AIforce was framed as AI replacing the UI. A Salesforce president demonstrated a working interface he had built by talking to Claude over a few days. The pitch is that people will update records and trigger workflows without ever opening the Salesforce app.

Enterprise software is admitting that the screen it spent twenty five years perfecting is no longer where the work happens. Now apply that to the thing you actually sell. If the buying journey increasingly runs through an agent that reads, summarizes and recommends, then the website, the campaign landing page and the beautifully art directed product page are all being mediated by something you did not design.

So what. Add one slide to every credentials deck that asks the client where their brand shows up when nobody opens the app. Most agencies have not asked it yet. The first one in the room who does owns the conversation.

Salesforce published its architecture five days before the keynote and named the thesis plainly. AI reasoning can be open ended, while enterprise execution often cannot be. The model layer, the cloud layer and the interface layer were all made into choices. The governed business context was not.

This is the single most useful idea of the week for an agency. Your tools are swappable. Your AI stack is swappable. Whatever you paid for last quarter will be free next quarter. What is not swappable is the accumulated, specific, hard won understanding you hold about a client's business, its customers, its internal politics and what actually sells.

So what. Stop pitching your stack. Start pitching your context. The credentials slide about your AI tooling has a shelf life of about six months. The slide about how deeply you understand the category does not.

The Agentforce keynote was almost entirely customers and numbers. Eighty thousand hours saved at Fulton Bank. Three hundred and eighty nine million dollars in loans and deposits attributed. Four days off complaint resolution. Thirty thousand customers live. Salesforce Ben had argued going in that the one job of the week was proving the thing works.

Two years ago a demo was enough. Now the biggest software company in the category cannot get through a keynote without a scoreboard. Your clients are in the same mood. The era of an agency winning on a beautiful idea with no measurement story attached is closing, and it is closing fastest in exactly the mid market accounts indies fight for.

So what. Every case study in your deck needs a number and a source. Not a vanity metric. A business number the client's CFO would recognize. If you cannot get one, call the former client and ask. Most will tell you.

Adecco piloted Agentforce Coworker in the UK and France, then announced a rollout across more than forty countries. It was held up all week as the model, and specifically as the opposite of the common failure pattern where a proof of concept on clean data gets declared a win and scaled sideways.

Indies have always been told to sell the big transformative engagement because that is where the margin is. That advice is now backwards. The buyer has been trained, publicly, at scale, to expect a small proof followed by an earned expansion. Selling the pilot is no longer selling small. It is selling the way they have decided to buy.

So what. Build a productized ninety day pilot with a fixed price, a named outcome and a defined decision gate at the end. Lead with it. The holding company cannot price it and will not bother trying.

The most concrete thing announced all week was seven agents with human names and job descriptions. Casey. Paige. Carter. Marshall. Piper. Fin. Hunter. Each mapped to a role a buyer already staffs and budgets for. They got more traction in the coverage than the architecture underneath them.

Agencies chronically overpitch capability and underpitch the named, scoped, staffed thing. We say strategic partnership when we should say a senior strategist, two days a week, owning your category intelligence. Specificity is not a downgrade from vision. It is what makes vision purchasable.

So what. Go through your offering list and give every service a name, an owner and a job description a procurement person could put on a purchase order. Watch your close rate move.

Salesforce showed a long horizon agent runtime built as a cycle of goal, plan, human in the loop, signal and continuous refinement. The message was that agents can work toward bigger goals over longer periods while people keep the ability to step in, redirect and approve.

The most valuable company in enterprise software just told the market that human judgment is a designed component of the system, not a workaround for when the machine fails. That is the permission structure independent agencies have been waiting for. You are allowed to charge for judgment again, out loud, without sounding like a Luddite.

So what. Put the humans back on the org chart slide with their names and faces. Describe where in the process a person intervenes and why that intervention changes the outcome. That is the sale.

The sharpest practitioner commentary of the week argued that the refusal set is the part teams skip and the part that determines whether a deployment survives its first bad month. Not a vague instruction to be helpful and safe, but an enumerated list of what the system must never do, with a test proving each refusal fires.

Every agency has an AI policy slide. Almost all of them are aspirational mush about augmenting creativity. None of them say what the shop will never do. No synthetic likeness of a real person without consent. No AI generated talent replacing booked talent. No model trained on a client's proprietary assets without a written agreement. No unlabeled synthetic imagery in earned media.

So what. Write your refusal set this month. Four to eight lines. Put it in the pitch deck. It will be the most memorable page in the room, because it is the only one that costs you something.

The consistent practitioner warning all week was that announcements arrive at vendor pace while adoption happens at the pace your data and permissions allow, and those two speeds have never matched. The recommended first move was an audit, not a platform decision.

Here is the agency version. When a client's AI systems start speaking on behalf of the brand, somebody has to author the layer those systems are grounded in. Tone. Claims that are legally cleared. Product truths. Category vocabulary. What the brand refuses to say. Right now that work is being done badly, by nobody in particular, in a shared drive.

So what. This is a brand governance retainer with a new job to do, and it is the most defensible recurring revenue available to an independent agency right now. Name it, scope it, price it, and sell it before a consultancy does.

Salesforce's applications president told the room that the company's real product is the trust customers place in it to hold their data, workflows and security rules. Benioff spent a full segment on zero data retention and said flatly that customer data is not going into models. That was not a footnote. It was a positioning move.

When capability becomes commodity, trust becomes the differentiator. That is the whole thesis of Relationship Selling, arriving now as the explicit strategy of a ninety billion dollar company. Independent agencies have an enormous structural advantage here and mostly waste it by talking about process instead.

So what. Your senior people are in the room, on the work, for the whole engagement. That is a trust claim, not a service claim. Make it early, make it specific, and be willing to write it into the contract.

Three of the most powerful people in AI took three different positions on the same stage in the same afternoon. Slow down and regulate. Self regulate and move fast. No new rules at all. The disagreement was the story, and it made mainstream news.

Marketing leaders are now expected to have a position on this in front of their boards, their employees and increasingly their customers. Most do not have one. They have a legal review and a vague feeling. That gap is the single best chemistry meeting opener available to you this quarter.

So what. Develop your agency's actual position and be able to defend it in ninety seconds. Not a fence sitting both sides answer. A real one. Clients hire conviction, and there is almost none of it on this topic right now.

Four of the most consequential items landed in the weeks before the keynote. Any recap that started on September 15 missed the architecture the whole week was built on. Salesforce published its thesis five days early, then spent three days demonstrating it.

That is a pitch orchestration lesson and a very good one. The frame goes out before the room. By the time everyone is in the theatre, you are not introducing the idea, you are confirming it. Agencies still treat the pitch as a reveal, hoarding the thinking for a ninety minute performance and hoping the room assembles it in real time.

So what. Land your organizing idea with the client before the final presentation. A memo, a call, a single provocation. Let them arrive already holding it. Then the room becomes the moment they agree, instead of the moment they decide.

Salesforce renamed Marketing Cloud to Agentforce Marketing last year, and by this Dreamforce the products were listed under their original names again. Coverage this week had to explain the reversal. Separately, at least one published transcript rendered Claudeforce as CloudForce.

Agencies do this constantly. We rebrand our methodology every eighteen months, invent a proprietary name for a thing the client already understands, and then wonder why the deck does not travel inside the organization after we leave the room.

So what. Audit your own vocabulary. If a client cannot repeat the name of your process to a colleague without the deck open, the name is failing. Plain beats proprietary almost every time.

Forty three thousand people. One thousand four hundred and thirty published sessions. The most talked about moments of the week were two CEOs wandering through the crowd riffing at each other, and an animated cartoon of a co founder popping up in the corner of a demo screen.

Salesforce spent a fortune on the biggest software event in the world and the things that cut through were human, unscripted and slightly ridiculous. Meanwhile a session on the program named attention as the scarcest resource in business. Both of those facts are the same fact.

So what. Look at your next pitch through this lens. What in that ninety minutes is genuinely unrepeatable by anyone else in the process. If the honest answer is nothing, you have a deck, not a pitch, and you are going to lose to whoever shows up as a human being.