For a period of several years, growth hacking was the dominant framework in early-stage technology companies. Clever, unconventional tactics that produced disproportionate results with minimal spend.
The canonical examples are well known and get repeated at conferences a decade later. What gets discussed less is why almost none of them would work now.
The conditions that made it work
The tactics depended on a specific environment, and every element of it has changed.
Open platforms. Social networks and email providers offered permissive access to contact lists and posting capabilities. The famous viral loops were built on integrations that platforms subsequently restricted heavily.
Uncrowded channels. Early adopters of a channel get outsized returns because attention isn't yet contested. That applied to email, then social, then content marketing, then influencers. Each was extraordinarily effective briefly and then became expensive.
Low user expectations. Practices considered acceptable then — importing contact lists to send invitations, aggressive notification defaults — would now generate complaints and, in many jurisdictions, regulatory attention.
Cheap capital. Growth-at-any-cost strategies were viable when funding was abundant and the market rewarded user numbers over unit economics.
Each of those has reversed. The tactics didn't stop working because people got better at defending against them; the environment they exploited was dismantled.
The arbitrage framing
The most useful way to think about what growth hacking actually was: a series of arbitrages.
An arbitrage exists when a channel is underpriced relative to its value. Someone notices, exploits it, others follow, the price rises, the opportunity closes.
That's a description of every channel in the last two decades. Search advertising was extraordinarily cheap and became competitive. Social organic reach was free and then throttled. Content marketing worked until everyone produced content. Each cycle lasts a few years.
Framed this way, growth hacking wasn't a discipline with transferable principles. It was a name for being early, and being early isn't a method.
What replaced it
The practices that have proven durable are less exciting.
Retention before acquisition. A product people stop using cannot be grown by any amount of acquisition — you're filling a leaking container. Working on retention first is unglamorous and it's the constraint in most businesses that plateau.
Unit economics. Acquisition cost against lifetime value, with an honest payback period. This was widely ignored during cheap capital and has become the central question.
Product-led distribution. Building distribution into the product itself — collaboration features that require inviting others, outputs that carry attribution, use cases that naturally involve more people. This is the one genuine survivor from the growth hacking era, and it's a product design discipline rather than a marketing tactic.
Positioning. Being clearly the right answer for a defined group. Less exciting than a viral loop and considerably more durable, because it doesn't depend on a channel remaining cheap.
The metrics problem it left behind
One lasting damage worth naming. Growth hacking established a set of metrics that persisted after the tactics didn't.
Signups, downloads, registered users, monthly actives with a permissive definition of active. These were the currency of the era and they measure exposure rather than value.
A great many organisations still report against these, and they can be improved by acquisition tactics that produce no economic benefit. A signup that never returns is a cost, not a result, and it appears in the same column as a customer.
The correction has been towards revenue-linked measures — activated users, paying customers, revenue retention — which are harder to move and mean something.
What still transfers
Some genuine principles survived, and they're methodological rather than tactical.
Experiment properly. Structured testing with adequate samples and honest analysis. The discipline was real even where the tactics were era-specific.
Instrument everything. Understanding where users drop out requires measurement built in from the start.
Look at the funnel end to end. Optimising acquisition when the problem is activation wastes money, and the two are frequently owned by different teams who don't look at each other's numbers.
Find the constraint. At any moment one thing limits growth. Working on anything else produces no result regardless of effort. Identifying it correctly is most of the job.
Where the next arbitrage is
Nobody knows, which is the point. If it were identifiable it would already be priced.
What I'd say is that the pattern suggests it will be in a channel that currently looks unpromising or beneath serious attention, that it will be exploited by people who don't have budget for the expensive channels, and that it will last two or three years before becoming competitive.
Which means the transferable skill isn't knowing the tactics. It's being willing to try things that don't look like what everyone else is doing, and being early enough that they're still cheap.
What the era got right
It would be unfair to end without crediting the genuine contribution, which was cultural rather than tactical. Before it, growth was treated as marketing's problem and product as engineering's, and the two rarely met.
The insistence that growth is a product concern — that distribution should be designed rather than bolted on afterwards — was correct and has largely been absorbed into how software companies are organised. Growth teams sitting inside product, with engineering resource and their own roadmap, are now standard, and that structure came from this period.
So the tactics expired and the organisational insight stuck, which is a reasonable legacy for any movement.