches Production Pipeline?

Write the rollback plan before the agent can write, send, or stage anything. Kill criteria stop the spend. Rollback reverses the blast radius. Seven action classes, one-page plan, named freeze owner.

The Teardown · 2026-09-07

The Signal Stack: What Common Room Buyers Learn After Go-Live

Signal platforms find intent. They do not hand you contacts, and they do not act. Here is the four-layer signal stack, the two costs buyers underestimate, and what the Zoom acquisition changes for renewals.

Playbook · 2026-09-07

The 30-60-90 Ramp Plan When New Reps Work Alongside Agents

Ramp plans built for a world of scarce information break when research, drafting and summarising are free. Here is a rebuilt 30-60-90 with competency gates, supervision skills, and the two things that must not be automated in week one.

Reality Check · 2026-09-07

Optimization Theater: Five Tells That Your AI Dashboard Is Lying to You

The metric improved and the business did not. Five reliable tells, the counter-question for each, and the one-page reporting format that makes theater impossible to sustain.

Playbook · 2026-09-06

AI Supplier Shortlisting: The Nine-Question Diligence Pass That Cuts 40 Vendors to 3

A shortlist is an elimination process, not a research project. Nine questions, five gates, and a scoring sheet that gets a revenue team from a crowded category to three defensible finalists in two weeks.

Benchmark · 2026-09-06

The AI Adoption Curve for Revenue Teams: L1 to L6, and Where Most Teams Stall

Six levels from personal experiments to a system that runs without heroes. Each level has an entry test, a failure mode, and a single exit criterion. Most revenue teams are at L2 and reporting L4.

Benchmark · 2026-09-06

The AI Revenue Team Org Chart: Which Roles Change, Which Appear, Which Go

Agents do not flatten the org chart. They move decision rights. Here is the before and after for seven revenue roles, the three jobs that appear, and the reporting line that decides whether any of it works.

The Teardown · 2026-09-05

Ebsta Reviewed: What the Revenue Intelligence Claim Actually Measures

Ebsta sells engagement data as forecast accuracy. Those are two different products. Here is what the review base supports, what the Fullcast acquisition changes, and the eight questions to ask before you swap a forecasting platform for an engagement layer.

The Teardown · 2026-09-05

11x After the Walk-Back: What Buyers Should Ask Now

The autonomous SDR category had a public credibility event and the vendor is still selling. Here is the documented timeline, what changed, and the reference-check script to run before any AI SDR contract reaches signature.

Reality Check · 2026-09-04

Does Digital Wallet Stop Agentforce When Credits Run Out, or Do We Need Our Own Kill Switch?

Digital Wallet alerts. It does not block. Salesforce says overages bill at contracted rates. Here is the RevOps control plane: Flow thresholds, named owners, and a written kill rule before production scale.

Reality Check · 2026-09-03

Who Audits the Meter on Our AI Agents, and What Audit Rights Do We Need Before We Sign?

Nobody audits it unless you write the right in. On a consumption AI contract the vendor defines the billable unit, detects it with its own system, and holds the only full-fidelity record. Cap vs audit, clause set, parallel count.

Reality Check · 2026-09-03

How Do You Forecast Agentforce Flex Credits When Overages Bill in Arrears?

Flex Credits look granular until month-end. Credits expire, overages invoice in arrears at the contracted rate, and Digital Wallet still has no kill switch. Here is the RevOps forecast-and-close model before the invoice contradicts the pilot case.

Reality Check · 2026-09-02

Do We Need Two Contracts to Buy Claudeforce, and How Do We Budget Two Meters?

Yes. Salesforce meters headless consumption. Anthropic bills inference on a separate contract. Stokes said you cannot buy it on one piece of paper. Two meters, two owners, no invented Claudeforce seat price.

Benchmark · 2026-09-01

The Real Cost of an AI SDR Is Not on the Pricing Page

Vendors sell you a seat price. Your P&L pays for list decay, deliverability repair, management overhead, and the meetings that never should have been booked. Here is the math that actually matters.

Reality Check · 2026-08-27

Wait, Buy, or What to Ask the Salesforce AE Before Dreamforce?

Claudeforce / Salesforce in Claude is open beta for all customers as of Dreamforce day 1 (Sep 15, 2026). Still no GA, no list price, dual-path with Agentforce. Four questions to put to the Salesforce AE in writing.

Reality Check · 2026-08-27

What Truth Tests Should We Run Before Buying Claudeforce, Agentforce, or an AI SDR?

AI on a dirty CRM is a prettier lie. Five truth tests a Director+ can run before Claudeforce, Agentforce, or AI SDR spend. Definitions and freshness first. Then the pilot.

Playbook · 2026-08-27

What Kill Criteria Should We Set Before Signing an AI SDR Contract?

Write the fail conditions into the order form before the demo becomes a two-quarter argument. Hybrid 1.9x, reply-rate floors, named redeployment, no headcount-cut ROI story.

Playbook · 2026-08-22

Just in Time Enablement: What AI Gives a Revenue Team That Nobody Had Before

Enablement never lacked intent. It lacked data and bandwidth. What changes when coaching, content, and practice can be measured on every rep, every week.

Reality Check · 2026-08-21

Systems Beat Talent: Why Good AI Fails Inside a Mediocre Revenue Process

Put a great rep in a mediocre system and you get mediocre results. AI follows the same rule, only faster. What to fix before the next deployment.

Playbook · 2026-08-20

The Editor's Mind: How to Review AI Output Without Trusting or Dismissing It

Most AI disappointment is a prompt problem or a review problem. A working method for critical review, and why outlandish claims in both directions deserve the same scrutiny.

Playbook · 2026-08-19

Text to Action, Not Text to Content: What Changes When Agents Execute

The shift is not better output. It is AI doing the next step. What a revenue leader has to own when the tool stops drafting and starts executing.

Playbook · 2026-08-18

If AI Did Half the Work, Who Gets Paid for the Deal?

Comp plans assume a human did the sourcing, the qualifying, and the closing. Agents now do parts of all three. Here is how to redesign quota and territory without detonating trust in the field.

Reality Check · 2026-08-18

Why Does the Same AI Stack Produce Different Results at Two Companies?

AI is power. Everyone has access to the outlet. The gap between two teams running the same model is domain expertise, systems, and process, not the model.

Reality Check · 2026-08-11

Bet on the Team, Not the Tool: Three Converging Shifts Your GTM Motion Is About to Be Tested On

A software selloff, agent sprawl, and buyers who shortlist you before your rep picks up the phone. Three stories, one root cause, and three structural decisions that make all of them easier to absorb.

Benchmark · 2026-08-11

Your Partner Channel Is the Least Instrumented Revenue Motion You Own

Direct sales got agents, dashboards, and forecast rigor. Partner-sourced revenue is still run on spreadsheets and goodwill. That gap is now a competitive problem, not an admin one.

Benchmark · 2026-08-04

4.8 Hours Saved, 72 Percent Wasted: The Reinvestment Gap Where AI Revenue Goes to Die

AI is giving sellers nearly five hours back every week and most of it disappears into low-value work. The organizations closing the gap are making two structural moves at once, not one.

Playbook · 2026-08-04

Nobody Owns AI in Your Revenue Org, and It Shows

AI in GTM sits between RevOps, IT, enablement, and whichever leader moved first. Ambiguous ownership is why capability stalls at the pilot line. Here is an ownership model that actually holds.

The Teardown · 2026-07-28

The Single-Player Problem: Why Your AI Wins Never Leave the Room They Were Built In

Only 4 percent of revenue teams describe their AI work as fully integrated across functions. The other 96 percent have pockets of brilliance surrounded by organizational silence. Four structural decisions fix it.

Reality Check · 2026-07-28

Everyone Published More. Nobody Got More Pipeline.

AI removed the cost of producing content, so the entire category produced more of it. Attention did not expand. Here is what actually creates demand when volume is free and credibility is not.

Reality Check · 2026-07-21

What Steam Engines Teach Revenue Teams About AI: Capacity Lift, Not Headcount Cut

80 percent of companies cutting staff for AI see no ROI advantage from the cuts. The ones winning redeployed their people toward growth that was never reachable at human cost. Here is the audit that finds it.

The Teardown · 2026-07-21

Seats, Consumption, or Outcomes: How to Negotiate AI Pricing You Can Forecast

AI vendors are moving off per-seat pricing and onto models that make your spend a variable you cannot forecast. Here is how each model actually behaves at scale, and the terms to fight for.

Playbook · 2026-07-14

You Already Run a Loop. The Question Is Whether It Learns.

Loop engineering is the new buzzword and your forecast is the original version. Before you ask your team to build one, know exactly when AI belongs in a GTM loop and when it does not.

Reality Check · 2026-07-14

AI Made Your Health Scores Confident. It Did Not Make Them Right.

Churn prediction models are now standard in customer success. Most of them learned from a book of business that no longer exists. Here is how to tell whether yours predicts renewal or just describes usage.

Benchmark · 2026-07-07

The 5 Percent Problem: The OAR Matrix for Getting Real Revenue Out of AI

59 percent of companies spend at least a million a year on AI. Only 5 percent generate value at scale. The gap is a decision problem, not a deployment problem, and the OAR Matrix is how to diagnose which side you are on.

Playbook · 2026-07-07

Enablement Is Not a Training Problem Anymore. It Is a Judgment Problem.

When agents draft the email, summarize the call, and suggest the next step, the scarce skill stops being execution and becomes knowing when the machine is wrong. Enablement has to be rebuilt around that.

The Teardown · 2026-06-30

Do This or Die Slowly: Building a PLG Motion Inside a 25-Year-Old B2B Company

A CRO at a PE-backed, 25-year-old platform said it plainly at one of our meetups. Here are the four decisions that separate companies actually building a second front door from the ones still planning it.

The Teardown · 2026-06-30

Twenty Questions That Separate an AI Vendor From an AI Demo

Every GTM AI vendor demos well. Diligence is how you find out whether the product works on your data, in your process, at your risk tolerance. These are the questions that change the answer.

Reality Check · 2026-06-23

Your Buyers Have Already Decided: What to Do When AI Builds the Shortlist Before You Know They Exist

94 percent of B2B buyers use AI during their research and 92 percent arrive with a shortlist already formed. The 61 percent of the journey that happens before a rep is involved belongs to nobody in most orgs. Here is how to take it back.

Playbook · 2026-06-23

Your CRM Was Built for Reporting. Agents Need It to Be True.

AI agents do not tolerate the fuzzy data humans quietly route around. Here is the readiness assessment to run before you point an agent at your pipeline, and what to fix first.

Playbook · 2026-06-16

You Are Not Behind. You Are Just Not Shipping: A Framework for Sequencing Your AI Backlog

97 percent of enterprises have deployed AI. 29 percent see significant ROI. The difference is not tooling, it is sequencing. Compound Moves, Capability Moves, and Noise Moves, with the receipts from teams that shipped.

Post-Mortem · 2026-06-16

Why Your AI Pilot Worked and Your Rollout Did Not

The pilot ran on your best rep, your cleanest data, and your most motivated manager. Production runs on none of those. Here is what breaks between the two, and how to design a pilot that survives contact with the org.

The Teardown · 2026-06-09

From Org Chart to Operating Intelligence: The New GTM Operating Model

The B2B revenue pyramid was built to move information through humans. AI does that job better. Here is the four-layer operating model that replaces it, and the five questions that expose how much of your org chart is still a workaround.

The Teardown · 2026-06-02

Your Org Chart Is a 160-Year-Old Workaround. Here Is the Migration.

Hierarchy existed because information had to route through humans, and humans have limits. Your agents do not. The revenue orgs redesigning around that now become structurally hard to catch.

Playbook · 2026-05-26

The Exponential Growth Playbook: Five Compounding Patterns and a 90-Day Roadmap

Clay went from 1 million to 100 million ARR in two years. The gap is not product-led growth. It is compounding systems, and there is a specific 90-day path to close it.

Reality Check · 2026-05-19

You Are Staffed for Stages. Revenue Lives in Moments.

78 percent of B2B buyers purchase from the vendor that responds first. Your org chart is built around stages, but the deal is decided in moments no human is scheduled to cover.

The Teardown · 2026-05-12

The Headless GTM Stack: What to Protect When Every System Becomes Callable by Agents

Salesforce is the signal, not the story. When your data, workflows, and business logic become callable from anywhere, the interface stops being the question and your operating model becomes it.