The Value Creation ProtocolContents
Part II — The two foundational traps
Chapter 6
The SaaS Trap → Value-First Context
My favorite part of this work is watching someone do their job.
You take a person who has been in a role for a long time and you ask them to describe their process. It's three steps. Then you watch them do it. They open ten different applications. They've got a paper binder over here. They're calling over to another department to get something. It's twenty-five steps. Over the years they built up little context clues all over the place, and then we try to train somebody new without any of those clues, and we wonder how the first person ever remembered it all.
That is the SaaS Trap, seen from one chair.
What this is
The SaaS Trap, as the methodology defines it:
Fragmenting the context your people rely on — where each rational tool purchase silos the intelligence needed to serve customers.
The word to notice is rational. Nobody set out to buy a bad tool. Every purchase solved the problem in front of the person who made it. The methodology puts the promise and the result side by side:
The SaaS revolution promised specialized tools that integrate seamlessly. Instead, each rational purchase fragments the common sense intelligence teams need to serve customers effectively.
To get a sense of the scale: the 2025 marketing technology landscape, published by chiefmartec, counted 15,384 solutions. That is marketing software alone. The methodology's picture for what this does inside a company is a chain of islands, and it is blunt about what connecting them has achieved so far:
Digital transformation promised to connect the islands; mostly it just automated the fragmentation — faster APIs moving data between islands that still can't see the customer the person on the next one is trying to serve.
This is one of the two foundational traps. The B2B Trap, in the last chapter, makes it normal to treat people as objects. This one breaks up the context you would need to treat them as anything else. That is why the pair comes first.
Its way out is Value-First Context. The principle:
Unified context enables better decisions than fragmented point solutions — and unifying context is much more than integrating tools. Context is not an input to operations; it is the substrate on which operations run.
The middle of that sentence is mine, and I'd underline it. Unifying context is much more than integrating tools. And the instruction that follows from it:
Stop managing individual tools and start preserving and multiplying the contextual intelligence your frontline teams have already developed.
What does unified context hold? Four things: relationship history, preference signals, prior commitments and current state. Two of the Protocol's claims in Chapter 1 stand behind this chapter. Context as Substrate says that "fragmented context produces fragmented operations regardless of methodology overlay." Just-In-Time Over Just-In-Case names what buying against a hypothetical need produces: "platform sprawl, integration debt, and operational complexity without proportionate value." And the belief to keep beside you is Wholeness over Fragmentation.
This is not the ERP Trap. People mix the two up, because both are about software. They pull in opposite directions.
| The ERP Trap | The SaaS Trap | |
|---|---|---|
| Definition | "Forcing business processes to conform to rigid systems rather than enabling natural operational flow" | "Fragmenting the context your people rely on — where each rational tool purchase silos the intelligence needed to serve customers" |
| The shape | One rigid system. The work bends to fit it. | Many tools. Each fits its own job. The context is scattered across them. |
| One of its signs | "Business processes redesigned to match system limitations" | "Every new tool solves one problem while creating three coordination problems" |
| Way out | Value-First Platform | Value-First Context |
The methodology calls the ERP Trap the "monolith counterpart" to this one. You can be in both at once. And the cure for one can be a road into the other, which is the first thing to watch for at the end of this chapter.
How to tell where you are
The hard part about this trap is that nobody defends it. Everybody agrees fragmentation is not good. And we are still bad at solving it, because it has become the problem we've just accepted: different teams, different KPIs, systems that don't talk to each other, and nobody making them talk, because each team cares about its own thing.
So you have to go and look. These are the trap's five signs, in the methodology's words.
- "Integration costs exceed software costs." Ask whoever holds the budget for two numbers: what you pay for the tools, and what you pay to make them work together. Include the people whose job is mostly keeping them in step.
- "Best insights stay trapped in individual experience." Do what I described at the top. Ask someone experienced for their process, then watch them do it. Write down every place they go that wasn't in what they told you.
- "Every new tool solves one problem while creating three coordination problems." Take the last tool your company bought. List what now has to be kept in agreement with it, and who does that.
- "Teams maintain shadow systems to actually do their jobs." Ask to see the spreadsheet. Ask kindly. It exists because someone needed the whole picture and built it themselves.
- "New employee onboarding means learning 12+ different systems." Count the logins your most recent hire was given in their first week.
Then look for the quiet version. A colleague described one on a show we did together in the summer of 2026. Two systems were supposed to stay in step. One of them held four records for a customer's account, and the other held one. A rule in the second system would only open a case if the account had nothing else open against it, and whichever of the four records it matched broke that rule. So the process died. No case was created. Nobody was told that it failed.
Nothing in that story is a bad tool. Each system was doing what it was set up to do.
What to do
Start with one relationship, not the tool list. Pick one customer and the person who serves them. Sit beside that person and watch them prepare for a conversation with that customer. Don't ask for the process. Watch it.
Write down every place they go. Every application, every inbox, every spreadsheet, the binder, the colleague they call. Next to each, write what they got from it. That list is the context this relationship actually runs on.
Sort what they gathered into the four things. Relationship history. Preference signals. Prior commitments. Current state. Whatever doesn't fit one of the four, ask whether they needed it. Whatever is missing from one of the four, ask where it lives. Often it lives in their head.
Ask for the view that already exists. It usually does, somewhere: a spreadsheet, a deck someone rebuilds before every big meeting. Treat it as the map. The person who built it has already done the work of deciding what matters.
Agree the view before anyone buys or builds anything. Get the people who serve the relationship to agree, in writing, what they need to see for the decision in front of them. This is the Unified Customer View, and Chapter 19, on the Four Unified Views, teaches it properly. What matters here is the order: agreement first, tools second.
Give every tool one test from now on. Does it move the view? That goes for the next purchase, the next feature and the next AI assistant. If it doesn't bring the agreed picture closer for the person doing the work, it is one more island.
Prefer what you already have. Before adding anything, see how far configuring what you own will take you. When you do have to choose between tools, weigh how well each one joins the agreed picture above how many features it has. Fewer places to look beats a better place to look.
How you know it worked
The person serving the customer can answer two questions without leaving the place they work and without asking anyone: who am I talking to, and where do things stand?
That's the first half. The second half is the customer. They notice that they didn't have to re-establish who they are. They don't get asked for something they already gave. A commitment someone made to them last spring is honored by a person who wasn't there when it was made, and the customer says so.
There is a third voice worth hearing. Ask your newest person, a week in, where they started. If they started from what the last person knew, the context is being held somewhere other than in heads. The AI-Native Shift has a name for that question, the Independence Test, and builds on it at length.
What won't tell you: the number of tools retired, integrations built or logins removed. You can retire six tools and still have someone opening a binder.
Where it goes wrong
- Swinging into the ERP Trap. "Too many tools" gets answered with one big system that everyone must conform to. Now the work bends to the software. You've traded scattered context for a rigid process, and people will build shadow systems around that too.
- Buying your way out. Something will be offered to you that promises to unify everything else. It may be good. It is also one more rational purchase, and that is how you got here. Agree the view first, then see what moves it.
- Making integration the goal. Data flowing between systems is not context reaching a person. You can connect every island and still have nobody who can see the customer.
- Punishing the shadow systems. The spreadsheet is not the offense. It is the best documentation you have of what the official systems leave out. Shut it down before you've learned from it and the knowledge goes back into someone's head.
- Fragmenting again with AI. An assistant inside every tool, none of which knows what the others know, is this trap at a new layer.
- Calling it something else. There is no "Integration Trap" and no "Software Trap" in this methodology, and the way out is Value-First Context, never a new architecture for its own sake.
Read further
- A Value-First Guide to Surviving the SaaSpocalypse (Book 1) tells this trap as story: what happened to the software we built our operations on, and what it proved.
- The AI-Native Shift (Book 3) returns to it in three chapters. Chapter 2, "The Org Chart Is a Confession," names the compounding fragmentation. Chapter 8, "The Platform Question," takes on the habit of buying the best tool for each job. Chapter 13, "The Second Person," shows the same trap forming again around AI.
- The Value Era (Book 0), Chapter 14, "Only the person who received it," is why the test here ends with the customer.
- In this handbook: the ERP Trap is Chapter 13. Chapter 19, on the Four Unified Views, teaches the view you agreed in the fifth move. Chapter 1 has Context as Substrate in full.