# The FDE 30-60-90 day plan: a six-week target leaves about four weeks with the customer

> The first two weeks usually go on internal work. A new FDE has to pick one small, high-impact task and agree it with the customer before hitting the wall in week five or six.

Original: https://fdetimes.net/en/guides/fde-30-60-90-day-plan/

Imagine you have just been handed a customer account, and your manager wants the customer to "see value in the first six weeks". That sounds generous until you open your calendar and count.

In practice, an FDE's first two weeks usually go on internal work: learning the company's deployment patterns, requesting access to systems, shadowing a senior FDE. Take those two weeks away and six weeks becomes about four weeks of real work with the customer.

That is why a 30-60-90 day plan is not paperwork. It is a tool for making sure you do not spend those four precious weeks on work that is correct but that the customer never feels.

## Start from the customer's clock, not yours

Entrans defines an FDE's Time-to-Value as the time from the start of an engagement to the customer's first real operational outcome. The key words are "real operational": a demo running on your laptop does not count.

A useful way to think about it, as Rocketlane suggests, is to ask what value can be unlocked at 30, 60 and 90 days from kickoff. Put the two ideas together and your plan should have two layers: your own learning, and the value the customer can see at each milestone.

## The first two weeks: lay the groundwork, don't wait around

An onboarding playbook from Tandem, written for a second FDE joining a live account, puts the account brief ahead of even the first-day office tour. If nobody hands you one, write it yourself: who the customer is, where it hurts, who makes decisions, and what the company has promised.

In week one you sit in on every customer call with the senior, and you both use the brief as a shared starting point. You do not need to say much. Your job is to note every task the customer mentions, because that list is the raw material for the next step.

By the end of week two you should have access, an account brief and a list of candidate tasks. If any of these is missing, that is the first risk to flag.

## Example: choosing a "home run" for an invoice-reconciliation customer

Imagine you take on a distributor that wants an agent to reconcile invoices automatically. After two weeks of listening to calls you have five candidate tasks. You score each on effort and on impact for the customer, on a scale of 1 to 3.

| Candidate task | Effort | Impact | Decision |
|---|---|---|---|
| Read PDF invoices from email and push them into the accountants' reconciliation sheet | 1 | 3 | Home run, do first |
| Two-way ERP integration | 3 | 3 | 60–90 day milestone |
| Overview dashboard for the director | 1 | 1 | Later |
| Fine-tune a custom model for invoice templates | 3 | 1 | Not now |
| Alerts on mispriced invoices via internal chat | 2 | 2 | Candidate for weeks 5–6 |

Rocketlane calls a low-effort, high-impact task a "home run", and recommends it as the first choice for the initial delivery of value. In the table above, only the PDF invoice task meets that bar.

The ERP integration is far more appealing, and every engineer wants the hard problem. But with four weeks, it will almost certainly not be finished when the customer starts asking "have we got anything yet?" You schedule it for the 60 to 90 day milestone, once you have earned the trust to ask for more time.

The next step is neither to decide alone nor to let the customer decide alone: you sit down with the customer and plan the route to first value together. A good opening line with the head of accounting: "If in three weeks your team no longer has to type up invoices from email by hand, would that be a meaningful result?"

If she nods, you have a measurable goal and someone on the customer side who shares responsibility for it. If she shakes her head and points to something else, you have just saved yourself three weeks of work in the wrong direction.

**Key point:** A good plan does not start with the hardest task, but with a small one the customer feels immediately.

## Weeks five and six: prepare for the wall

FDE Academy, writing about an FDE's first 90 days, describes how most new FDEs hit a "wall" of their own around week five or six. Those who get past it well receive their first signal of trust from the customer.

What that wall looks like differs from account to account. But on the timeline in the example above, it would most likely land just as the home run starts running on real data: the data is dirtier than you expected, access is missing a step, users work differently from the process they described.

So keep week six for fixing things and putting them into production. Do not schedule new features there.

FDE Academy also notes that a new FDE's first assignment is rarely the team's highest-risk project. If your task is smaller than you hoped, that is normal. Do it cleanly.

## Three checkpoints: 30, 60, 90

Day 30 is a readiness test: you present the whole account back to the senior. Rehearse until you can explain where the customer hurts, who decides and where the home run stands, without opening any documents.

By day 60, you become the customer's first responder; the senior steps in only for exceptions. This is also the milestone Tandem's playbook sets. Now larger work such as the ERP integration starts to make sense, because the customer is used to coming straight to you.

At day 90, set expectations correctly. According to Tandem, a full engineering ramp takes 3 to 9 months on average, so day 90 is about early ownership of the account, not full productivity.

## Mistakes that sink the plan early

The most common mistake is picking the biggest task first because it looks impressive on a slide. By week six you have plenty of progress but the customer has no operational result.

The next is letting the customer choose the outcome without discussing it, and getting back a request that cannot be done in four weeks. Another is measuring only one number.

Entrans recommends a balanced scorecard rather than a single metric, to avoid creating the wrong incentives, such as rushing Time-to-Value with a solution nobody uses after month three.

The last is treating the two internal weeks as dead time. They are the cheapest time to learn, ask questions and take notes. Waste them and you will pay for it when the wall arrives.

## How to show this skill on a CV

When reading FDE job descriptions, look for phrases such as "time-to-value", "first value" or "customer outcomes". Palantir's Forward Deployed Software Engineer posting, for example, says the role works directly with customers to rapidly understand their biggest problems.

When you see descriptions like that, have a story ready about how you sequenced work in the first weeks of a project. With four weeks, that word "rapidly" is exactly the pressure described above.

On your CV, do not just write "implemented integration X". Write it as a sequence: which task you did first, why, what operational result the customer achieved, and how long it took.

## Exercise: rescore your last project

Take a project you worked on in the past year, including at a product company. Write out its first six weeks week by week, then mark the week when users first got a real operational result.

Then rebuild a scoring table like the invoice example for the tasks the project had. If the task done first was not a home run, rewrite the order and estimate how many weeks earlier the first result could have come. That rewrite is also the story you take into the interview.

Next time you take on a new project, count back from the milestone the customer is waiting for. Four weeks sounds short, but it is enough for a home run done cleanly and agreed with the customer in advance.

**Try this week:**

- List five things your customer or current project is expecting, score each on effort and impact from 1 to 3, and circle one 'home run'.
- Write a one-page account brief for the project you are working on: who the customer is, where it hurts, who makes decisions, what has been promised.
- Try presenting the project's status to a colleague in five minutes without opening any documents, as in the day-30 test.

## Sources

- [Forward Deployed Engineer First 90 Days: What Actually Happens on the Job](https://fde.academy/blog/forward-deployed-engineer-first-90-days)

- [Onboarding a second FDE onto a live account: the ramp playbook, and the context it depends on](https://usetandem.ai/blog/onboarding-second-fde)

- [How to showcase value to your customers right from the very beginning](https://www.rocketlane.com/blogs/how-to-showcase-value-to-your-customers-right-from-the-very-beginning)

- [Creating early wins with the first value delivery framework](https://www.rocketlane.com/blogs/creating-early-wins-with-the-first-value-delivery-framework-irit-eizips-for-propel22)

- [Forward Deployed Engineer Metrics: How to Measure FDE Performance and ROI](https://www.entrans.ai/blog/forward-deployed-engineer-metrics)

- [Palantir Technologies - Forward Deployed Software Engineer](https://jobs.lever.co/palantir/bf718bd3-b2ef-451e-8033-cb4d2d9c094b)