Tibo Louis-Lucas on Validating and Growing SaaS Products

Fast Evidence, Not Fast Shipping

The shallow version of Tibo Louis-Lucas's playbook is to ship more products. The useful version is to shorten the time between an assumption and evidence strong enough to change course. His earlier startups took roughly two years each to fail, including one where the team raised money and hired before finding product-market fit. It attracted users but few paying customers 5:37, 6:02. The lesson he carried forward was not that every idea deserves frantic execution. It was that an unproven idea should not consume two years before the market gets a vote.

He and a co-founder responded by trying to release roughly one new product each week, then giving an idea only a week or two to show a reason to continue. One handled marketing and design while the other focused on building, and they tested around ten software products 7:43, 7:59, 8:30. Revenue was the decisive signal. Several experiments made money, while the others were stopped rather than protected by a persuasive story 9:05.

That cadence was a response to a particular failure mode, not a universal command to launch every seven days. The transferable principle is to define the cheapest credible test before building the full system. Speed matters because it creates more opportunities to learn; it is not evidence by itself. Louis-Lucas describes product-market fit as the point where the problem changes from finding demand to struggling to fulfil it 9:39. Until that happens, the job is to keep assumptions inexpensive.

Payment Is Stronger Evidence Than Attention

Louis-Lucas repeatedly separates usage from willingness to pay. In a later interview, his first validation rule is to charge from the beginning because a payment reveals a different level of need from a sign-up or compliment 4:48, 13:08. Tweet Hunter crossed its first meaningful recurring-revenue threshold quickly, which told the founders that customers did not merely enjoy the concept; they would exchange money for it 10:16.

This changes what an early launch is designed to answer. A free product can test whether someone understands the proposition, reaches the result, or returns. It cannot by itself establish that the problem belongs in a budget. Charging early tests the whole commercial promise: who the buyer is, how urgent the job feels, whether the result is valuable enough, and whether trust survives the checkout screen.

The metric must match the business. Louis-Lucas and his partner explicitly chose revenue as the filter for their software experiments 9:05. That does not make visits, activation, retention, or referrals irrelevant. It stops them from becoming substitutes for the result a paid product ultimately needs. A useful sequence is therefore: observe interest, ask for payment, watch whether the buyer reaches value, and then see whether they stay or recommend it. Each step invalidates a different kind of optimism.

Follow Behaviour, Not the Original Story

Fast shipping only works when the founder is willing to be corrected. Louis-Lucas's later summary is to follow what users actually do, even when their behaviour points away from the original plan 9:55. One of his abandoned products searched online communities. It solved a real problem, but users needed it infrequently, so it lacked the repeat use required for a durable subscription 14:32.

That example exposes a common validation error. A user can genuinely value an answer while rarely needing another one. The product then has utility but weak recurring economics. The right response is not to argue that the feature should become a habit. It is to decide whether the business should charge per use, serve a higher-frequency job, bundle into a broader workflow, or stop.

Tweet Hunter began with a much smaller surface than the later product: essentially a wall of examples intended to help people find ideas for posts 15:35. Louis-Lucas's stated reason for starting narrowly was to validate and capture attention around one specific need before adding the rest 15:59. The initial product was not a miniature version of an imagined final suite. It was a probe into where users already felt friction.

A Narrow Audience Produces Better Product Information

Louis-Lucas treats an audience less as a broadcast count and more as a network of two-way relationships. He credits that network with validation, feedback, early sales, partnerships, and access to people who could move the product forward 2:22, 4:29. The important qualifier is relevance. A large audience with the wrong interests supplies noisy feedback, while a much smaller group of likely buyers can reveal what to build and why they would pay 7:06.

This is why his experiments were related rather than random. They were aimed at creators and distributed through the same mailing list, so each launch taught the team more about a shared buyer and made the next launch more credible 13:49. A portfolio compounds when products reuse customer knowledge, trust, and distribution. Ten unrelated products may create ten isolated starts; ten tests around one audience can create a learning system.

Early publishing serves the same purpose. Louis-Lucas recommends producing enough work to discover a distinctive voice and learn what resonates, rather than waiting to perfect a content strategy in private 11:07. But he does not present social media as the only path to a business 13:30. The underlying asset is not a follower count. It is a reliable way to reach the right people, hear their response, and earn another interaction.

Distribution Is Part of the Product

Tweet Hunter's early growth did not come from one channel. It came from several distribution mechanisms that fit the same audience. The founders acquired a small adjacent tool whose existing reach became an important growth driver 22:42. Louis-Lucas suggests that dormant tools can sometimes be found by looking through launch platforms and contacting their makers, turning acquisition into a way to buy useful product surface, attention, or both 23:39.

They also formed a distribution partnership with a creator whose teaching already matched the product. The arrangement connected ongoing promotion to a share of product economics, rather than treating the creator as a one-off sponsored placement 24:42, 26:55. The general lesson is not to copy that deal structure. It is to seek a partner for whom recommending the product is consistent with the value they already deliver, then align incentives with durable outcomes.

Free mini-tools supplied a third channel. The team initially made playful, shareable tools, then shifted toward tools built for specific search queries. Louis-Lucas says the search-oriented versions became a major source of growth 28:55, 29:34. He later returned to the same pattern: build a genuinely useful free tool around a narrow search intent, then connect the result to the paid product's next step 20:50.

All three mechanisms share a property: they deliver value before asking for attention. The acquired tool already served users. The creator partnership matched existing instruction. The search tool answered an active question. Distribution worked because it was attached to a job the audience already wanted done.

Depth Beats a Shallow All-in-One Product

The founders positioned Tweet Hunter against broad social-media suites. Instead of supporting every network lightly, it concentrated on one kind of creator and one platform 13:11. Louis-Lucas later made the same product argument explicitly: different networks reward different content and behaviour, so a dedicated tool can fit the workflow more deeply than a generic interface spread across all of them 21:44, 25:33.

Narrowness is commercially useful when it lets a product solve the complete job. It clarifies the landing page, constrains the roadmap, improves onboarding, and makes word of mouth more precise. Louis-Lucas noticed that revenue continued after a promotional push ended, which he interpreted as evidence that recommendations were carrying the product 15:11. A customer can repeat a specific promise more easily than a catalogue of loosely related features.

There is a corresponding risk. A product deeply tied to one platform inherits that platform's policy, API, and market changes. Louis-Lucas identifies changes at Twitter and LinkedIn as part of the risk that influenced the decision to consider a sale 31:16. The practical balance is to go narrow enough to win a job, while understanding which dependency could erase the advantage.

Founder Support Turns Reliability Into Trust

Early support was deliberately direct. A help link sent users into Louis-Lucas's messages, where he could see bugs and answer in real time 17:16. When a defect appeared, the team could fix it within minutes and tell the affected user what had changed 18:20. He argues that a technically small fix can create a disproportionately strong emotional response when the customer sees that the product team is listening 18:41.

That loop is more than good service. It converts support into product discovery, reveals the language customers use, and shows which failures destroy trust. A scheduling product, for example, must work reliably at the promised moment; a missed post undermines the core reason it exists 17:53. Reliability is not polish added after validation. For workflows with a deadline, it is part of the value proposition being validated.

Louis-Lucas also connects this learning speed to team shape. He views hiring a larger team before understanding the core product as one of his earlier mistakes, because founders need to experience the central customer problems firsthand 19:11. In a later interview, he describes operating several businesses without building a conventional large team around each one 26:58. Lean is valuable here because it keeps information close to decisions, not because headcount is inherently bad.

Relationships Compound More Than Reach

Louis-Lucas says a small number of relationships account for much of his progress, including meeting a business partner through Twitter 17:56. His advice for direct outreach is correspondingly relational: provide something useful or ask a thoughtful question before making a request 19:54.

This explains why audience quality, founder support, partnerships, and product distribution reinforce one another. Each is a way to create repeated, useful contact with the same market. The founder who listens in support learns what to publish. The useful post attracts the right buyer. The buyer's behaviour shapes the next feature. A trusted operator introduces the product to another relevant audience. None of those loops requires mass reach at the start.

The Portfolio Rule: Kill, Acquire, or Double Down

The deeper discipline in Louis-Lucas's portfolio is allocation. A test that produces no credible signal is stopped. A small product with useful distribution can be acquired. A product showing paid demand, repeat use, and word of mouth receives more attention. His retrospective account emphasises that many failed experiments preceded the hit, and that the ability to keep running bounded tests mattered more than knowing in advance which one would win 40:16, 34:10.

This is not permission to abandon a good product whenever growth becomes difficult. Louis-Lucas's own standard for a promising product includes demand strong enough to create operational pressure 9:39. The decision should follow evidence defined before the test: payment, activation, repeat use, retention, referrals, or another behaviour tied to the business model. Persistence is rational when those signals improve. It becomes sunk-cost protection when the founder keeps changing the definition of success after the result arrives.

What This Playbook Does Not Prove

These interviews are founder retrospectives, not controlled comparisons between business strategies. They are strongest as evidence of how Louis-Lucas says he made decisions, and weaker as proof that the same tactic will produce the same outcome elsewhere. Revenue and growth outcomes discussed in the episodes are self-reported, which is why this synthesis uses them sparingly and does not treat headline numbers as audited results.

There is also selection bias in any portfolio story told after a breakout product. Many founders can charge early, publish frequently, and still fail to find a durable market. The defensible lesson is narrower: those actions can make weak assumptions visible sooner and preserve resources for a better bet. They improve the quality and speed of a decision; they do not guarantee which decision the market will reward.

A Practical Tibo-Inspired Decision Loop

The interviews combine into a repeatable sequence:

1. Choose one buyer and one recurring job. Prefer a narrow group you can already reach and understand. Check that the problem happens often enough for the intended pricing model.
2. Build the smallest complete result. Remove surface area, not the core outcome. A narrow product should solve one job properly rather than demonstrate ten unfinished possibilities.
3. Ask for money immediately. Payment tests urgency and trust. Define the price and success threshold before the launch so attention cannot be mistaken for validation.
4. Watch behaviour after purchase. Measure whether customers reach the result, return, ask for the same thing again, and recommend it. Follow the repeated behaviour even when it contradicts the original roadmap.
5. Keep support close to the founder. Treat every failure and question as product information. Fix reliability problems that compromise the promised result before expanding the feature set.
6. Attach distribution to useful work. Publish for the same buyer, build free tools for specific search intent, acquire adjacent assets selectively, and partner where the product genuinely extends what the partner already teaches.
7. Make an explicit allocation decision. Stop, reshape, acquire, or double down based on the evidence. Do not let the calendar make the decision by default.

The high-conviction lesson is not to imitate Louis-Lucas's launch count, product category, or partnership terms. It is to build a company that is corrected by reality early. Paid demand identifies a real commercial problem. User behaviour identifies the product. Relevant relationships improve the signal. Useful distribution lowers the cost of the next test. Ruthless allocation protects the time required to find the one worth scaling.

Sources - From Bankruptcy to Millionaire in 2 Years: How Tibo Louis-Lucas Built and Sold Tweet Hunter & Taplio - E168: Thibault Louis-Lucas, Co-Founder at Tweet Hunter - Creator Economy interview with Tibo Louis-Lucas