Self-serve is the wrong model for revenue systems

Point tools can be configured by the buyer. A revenue system has to encode how you sell, so it should be set up around you by an engineer, not assembled from a dashboard.

Abhishek Singla, 6 November 2026

Self-serve is the wrong model for a revenue system. It works for a point tool. You sign up, connect a few accounts and configure it yourself. A revenue system is different. It has to encode how you sell: your market, your motion, who owns an account, when it is ready and what the next person needs at handoff. Those rules are never the same twice.

So teams stitch a router, an enrichment tool and a signals feed together, or they hire a GTM engineer to build the system in house. Each path takes months and depends on the person who built the glue.

The work is the same everywhere. Only the market, the motion and the team are different. The right model is a proven core, set up around how you sell by a forward deployed go to market engineer, in your own instance, with your data staying yours. In 20 days your team is running it. Those engineers stay on to fix and extend it.

Why does self-serve work for point tools?

Self-serve works when the product is a point tool. A CRM you can set up yourself. A sequencer you can load yourself. A dashboard you can connect yourself. The buyer already knows the job, and the tool has one job.

Most software companies sell the same way: put the product in the buyer's hands and let them configure it. That motion fits a single job. It does not fit a system.

Revenue software copied that motion. You buy a router, an enrichment tool, a signals feed, an ads platform. Each one is self-serve. Each one does what it says. The work of turning them into a system is left to you.

I am a GTM engineer. I build the systems revenue teams run on every day: signals, scoring, routing, qualification and the handoff to sales. I ran that system as the sole GTM engineer at a B2B SaaS company that grew from 5M to 20M ARR in nine months. In Berlin's GTM community, every GTM engineer I compared notes with had built their own version inside their company: a CRM, an enrichment tool, a router and a signals feed, joined together by hand. Each one took months, and each one depended on the person who built it.

That pattern is the tell. If every practitioner in a field independently rebuilds the same system, the market is selling the parts and asking the buyer to invent the product.

What does a revenue system have to know?

A revenue system is not another tool in the stack. It is the path from first signal to signed deal, and then into customer success. It has to know how you sell.

It has to know your market: your ICP and the TAM built from it, kept current, tiered by fit.

It has to create demand across that market, not wait for inbound. Campaigns run on your own ad accounts and budget. Every response becomes a signal on the account.

It has to read buying signals together from campaigns, website, product, CRM and the tools you already use, so a ready account does not sit in three dashboards.

It has to move that account to the right owner, with your territories, your handoffs and a queue, in minutes, with the reason attached.

It has to hold every deal to the qualification standard your team chose, with the buying committee on it.

And it has to do all of this without replacing the CRM your team already trusts. HubSpot is supported out of the box. Salesforce is set up on request. The system reads from the CRM and writes stages, owners and notes back.

None of that is a setting you toggle on a pricing page. It is a set of rules about how your company sells. Tools fit the average team. You are not the average team.

How do teams try to build a revenue system themselves?

They stitch the stack themselves

You keep the CRM. You add a router so inbound does not pile up. You add enrichment so reps see a fuller account. You add a signals tool so someone can watch buying intent. You add an ads platform, or an agency, so demand is not only inbound.

Each purchase is reasonable. The problem starts in the gaps. Nobody owns what happens between the tools: which signal means the account is ready, who should get it, what the next person needs to know, and whether the deal still matches how you qualify. RevOps becomes the integration layer. Data splits across vendors. When something breaks, the first question is whose tool it is.

You can make this work, for a while. The cost is not the licenses. It is the months spent connecting tools that were never designed to be one system, and the work of keeping the glue from rotting.

They hire a GTM engineer to build it

This is the more honest path, and the one I know from the inside. You hire someone whose job is the system: the TAM, the campaigns, the signals, the routing, the qualification, the handoff. They build it around how you actually sell. It fits.

It also takes months, and it lives in that person's head. The rules are implicit. The exceptions are tribal knowledge. The workflows work until the person who wrote them leaves, or gets pulled onto something else.

An in-house build is yours, which is the point. It is also a single point of failure dressed up as a competitive advantage.

The system should not leave when the person does

The mistake is treating delivery as a configuration problem.

Self-serve assumes the buyer can assemble the product. That is true for a point tool. It is not true for a system whose value is the way the pieces fit your market, your motion and your team. If the buyer has to invent that fit, you have not sold a system. You have sold parts and a dashboard.

The opposite mistake is treating the system as a custom build that only one person can touch. Then the company does not own a system. It rents one from whoever happens to understand it.

There is a third path. The work is the same everywhere. Only the market, the motion and the team are different. Build the common core once. Set it up around how each company sells. Write the rules so the team can read and change them. Keep the customer's data in the customer's own instance. Stay on after go-live so the system does not depend on one person remaining in the seat.

That is a forward deployed model. It is slower to start than a sign-up page and much faster than building from scratch. It is also the only model that matches what the work actually is.

What are the three ways to get a revenue system?

Three paths show up again and again: stitch point tools yourself, hire someone to build in house, or start from a proven core set up around how you sell.

GTM Pilot

What it is
A proven core, set up around how you sell by a forward deployed go to market engineer
Time to live
20 days
Who owns the rules
Your team: your territories, handoffs and qualification, written so you can read and change them
Where your data lives
Yours, in your own instance
What happens when the builder leaves
The core stays. Your team keeps running it. We keep fixing and extending it

A stitched stack

What it is
A router, an enrichment tool and a signals tool, connected by your team
Who builds it
Your team, after buying the tools
Time to live
Weeks per tool, then the work of connecting them
Who owns the rules
Scattered across tools, within what each one allows
Where your data lives
Split across vendors
What happens when the builder leaves
The glue is yours to keep working

An in-house build

What it is
A GTM engineer you hire, building from scratch
Who builds it
One person you hire
Time to live
Months, after the hire
Who owns the rules
The person who wrote them
Where your data lives
Yours
What happens when the builder leaves
The system leaves with them

GTM Pilot is that third way. It is a revenue system for Series A and B B2B SaaS companies with product-market fit. It creates demand across your whole market and moves every ready account to the right rep.

It is not an agency. It is not a CRM, a sequencer, a dialer or a lead-list vendor. It is not a self-serve product you configure alone. Campaigns run on your own ad accounts. Outreach and enrichment stay the tools you already use. We hand them audiences and ready accounts; we do not replace them.

What does setup by a forward deployed go to market engineer look like?

A forward deployed go to market engineer sets the system up around how you sell, in your own instance, instead of leaving you to assemble it from a dashboard. You start with a core that is already running in production: workflows, integrations and data models that are built and tested, and security built to SOC 2 controls from day one. You are not buying a blank workspace.

A Ziel Lab forward deployed go to market engineer then builds the rest around you, in your own instance, on your own domain, with a separate database hosted in the EU. You own the data and can export it at any time.

The standard deployment is 20 days, scoped on a teardown call:

  1. Days 1 to 5. Connect CRM, product and website data. Agree the ICP. Build the TAM.

  2. Days 6 to 12. Audiences and campaigns go live on your ad accounts. Routing, handoff and qualification are set up with your team.

  3. Days 13 to 20. Reps work from GTM Pilot. Every role is trained. We hand the system over.

On day 20 your sales, marketing and RevOps team is running it. No GTM engineer is required on your side. The rules are written so you can read them and change them.

When something breaks, or the business changes (a new territory, a new product, a new signal) Ziel Lab engineers fix it or build what comes next, under the subscription. Pricing is a fixed deployment fee for the 20-day setup, then that subscription. It is never hourly. The numbers are shared on the teardown call, not on a public page.

This is the bet

The market will keep shipping point tools a team can turn on in an afternoon. That is useful. It is not the same as owning a revenue system.

A revenue system has to be designed for how you sell, deployed into your own instance, and kept after the person who set it up moves on. Self-serve cannot do that work. An in-house build can, until the builder leaves.

We built GTM Pilot so you do not have to make that trade. The core is proven. The setup is yours. Your team runs it. We stay.

If you want to see where demand leaks in your funnel and what this would look like for your team, book a teardown: 30 minutes with a forward deployed go to market engineer.

Leave comment

See where your pipeline leaks

Thirty minutes with a forward deployed go to market engineer. We map where your demand comes from, where accounts get stuck on the way to a rep, and what we would deploy in the first 20 days.