Why the customer 360 keeps failing

Almost everyone in revenue operations has been asked for a customer 360. Most of them were asked by someone who thinks it’s a report. It gets scoped, it gets a project code, and a year later the renewal desk is still working from a spreadsheet somebody rebuilt on Monday. The data was there the whole time. The answer just never reached the person who needed it, at the moment they needed it.

This post is about why that keeps happening. The short version: a 360 has to serve several people who want different things, from data that disagrees with itself, and putting it together is no team’s job.

Growth now comes from the base

The business case starts with where growth comes from. Benchmarkit’s 2026 benchmarks put median ARR growth at 20% in calendar 2025, down from 30% in 2022, and Benchmarkit describes it as the fourth consecutive year of decline.¹ SaaS Capital’s 2026 survey, drawn from a different and larger population of private B2B SaaS companies, lands at a different level and points the same way: median growth of 22% in 2025, down from 25% the year before.²

So the growth that’s left has to come from customers you already have. In the same Benchmarkit data, expansion accounts for 40% of net new ARR at the median and 44% among low-growth companies. Benchmarkit’s own reading of that threshold is the useful part: once expansion crosses 40%, it signals that expansion is replacing new-logo growth.¹

There’s a fair objection here, and I’ll concede it straight away. The same report shows acquisition getting more efficient, and headlines that section as operating efficiency having recovered. That’s true. My claim is narrower: acquisition now delivers less growth, whatever it costs. If your growth depends on the base, you need to know the base cold, account by account, and most companies don’t.

Every team needs a different view

Here’s where it gets harder. “Know the account” is four different questions, asked by four different people, in four different places.

• The renewal desk works a queue. It wants every account renewing in the next 90 days, sorted by risk.
• The CSM opens an account page thirty seconds before a call and wants to know what changed.
• The CRO reads one page on a phone on Thursday morning before the executive review.
• The rep never leaves the CRM, so the answer has to arrive as a prompt in the flow of work.

They’re asking about the same account. What differs is the thing each of them opens: four surfaces, four moments, and four very different amounts of patience.

The data underneath is wrong

Then there is the material the answer is built from. In Openprise’s 2025 survey of revenue and marketing operations practitioners, 99% reported struggling with at least one dimension of data quality: data that is incorrect, data that isn’t unified across sources, or data that lacks enough detail.³ That figure comes from a vendor with an interest in the answer and a modest sample, and the footnote says so. It also matches what anyone who has run a renewal forecast already knows.

Each question needs six to eight systems

Take those four questions and trace where the answer lives. What follows is domain description, with no research behind it. It’s how these questions break down in the accounts we work with, and your stack may differ.

Is this account healthy, 90 days before renewal? That touches the CRM for stage and owner, billing for what’s invoiced and owed, support for open tickets, product analytics for whether the seats are still in use, the customer success platform for the health score and QBR notes, and call recordings for what the customer actually said. Six systems, at the least. Where is there room to grow, and who went quiet and when, need a similar six. The brief your CRO asks for on Thursday needs all of them plus the warehouse and marketing engagement: eight.

Each of those systems is right about its own slice and wrong about the customer. The CRM says the account is healthy because the stage says so. Support knows it isn’t. Product analytics knows usage dropped in June. The health score predates all of it. Most readers will be adding a ninth system in their heads by now, which makes the point better than any count.

Data times perspectives

Put the last two sections together and you have the whole diagnosis. The problem is data multiplied by perspectives. Every one of those systems has to be readable by every one of those people, in the form each of them will actually open.

Most teams have solved the data half at least once and still never had a 360. The usual response to four audiences is one screen designed to satisfy all of them, scoped by committee. One shared dashboard serves none of them. The renewal desk can’t work it as a queue, the CSM can’t find what changed, the CRO won’t scroll it on a phone, and the rep never sees it at all. Nearly every 360 that died, died that way.

If your team only needs one of the four today, that’s good news. Start with that one. Just make sure the fourth won’t require rebuilding the first.

The join has no owner

Every system in that list has an owner and an admin. The CRM has a team. Billing has finance. Support has its own ops lead. The join between them has neither. So the join gets done by a person, by hand, usually on a Monday morning, and it gets done again next Monday because nothing they built can be reused by anyone else.

The join is where the time goes

The time cost shows up in two places. In Rattle’s 2025 survey, 46% of revenue ops teams said they run mostly or entirely on manual processes.⁴ Two caveats: the figure counts reliance on manual work, with no hours attached, and it comes from a vendor. Still, it’s RevOps people describing their own function: the join, done by hand, every week.

The second figure is about who has room to fix it. Gartner reported in April 2023 that only 27% of sales operations support time goes to the sales function; the rest goes to finance, supply chain, HR and analytics.⁵ Read it as a snapshot of one moment, and note its age: the survey was fielded in late 2022, among sales leaders describing their ops teams. Even so, the team best placed to build the account view is the team with the least time left to build anything.

What teams do today

Most teams have tried at least one of three things, and two of them work for a while.

The data project. A warehouse, a model, and a view on top. This is the right answer to a different question. Warehouse work pays off for analytics, and it should keep going. It doesn’t give a CSM an account page, and in our experience it takes something like a year and a half, by which point the definition of a healthy account has moved.

The spreadsheet. The most common answer and the least admitted. Somebody rebuilds it every Monday, often from a script that only runs while their machine is on. In the engagements we see, a weekly spreadsheet ritual or a laptop-bound script is the thing teams replace most often. One team’s sales dashboards depended on a manager’s laptop staying awake. That’s a single point of failure with a person’s name on it.

The point tool. It works, and it’s fine at what it does. It also has its own login, its own copy of your data, and it reaches one system.

None of the three produces something the rest of the team can open.

The shape of the fix

The shape that works has two parts. Underneath, the heavy work done once: connecting the systems, moving and joining the data, and keeping it current. On top, a separate light app for each audience, built for the way that person works: a queue for the renewal desk, a page for the CSM, a brief for the CRO, a prompt for the rep. Tray iPaaS does the heavy integration lifting, and Tray Helix is where the apps run. A team builds each one in the AI assistant it already uses, and Helix deploys it with SSO, managed credentials, a named owner and a log of every run, so the person who wasn’t in the room can open it tomorrow.