Who buys custom development, and what pushes them?
Non-technical founders after a raise
They have a prototype, a seed round and investors who expect a real product. They need a team that can build version one, or rebuild the prototype properly. Watch funding signals and fresh listings in startup directories (new companies).
Funded teams whose roadmap outran hiring
The engineering roles have been open for two months and the launch date has not moved. Staff augmentation or a dedicated squad fills the gap. Hiring signals show it.
Companies stuck on legacy software
An internal tool built ten years ago, a framework nobody wants to touch, a cloud migration that keeps slipping. Posts about these show up as tech stack changes.
Rescue projects
A freelancer vanished, an agency missed three milestones, the app crashes on launch day. Founders say this in public, on Reddit and X. These buyers move fast because the pain is acute; treat the posts like competitor complaints.
Where do you find funded teams that are short on engineers?
Each source below answers one of the three questions a dev shop cares about: is there money, is there a deadline, and is the team too small for it. None answers all three, which is why the best leads tend to appear in two sources at once, such as a recent raise and an engineering role open for six weeks.
| Source | What it tells you | What it misses |
|---|---|---|
| SEC Form D filings | A US company raised money under Regulation D. The notice is due within 15 days of the first sale and lists executive officers and directors (SEC guidance). | Not every raise is filed this way, and the form shows amounts offered and sold, not what the money is for. |
| Accelerator directories, such as the Y Combinator company directory | Young companies by batch and industry, usually with small teams and fresh capital | Many have strong technical founders who will never outsource the core product |
| Job posts and careers pages | Which engineering roles are open, for how long, and in which stack | Whether the company would consider an agency at all: ask, do not assume |
| Product Hunt and launch posts | A shipped product with a date attached, and public feedback on bugs and missing features | Budget: a launch can come from a side project or a funded team |
| Public posts on Reddit and X | Rescue situations, vendor complaints, requests for a recommendation | Company details: many posters stay anonymous until you resolve the company |
Which engagement fits which signal?
Dev shops lose deals by pitching the engagement they like selling instead of the one the signal points to. A funded founder without a CTO does not want three contractors billed by the hour. A CTO with open roles does not want a fixed-scope MVP. Match the offer to the moment, and make the first purchase small.
| Signal | Engagement that fits | Pricing model | First paid step |
|---|---|---|---|
| Fresh round, no technical co-founder | Version one, built by a full team | Fixed price per milestone | Two-week discovery ending in a scope and a quote |
| Engineering roles open 30 days or more | Staff augmentation or a dedicated squad | Monthly fee per engineer | One engineer on a one-month trial |
| Rescue post or missed vendor milestones | Takeover: audit, stabilize, then build | Fixed-fee audit, then time and materials | One-week written code audit |
| Legacy system or stalled migration | Modernization in phases | Fixed fee per phase | Architecture assessment |
| Launch with traction and public bug reports | Performance, reliability and integrations | Monthly retainer | Review of the slowest user flows |
An example ICP and the email it produces
| Field | Example |
|---|---|
| Company | Seed to Series A startups with 3 to 30 people, based in the US |
| Trigger | Raised in the last 120 days, a founding engineer role open 30+ days, or a public rescue post |
| Decision-maker | CEO when there is no CTO; otherwise the CTO or VP Engineering |
| Stack fit | React, Node and React Native, matching the studio's strengths |
| Disqualify | Teams with more than 10 engineers in-house, hardware products |
| First project (assumption) | $40,000 to $120,000, starting with a paid two-week discovery |
The signal
The founder of a fictional logistics startup posts on X that the freelancer who built their MVP has moved on, with the payments integration half finished.
The first email
"You mentioned your MVP's developer has moved on with the payments flow half built. Before anyone writes new code, we start takeovers with a one-week written audit: what works, what is risky, what it would cost to finish. Happy to share the template we use."
Why it holds up
It names the event, describes a low-risk first step and offers something useful. It does not claim the code is bad, because you have not seen it. Our cold email templates include more openers for takeovers and staff augmentation.
Two buyers, two sequences: the CEO without a CTO and the CTO without enough engineers
The same studio often writes to two very different people. A non-technical CEO needs to trust that you will make good technical decisions on their behalf. A CTO needs proof that your engineers will not slow the team down. Here is how one fictional studio's sequence changes between them.
| Step | Non-technical CEO (seed round, no CTO) | CTO (Series A, two backend roles open 45 days) |
|---|---|---|
| Email 1, day 0 | "Congrats on the seed. The announcement mentions a mobile app by spring, and your team page lists no engineers yet. We build version one for founders in that spot, starting with a paid two-week discovery, so you see the plan and the price before any code." | "Your two backend roles have been open since August. If the roadmap did not move with them, we can put two Node engineers into your sprints within two weeks, working in your repository, your standups and your review process." |
| Email 2, day 4 | How decisions get made: a written note for each major technical choice, in plain English, with the trade-off stated. | A redacted pull request with its review comments, so they can judge code quality before any call. |
| Email 3, day 10 | An offer of 20 minutes to look at their prototype or spec together, with no deck. | A one-month trial with one engineer, and a clean exit if the fit is wrong. |
| Email 4, day 18 | "If you have found a technical co-founder, great. Keep the discovery template anyway; it works for in-house teams too." Then stop. | "If the hires landed, ignore this. If not, the offer stands next quarter." Then stop. |
Reach companies with a reason to buy this week
Startories finds the buying signal, verifies the decision-maker and runs the outreach until they book a call.
How do you rank this week's leads? A scoring example
The source table above says the best leads show two traces at once. A simple score makes that rule usable on a Monday morning: one point each for money, a deadline, a team too small for the work, and a stack you know well. The three companies below are fictional.
| Lead | Money | Deadline | Team gap | Stack fit | Score and next step |
|---|---|---|---|---|---|
| Fintech startup: seed round five weeks ago; the founder posts that the app must ship before a partner pilot in March | Yes | Yes | Yes: no engineers on the team page | Yes: React Native | 4: write today and offer the paid discovery |
| Series A marketplace: two backend roles open for 50 days, no public launch date | Yes | No | Yes | Yes: Node | 3: staff augmentation email this week |
| Small bootstrapped agency asking on Reddit for a cheap developer to fix its WordPress site | No | No | Yes | No | 1: skip; not your buyer |
Reading the scores
A 4 justifies a senior person's time on the first call. A 3 gets the sequence, but not a custom proposal until they reply. Anything below 3 is a later lead or someone else's client. Write the reasons next to each score, as Startories does for every lead it keeps, so a teammate can argue with the evidence rather than the number. Once a few months of results exist, our lead scoring guide shows how to reweight the four criteria toward the signals that actually turned into signed projects.
Earning trust when every buyer fears the wrong vendor
Most founders who hire an agency know someone with a horror story. Your proof has to answer that fear, not just show nice screens, and your answers to their objections have to sound like someone who has been through a bad engagement from the other side.
- Live products they can click or download, with your exact role stated (design, build, maintenance).
- A sample of how you work: a real architecture note, a sprint report, a pull request with client details removed.
- References they can call. One conversation with a past client outweighs any case study.
- When clients signed NDAs, an anonymized case study the client approved: the problem, the stack, the timeline, what shipped.
- Clear terms on ownership: the code lives in the client's repository from day one and the IP is assigned to them.
- A paid discovery sprint as the first purchase, so the buyer risks a small amount before the big one.
| Objection | Answer |
|---|---|
| "We want to hire in-house." | Fine for the long term. Offer to ship version one while they recruit, then help interview and onboard the hires, with a written handover. |
| "Freelancers are cheaper." | Per hour, yes. Compare the cost of a missed launch or a rewrite. A fixed-fee discovery shows what a team adds: planning, QA, continuity when someone is out. |
| "We had a bad offshore experience." | Ask what went wrong: time zones, turnover, unclear scope. Answer each point with working-hour overlap, named team members and weekly demos. |
| "What is your hourly rate?" | Give a range and the model (per engineer per month, or fixed per milestone), then steer to scope: the total depends on what gets built, and a discovery produces a firm number. |
| "Can you sign our NDA first?" | Usually yes, and quickly. Read it, sign a standard mutual NDA and move on; hesitating over one makes you look like the vendor they fear. |
| "We are not ready yet." | Ask what has to happen first (the round, the hire, the spec) and follow up when it does. |
Deal math: why one project can pay for a year of outbound
A worked example, with assumptions you should replace. Your studio staffs a first project with two engineers and a half-time project manager for four months. At an assumed 150 billable hours per person per month and a blended rate of $90, that is 2.5 people × 150 hours × 4 months = 1,500 hours, or $135,000. At an assumed 35% gross margin, the project earns $47,250 before sales costs.
Against that, a year of Startories Growth costs $5,988, and done-for-you setup adds $1,500 to $2,500 once. Even fully managed acquisition at $1,999 a month comes to $23,988 a year, roughly half of that single project's margin.
The harder constraint is time. Assume six to ten weeks between a first call and a signed contract, or use your own history. Either way, the leads you contact in January fill your bench in spring, so outbound has to run while your team is still busy.
What to measure while you wait
Development projects take weeks or months to close, so judge outbound on the early steps first and on revenue later.
- First month: replies and calls booked, by signal
- Second and third months: paid discovery sprints sold
- After that: projects signed, first-project value, and how many turn into ongoing work
- All along: time from the signal to your first email. Rescue posts and recommendation requests attract replies fast, so speed matters.
Mistakes that waste a dev shop's outbound
- Leading with headcount. "200 engineers across three time zones" tells a seed-stage founder you are built for someone else. Lead with the one thing you would do for them.
- A stack list in the first email. Every shop lists the same frameworks. Name the stack only when it matches something in their job post or public code.
- Pitching "cheaper than hiring" to a CTO who is hiring. It sounds like a challenge to their plan. Pitch speed and a clean handover instead.
- Sending senior engineers to unqualified calls. Score and disqualify first, and protect the people who bill.
- Treating every funded company as a lead. A team of twelve engineers that just raised is hiring, not outsourcing. Check the team before you get excited about the round.
- Going silent after "not now". Ask what has to happen first, write it down, and write again when it does.
Where Startories fits, and how to start
Startories watches Reddit, X, Product Hunt and startup directories for funding news, launches, hiring, stack changes and rescue posts. It matches each one to a company, checks it against your ICP and explains the score, finds the founder or CTO, verifies the business email and drafts the first email around the event. That is signal-based outbound for a dev shop: fewer, better-timed conversations.
Your outreach must still follow US email law. The FTC's CAN-SPAM compliance guide asks for an honest subject line, a valid physical postal address and an opt-out you honor. Startories adds the opt-out to every email and suppresses unsubscribed contacts across campaigns.
Growth ($499 a month) suits most studios: three funnels such as funding, rescue posts and launches. If you would rather not run it yourself, the done-for-you service sets everything up for a one-time $1,500 to $2,500, or runs it entirely from $1,999 a month. See all plans. If you build AI products, the AI agency playbook covers that market, and lead generation for B2B startups shows your buyers' side of outbound.
Frequently asked questions
How do software development companies get clients?
Mostly through referrals, directories and outbound. Outbound works best when it is timed: funded startups without a CTO, teams with engineering roles open for weeks, companies migrating old systems, and founders posting about a failed vendor. Reach them about that situation with a small, paid first step.
Is outbound worth it for a small dev shop?
Usually, because one project can be worth tens of thousands of dollars, so a few new clients a year cover the cost. The key is narrow targeting, so your senior engineers do not spend their time on calls with companies that were never a fit.
Where can a dev shop find recently funded startups?
Follow funding announcements on X and in startup directories, check SEC Form D filings for US raises under Regulation D, and browse accelerator directories. Then confirm the team is small for its plans, through open engineering roles or a founder with no CTO.
Which buying signal works best for a development agency?
It depends on your offer. Rescue posts and recommendation requests show urgent need. Funding rounds show budget. Long-open engineering roles suit staff augmentation. Track replies and signed projects per signal for a few months, then put more effort behind the ones that convert.
How do I show my work if clients signed NDAs?
Ask past clients to approve an anonymized case study that describes the problem, the stack, the timeline and what shipped. Offer reference calls, and share samples of your process, such as a redacted sprint report or architecture note.
Can Startories find non-technical founders?
It finds the person who owns the problem at each company. In an early startup without a CTO that is usually the CEO or founder; in larger teams, the CTO or VP Engineering. Every business email is verified before anything is sent.