Field notes
Most go-to-market rebuilds fail the same way: they build in week one.
Decide, build, correct - in that order. Thirty days to decide and build nothing, thirty to build to the model, thirty to run it and watch where the system fights people.
The order is decide, build, correct - not build first. Spend the first thirty days deciding and building nothing, the next thirty building to the model, and the last thirty running it and fixing where the system fights people.
On this page
Most go-to-market rebuilds fail in the same way. Someone buys the tools in week one, starts configuring, and three months later has a beautifully built system that answers a question nobody asked.
Build in the wrong order and effort does not compound. It accumulates, which is different, and worse.
Days 1-30: decide. Build nothing.
This is the hardest part because it looks like no progress. The ICP on one page. One primary motion, chosen and defended. The data model written down and agreed out loud by everyone who will touch it - objects, properties, stages, what each one means.
No tool gets configured yet. If you build before you decide, the tool becomes the decision, and you spend the next year working around a choice you never made.
Days 31-60: build to the model
Now, and only now, the CRM gets configured - to the model from phase one, not to the software's defaults. Objects, properties, stages, statuses, source capture. Routing live, so no lead waits in a queue nobody answers. Handover gates live. The dashboards each team needs, and nothing else.
Because you decided first, this phase is assembly, not invention. That is why it fits in thirty days.
Days 61-90: run it, and watch where it fights people
Work the motion for real. And watch, closely, for where the system fights the people using it - because that is where the design is wrong, not where the people are lazy. A rep avoiding a field is data. So is a stage everyone skips.
Fix the friction. Then automate the admin so nobody dreads the CRM. The first ninety days will not hand you a finished system. They give you one that is correct enough to improve and used enough to tell you how.
Decide, build, correct. In that order. Every team that runs it backwards ends up calling the result a data quality problem, when it was a sequencing problem all along.
This is one of ten decisions in the GTM blueprint - the go-to-market system we build for our clients, free and written down.