They Consolidated the Platform. Now They Want to Consolidate Your AI.

Two things happened in association technology this year that most buyers filed under "innovation." In January, Momentive Software announced its acquisition of Personify and, the same week, launched MomentiveIQ, an AI layer stretched across fundraising, membership, events, and learning.¹ The combined company now sits under one roof serving more than 37,000 organizations, its fourth acquisition after VolunteerMatters, Cobalt, and Blue Sky eLearn.² This month, Bloomerang switched on Dataro's predictive donor intelligence inside its Giving Platform, the first of what it calls several intelligence-native features to come.³

I do not think either move is bad technology. Predictive donor models and agentic analytics are genuinely useful, and I would rather my analysts spend their time on judgment than on exporting spreadsheets. What I want to flag is the pitch underneath them, because it is the same pitch the industry has been making for twenty years, and this year it put on a new costume.

The old argument, rebranded

For most of the last decade, the reason to buy the whole suite was operational. One vendor, one contract, one support line, one integration you never have to maintain. I argued in the first issue of this column that the platform was consolidating and your architecture should not follow it there. The counterargument was always convenience.

AI changes the argument, but not in the direction the marketing suggests. The new pitch is that intelligence only works when everything lives together. If your members, donors, events, and learning records all sit in one vendor's tables, the story goes, then the AI can see across all of it and hand you the unified view. Fragmentation is the enemy and the suite is the cure. MomentiveIQ is sold, in almost these words, as turning fragmented technology into a single intelligent experience.

Read that carefully, because it quietly redefines the problem. The thing making your data hard to use was never that it lived in different systems. It was that it lived in different systems with no shared definition of a member, no governed layer underneath, and no clean way to query across it. The suite solves that by moving everything into one place it controls. A governed data layer solves it by putting everything under definitions you control. Those are not the same solution, and only one of them is still yours the day you change vendors.

Lock-in wearing an AI costume

Analysts have spent 2026 warning that AI lock-in is worse than the ordinary software kind, because it compounds.⁴ It is not only your records that get stuck. It is the models tuned on your history, the workflows built around one vendor's agents, the dashboards your staff learned, and the institutional muscle memory of "the way our platform does it." When the intelligence lives inside the suite, every one of those becomes a reason you cannot leave, and none of them shows up as a line item you can price. API dependency means your architecture quietly bends around one vendor's design choices.⁵

For an association this is a sharper problem than for a generic business, because the asset is member-contributed data. Your members handed you their history on the understanding that you would steward it. Handing the intelligence layer to whichever vendor happens to own the most of your stack is a stewardship decision, not a procurement one, and it usually gets made by default, in a renewal, with nobody in the room framing it that way.

Keep the intelligence, skip the cage

The good news is that the open direction and the smart-technology direction are, for the first time, pointing the same way. The Model Context Protocol, the emerging standard for letting AI systems reach data and tools, was handed to a Linux Foundation body late last year and is finalizing its next specification this year.⁶ The whole premise of it is that the intelligence and the data do not have to belong to the same vendor. An agent can read your governed warehouse through an open interface without that warehouse being absorbed into anyone's suite.⁷

That is the architecture I would build toward. Put your system of record in a warehouse you own, not in whichever application happens to hold the most rows. Govern the definitions there. Then let the vendors' AI features connect to that layer instead of the other way around. You still get the predictive donor score and the agentic analyst. You just get them on top of data that stays portable, defined, and yours. The composable-stack argument I have made before was about resisting the monolith for flexibility. The AI wave raises the stakes, because the monolith is no longer just where your records live. It is where your judgment gets encoded.

Quick takes

Four deals, eighteen months, one roof. Personify was Momentive's fourth acquisition in that span, following VolunteerMatters, Cobalt, and Blue Sky eLearn. It is worth pausing on how quickly "the whole stack from one vendor" went from a slide to a fact of the market. The AI layer is being poured on top of a base assembled at acquisition speed, not designed as one thing.

The open standard grew up. The Model Context Protocol moved to a Linux Foundation body at the end of 2025 and its next specification lands this year, with thousands of public connectors already in the wild. Two years ago "keep your AI independent of your platform" was a whiteboard aspiration. It is now a shipping option, which changes what you are allowed to demand from a vendor.

Watch the word "first." Bloomerang framed the Dataro integration as the first of several intelligence-native features headed into its Giving Platform. Read that as a roadmap, not a footnote. The next eighteen months of association software will be a steady stream of AI wired directly into suites, each one useful, each one another quiet strand of the rope.

Worth a read

The NonProfit Times on the MomentiveIQ launch. The clearest neutral write-up of what the unified AI platform actually bundles, from the trade press rather than the press release.

The Register on AI vendor lock-in biting budgets. The cost side of the story from well outside the nonprofit bubble, which is exactly why it is useful.

Anthropic on donating the Model Context Protocol. The open-standard move at the source, if you want to understand why "the AI without the suite" is suddenly a real architectural choice.

The vendors consolidating the market are not wrong that AI works better on unified data. They are wrong about who has to own the unification. My guess is that within two years the associations that treated AI as a connector to a governed layer will be quietly outrunning the ones that let a suite become their brain. The question worth asking your account rep this quarter is a plain one. When we leave, does the intelligence come with us, or does it stay with you?

---

From the Mind of Ravi Rooprai is a weekly column on association tech, data, and AI. Read the perspectives for the longer arguments behind it.

Researched with AI assistance and fact-checked against primary sources. The analysis, judgment, and writing are mine. How this column is made →