Your website is already doing outbound. You just can't see it.

Most companies run inbound and outbound as two teams with two stacks that never talk. The website is the one asset both motions already touch, and the only place they can be joined.

At Webflow, I started our sales and marketing teams during our fastest growth years. Hiring the founding members, testing tactics, opening up new segments. That meant I spent a lot of time sitting on both sides of a wall I didn’t fully understand at the time.

On one side was inbound: a content calendar, an SEO consultant, a website nobody wanted to touch, and a form. On the other was outbound: a list, a sequencer, an enrichment vendor, and a rep working through it. Different owners, different budgets, different tools, different definitions of a good week.

The only bridge between them was the form. Somebody filled it out, it became a lead, and it crossed over.

Every other visitor fell into the gap. The one who read the pricing page twice, compared us against a competitor, and left. In most B2B funnels the share of traffic that ever identifies itself sits in the low single digits, and I’d guess yours looks about like that. Which means the overwhelming majority of the demand your inbound work generates is invisible to the team whose whole job is finding demand.

Meanwhile that team is buying a cold list.

When we launched Ploy, I wrote that the craft of marketing got buried under dashboards. This is the operational version of the same problem. Two teams, one website, and no shared memory between them.

Inbound and outbound were never two things

I don’t think this split is a strategy. I think it’s a tooling accident that hardened into an org chart.

Inbound tools were built around publishing: a CMS, a page builder, an analytics dashboard. Outbound tools were built around contact records: a database, an enrichment API, a sequencer. Neither category had a reason to know about the other, so they didn’t. Companies that wanted both bought both, hired someone to reconcile them in a spreadsheet, and called the spreadsheet attribution.

Think about what actually happens when someone lands on your site from a comparison query, reads your integrations page, checks pricing, and bounces. One person, one intent, one signal. Whether the follow-up gets called nurture or prospecting is an org chart question. The buyer gets a relevant follow-up or they get silence.

Companies pick silence because the signal never left the analytics tool.

Answer engines made this urgent

I’ve made the argument before that the website matters more in the age of AI, not less. Here’s the part I didn’t spell out.

When a buyer asks an AI assistant to compare vendors in your category, the assistant reads your site, synthesizes it against your competitors, and answers. The buyer may never click. When they do, they show up already briefed.

So raw traffic falls while intent per visit rises. Fewer sessions, each one worth more, which means a funnel built on volume assumptions starts to leak.

The content that wins in that world is built to be cited rather than ranked. Structured, specific, machine-legible. Comparison pages, technical documentation, real answers to real questions. And the accounts that arrive after reading an AI summary of your category are pre-qualified in a way no purchased list can match.

Fewer, better-informed visitors make anonymous exits a lot more expensive than they were three years ago.

The loop, in four moves

Four stages. The fourth one feeds the first.

1. Get found by machines

Your site has two audiences now, and one of them is a crawler working on behalf of a buyer.

That makes the technical layer matter again. Structured data on every page. Clean sitemaps and a current llms.txt. Comparison and alternatives pages written to be quoted. Documentation that answers questions in complete, self-contained blocks, because that’s the unit an answer engine extracts.

Test it on yourself. Ask ChatGPT, Claude, and Perplexity to compare you against your three closest competitors. If they describe a product you stopped selling, or cite a competitor’s page for a claim that should be yours, you’ve found a content gap with a name and an address. We built our AEO audit around exactly this, because it was the first thing I wanted to know about our own site.

2. Identify who showed up

Reverse IP resolution and firmographic enrichment turn a slice of anonymous sessions into company-level records: name, size, industry, funding stage, tech stack. The company, plus the pages they read.

There are plenty of vendors that can hand you some version of this data. 6sense and Factors.ai identify visiting accounts and attach firmographic context. RB2B and Common Room push further toward person-level identification, particularly for US traffic. The coverage and method vary, but the output is familiar: a company, the pages it read, and sometimes a likely person behind the visit.

A cold list tells you a company exists and roughly matches your ICP. Visitor identification tells you a company was on your pricing page yesterday, twice. That is a much better signal. But it is still just data.

A company name in a dashboard is not an outbound motion. To become useful, the signal has to be matched to the right CRM account, enriched, scored against your ICP, checked against open opportunities and existing outreach, routed to an owner, and made available to your sequencer, website, paid audiences, and attribution. That stitching is the hard part. Without it, the data lives in a point solution, gets piped into Slack as a notification nobody acts on, and evaporates by Thursday.

3. Act while the signal is warm

Three moves, and which one you reach for depends on how strong the signal is.

Personalize the site. A known account coming back to your homepage shouldn’t see what a stranger sees. Swap the logo wall for their industry. Lead with the integration they care about. This is ABM that costs nothing per account, because the page assembles itself from account data. Hex runs this at a scale their team could never produce by hand, with every page tied to live account data and nobody waiting on engineering.

Build the page before the conversation. When a target account starts showing up in your traffic, spin up a page for them. Their use case, their stack, their objections, on your domain. The most useful thing a rep can send already exists before anyone asks for it. This is the gap I keep coming back to: the ABM page you know you should build almost never clears the bar of “important enough” to pull a designer and a developer off other work. So it never gets built.

Reach out with a reason. “I noticed you visited our site” is unsettling. “You were comparing tools in this category, here’s the honest teardown of where we win and where we don’t” is a useful email, and it’s only writable because you know what they read.

Paid closes the same loop from the other end. Build the audience from identified accounts, run creative against it, and give every variant its own landing page. TNT Growth pairs each ad creative and copy variation with its own targeted page instead of dumping traffic onto a homepage, so the ad and the page ship as one unit.

4. Close the loop back to step one

Once you can trace which accounts arrived, what they read, and which ones closed, you can answer a question most companies can’t: which content earns the right visitors?

A comparison page that pulls two hundred visitors from your exact ICP beats a listicle that pulls twenty thousand from nowhere. Attribution built on account quality instead of session volume changes what you publish next, and that decision belongs back at step one. Clay is running this loop on programmatic SEO, using Ploybooks to turn their own data into hundreds of on-brand pages, where a single Ploybook spins up an entire page type in minutes.

Content attracts the right accounts, identification names them, outreach and ads engage them, and the outcome tells content what to make more of. Continuously, in the background.

Why it only works as one system

You could assemble all of this from parts. A site builder, an SEO platform, a visitor identification vendor, a sequencer, an ad manager, a CDP to reconcile them, and a data engineer to keep the pipes from breaking.

Plenty of companies have. Every seam is a place where context dies. The enrichment tool knows the account is a 400-person fintech but not which pages they read. The analytics tool knows the pages but not the account. The ad platform knows the click but not the outcome. The CRM has a record with no behavior attached. Each system holds a fragment, and fragments only become a decision when a human sits down and assembles them, which means it happens for the ten accounts someone has time for and never for the other four hundred.

What closes those seams is shared state. One system where the pages, the traffic, the identities, the campaigns, and the outcomes are the same data, so acting on a signal doesn’t require a handoff.

That’s what we built Ploy around. PloyWeb builds and optimizes the site. PloyGrow identifies who arrives and reaches them. PloyAds creates the campaigns and attributes the results. All three read and write the same context, so each one can act on what the others find without a human in the middle passing notes.

Turn your site into automated outbound today

Here’s how I’d run the whole loop on Ploy, in order. Connect your domain and Ploy slurps your existing site first, so it already knows your design system, components, and brand before any of this starts.

1. Audit your AI visibility. Run the AEO audit on your domain. It shows you how the major assistants describe you against your competitors, and where they’re citing someone else’s page for a claim that should be yours. Every wrong answer is a page brief.
2. Turn on visitor identification and read it for two weeks. PloyGrow resolves the anonymous sessions into companies and ties each one to the pages they actually read. Sync it to HubSpot and let your reps see it next to the accounts they’re already working. You’ll find companies your sales team has never heard of, and companies they’ve been chasing for months.
3. Build a page for each of your top ten identified accounts. Ask PloyWeb for it. It pulls the design system from your live site and the account data from your CRM, so the page comes out on-brand and specific to their situation instead of a template with a logo swapped in. Then watch whether reps send them.
4. Schedule a Ploy to refresh your account data every day. Have it re-enrich the accounts that showed up overnight, attach the pages each one read, push the updates into your CRM, and drop the accounts that moved into Slack. Attribution stops being a quarterly reporting exercise and becomes something your reps read with their coffee, scored on ICP fit and pipeline instead of sessions.
5. Turn the winners into a Ploybook. Once you know which page type earns the right accounts, a Ploybook spins up that whole page type on a schedule or off a trigger. Clay turned their own data into hundreds of on-brand pages this way. The ones that pull volume and nothing else get retired.

None of this needs new headcount. It needs the data in one place.

Your website is already doing the work

The old model treated the website as a destination. You spent money getting people there, and the site’s job was to catch the small fraction willing to fill out a form. Everything else was waste, and the waste was invisible, so nobody counted it.

That made sense when your site couldn’t tell you anything. It can now.

The next generation of visitors won’t just be humans browsing. They’ll be agents gathering information and making decisions on someone’s behalf, and the companies that win will be the ones whose website is awake for both. Most companies are already running half of that loop. The other half is sitting in their analytics, unread.

Bring your domain and watch Ploy get to work. Start free.