Personalization Has Maxed Out – Here's What Comes Next

01 /The Concept

Standard personalization hits a wall: what happens inside a product tells you only so much about the person using it. Crosshatch was built for what comes next – "hyper-personalization" that draws on a significantly broader and more accurate picture of who each user actually is.

Conventional personalization is bounded by what happens inside the product itself. It draws on what users enter in their profiles and on their in-product behavioral history: purchases in a store, liked posts in a social feed, frequently accessed features in a SaaS tool. Apps do try to supplement this with third-party data, but that data is increasingly inaccurate, incomplete, and constrained by browser security policies and platform restrictions that keep tightening. There's a hard ceiling on how much can be personalized this way – and it's getting lower.

Crosshatch takes a different approach entirely, and explicitly declines to surveil users to do it. When an app integrates Crosshatch, users are presented with an opt-in list of external services. They can choose which ones – Gmail, fitness trackers, calendar, financial data connectors, whatever's available – to share data from with that specific app. They can revoke any of those permissions at any time through their profile settings.

With those permissions active, the app gains access to a much richer picture of who the user actually is and what they're doing right now – enabling more useful, more timely personalization than in-product data alone could support.

At the most basic level, this means better recommendations. But it opens the door to more genuinely helpful interactions. A healthy meal delivery service could detect that a user has travel plans coming up and proactively suggest pausing their subscription for that period – something many users forget to do. A travel marketplace could see where a user has been browsing lately on Instagram and what their location history looks like, then surface relevant trip ideas without requiring any manual filtering. A beauty marketplace could see what a user has purchased before and which creators she follows and admires, then suggest products she hasn't tried that fit the aesthetic she's drawn toward.

Apps connected to Crosshatch receive continuous, automatic updates from permitted external services. They can then query Crosshatch's API – specifying which AI model to use and passing their own prompts – to draw conclusions about what to recommend or surface next, automated and without manual configuration.

Current integrations include Plaid, Gmail, Fitbit, Strava, and Google Calendar – a small but meaningful set. A limited free tier is available for testing; commercial plans are $39 or $399 per month depending on API call volume.

Crosshatch raised $2.7M last fall and surfaced on Product Hunt recently.

02 /Why It Matters

Most people have had the uncanny experience of seeing an ad or a social post related to something they were just talking about out loud. The popular theory is that apps are listening through the phone microphone.

The more accurate explanation is a cognitive bias called the frequency illusion: once something is on your mind, you notice it everywhere – including in content that's always been there but that you previously filtered out. Whether the paranoid version is entirely wrong is harder to say.

What isn't in doubt: services do try to infer and track user preferences through every available channel, using methods that are technically legal but feel opaque. This divides users into two camps.

The first camp reaches for tinfoil – minimizing digital footprints, avoiding social accounts, using privacy-focused browsers and VPNs. The extreme end of this: operating online as if leaving no trace.

The second camp embraces the trade-off. If data collection is happening regardless, they'd rather it produce something useful. Crosshatch users belong to this group: give apps more information about them so the experience becomes noticeably better.

Who wins this standoff? The data suggests the pragmatists have the numbers:

- 90% of Americans say online privacy is important to them. - Only 71% of global users take any active steps to improve their privacy. - 62% of Americans believe it's simply impossible to live a normal life without companies collecting some data about them. - 58% say they're comfortable with collected data being used transparently to provide them some benefit. - 94% of companies say they can strike a balance between respecting user privacy rights and collecting personal information for marketing purposes.

03 /Opportunities

Crosshatch's model – transparent, consent-based personal data sharing in exchange for a better product experience – lands squarely in the intersection that 94% of companies and a majority of consumers say they're comfortable with.

As it becomes increasingly clear that some form of data collection is unavoidable, the share of users willing to trade data explicitly for tangible benefits will likely grow. A hard-core minority will always resist. But the trend runs the other way.

The opportunity is building platforms that let users share personal information with companies in an ethical, transparent, and genuinely beneficial exchange. The "what's in it for me" can be better recommendations, more relevant product discovery, timely proactive service, or any number of other convenience upgrades.

The specific mechanism Crosshatch uses – external service integrations, user-controlled permissions – is one model. But the design space for ethical personal data exchange is much broader: what kinds of information users would share voluntarily, what companies would do with it that's genuinely useful, and what forms of value that exchange creates for the user are all open questions with many potential answers.

There are likely many untapped answers to those questions. Find one, and your startup catches a rising tide.