Why your tech stack does not talk to itself

Your CRM, custodian, and document system each do their job. The work is in the handoffs between them, which nobody sells and nobody owns.

Part of Running the operation

Most advisory firms have four to six systems of record, and every one of them was chosen carefully. The CRM holds relationships, the custodial platform holds accounts, the planning tool holds projections, the document system holds files, and the email archive holds everything anybody said about any of it.

Each is good at its job. None of them is responsible for what happens between them, and that is where your staff spend their week.

What "integrated" usually means in a vendor demo

Vendors do build integrations, and they are worth having. What they generally provide is a one-directional sync of a defined set of fields on a schedule, which solves a real problem and stops well short of the actual work.

A stock integration does not decide which system wins when two of them disagree, or what happens when a record exists in one and not the other. It does not decide what a document should be called, or who is responsible when a sync silently stops. Those are firm decisions, and no vendor can make them for you.

The four decisions that connect a stack

  • Which system owns which record, so there is one authoritative copy and everything else reads from it
  • What triggers what, and who owns each automation by name
  • What a document is called and where it lives, so it can be found by somebody who did not file it
  • What happens on an exception, because an automation that fails silently is worse than one that never existed

That last one matters more than it sounds. Automations do not usually break loudly. A vendor ships an update, a field is renamed, and the connection stops working for six weeks before anybody notices the pattern. This is why every automation we build has a named owner, a log of what it did, and defined behaviour when it hits something it cannot handle, which is set out on the what we do page.

The same client fact entered in three systems is not three records. It is one fact and two chances to be wrong.

The first thing to trace

Pick the fact that gets typed most often, which for most firms is a new household’s core details, and trace every place it lands. Count the manual entries. That count is also a decent proxy for how much capacity you are losing. That count is your integration backlog, in priority order, and it is usually shorter than people expect.

You almost certainly do not need new software to fix it. You need the connections between the software you already license, built against decisions somebody has actually made.

Running the operation

What is actually consuming the hours, and what to do about it. Start with Advisory firm operations: a guide for firms between $75M and $500M.

Using the fourth quarter to fix what slowed you down this year

Q4 is the only stretch where an advisory firm has both the visibility and the slack to change how it operates. Here is what fits in it.

Why another platform will not fix your operations

Most advisory firms have already bought good software. The hours are going somewhere else, and no vendor sells the thing that is actually missing.

What a CRM migration actually costs an advisory firm

The licence is the smallest number in a migration. Here is where the rest of the cost sits, and how to tell whether you need one at all.

How to write SOPs an advisory firm will actually use

Most advisory firm SOPs are written once, filed centrally, and never opened again. The fix is where they live, not how they are written.