The standard DTC agency model puts a hard line between acquisition and retention. In 2018, that made sense. In 2026, the seam between the two has become the most expensive part of the system. Audience targeting has flattened, attribution has degraded, and customer leakage has become the dominant economic problem. The connected agency model is the structural response, and three diagnostic questions tell you whether your current setup is integrated or fragmented.

The standard way DTC brands buy marketing services has not changed in eight years. One agency runs paid media. A second agency runs email and SMS. A third runs SEO. A fourth runs influencer or UGC. Each one reports against its own metrics, on its own calendar, with its own quarterly review. Each one is accountable for a slice. None of them are accountable for the whole.

This structure made sense in 2018, when each function was specialised enough that separating them gave brands the best operators in each lane. It has stopped making sense in 2026. The economics of DTC have shifted in a way that makes the seam between acquisition and retention the most expensive part of the system, and that seam is precisely where the agency model puts a hard line.

This is the case for why the connected agency model has become structurally superior, what it actually changes in practice, and the questions worth asking of whatever setup you are currently running.

Why the old structure worked

The acquisition-only agency model worked when three things were true.

First, audience targeting was the lever. Sharper interest stacks, smarter lookalike construction, and better demographic filters were where performance came from. That work was deep, technical, and benefited from agencies that did nothing else. Specialisation was the moat.

Second, attribution was clean enough to trust. iOS still gave the platforms enough signal to credit conversions accurately. Last-click reporting roughly matched reality. Agencies could be measured against platform ROAS and the number meant something. Accountability worked.

Third, retention was a quieter line in the P&L. CAC was lower. Margins were higher. The cost of losing a customer was bearable because the cost of replacing them was modest. Brands could afford to treat retention as a secondary system, run by a separate vendor on a separate calendar.

In that world, three or four specialist agencies stitched together on the brand side was a defensible operating model. Each agency went deep on their lane. The brand kept the strategic view in-house. The seams between agencies were narrow enough to manage with weekly standups.

Why the structure has stopped working

Three structural shifts have made the seams between agencies the most expensive part of the system.

The audience layer has flattened. Advantage+ has eaten manual targeting. Lookalikes are deprecating. Interest stacks are largely irrelevant. The specialism that justified acquisition-only agencies in 2018 has been absorbed by the platforms themselves. We wrote about this in the Meta Q2 2026 targeting changes piece. What used to be the agency's edge is now the platform's default.

Attribution has become noisier than the metrics that depend on it. iOS, Advantage+, modelled conversions, and multi-touch journeys spanning weeks have eroded the precision that made platform ROAS a useful accountability tool. The metric the acquisition agency is reporting against no longer cleanly maps to what is happening in the business. See MER vs ROAS for the long-form version of this argument.

The cost of customer leakage has become the dominant economic problem. CAC has climbed 25 to 60 percent across categories since 2022. Margins are thinner. The cost of acquiring a customer who lapses without producing a profitable LTV has become large enough that retention quality now determines whether a DTC business compounds or stalls. The retention engine is no longer a quieter line in the P&L. It is the lever the P&L turns on.

When the audience layer has flattened, attribution has degraded, and retention has become the dominant cost lever, the case for separating acquisition and retention into different agencies on different reporting structures has collapsed. The seam is now the most expensive part of the system. Continuing to build the agency relationship around the seam is the structural problem.

The four costs of running separate agencies

In practice, four costs show up consistently when we audit brands running multi-agency setups.

Strategic incoherence

The acquisition agency optimises for new customer volume. The retention agency optimises for repeat revenue. Both metrics are reasonable. Neither is the metric the business should be optimising for, which is contribution margin per customer over time. With two agencies pulling against two different metrics, the strategic decisions usually default to whichever team is loudest in the monthly review, rather than to whichever decision compounds the business.

Calendar misalignment

Promotional campaigns, product launches, and seasonal pushes need to land synchronously across paid media and email. In practice, the two agencies run on different planning calendars, with different lead times. The result is paid media driving traffic to landing pages that have not been updated to reflect the email campaign, or email flows that promote products the paid media team has already pulled spend from. Each leak is small. The compound effect over a quarter is large.

Reporting fragmentation

Two agency reports arrive in different formats, with different metrics, on different cadences. Reconciliation against the P&L is the brand's job. Most leadership teams do not have the bandwidth for it. The result is leadership decisions made on the metrics each agency leads with, rather than on the metrics that map to cash. The brand pays for reporting twice and gets reconciliation never.

Cohort-level invisibility

This is the most damaging one. Repeat purchase rate by acquisition cohort, contribution margin per channel, payback timeline by source. These are cohort-level metrics that require both acquisition and retention data in one view. Multi-agency setups almost never produce them. The result is structural blindness at the exact level of analysis that determines whether the business compounds.

Why the dominant DTC agencies have not fixed this

It is a fair question. If the connected model is structurally superior, why have the largest DTC growth agencies not moved to it?

The honest answer is that the business model is the constraint. Kynship, RISE DTC, and Soar With Us are all acquisition-focused agencies with operating models, talent stacks, and commercial structures built around media buying and creative production. Each of them has talked, at various points, about wanting to do more retention work. None of them have built it into their core engagement model. Some have referral partnerships with Klaviyo or with specialist retention agencies. None run acquisition and retention as one connected commercial offer, on one P&L, with one accountability model.

The reason is that connecting the two is structurally hard. It requires hiring different talent, building different reporting infrastructure, and selling a different commercial model. For an agency with a large existing book of acquisition-only clients, the cost of the transition is high and the migration risk is real. The path of least resistance is to keep optimising the acquisition-only model and hope the structural shift slows down.

Webtopia built the connected model deliberately, with Oaks as the retention sister brand running full-time on Klaviyo and SMS strategy. The reason we built it this way, rather than bolting retention onto a Meta-focused agency, is that the connection has to live in the operating model itself. It cannot be a service line. It has to be the shape of the company.

What a connected agency actually does differently

In practice, the connected model produces four operational differences worth knowing about.

One reporting view. Acquisition and retention metrics live in the same monthly report, with cohort-level breakdowns that let leadership see the full picture in one place. The reconciliation work the brand had to do internally is now done at the agency level.

One calendar. Promotional campaigns, product launches, and seasonal pushes are planned across paid media and email in the same brief, with the same lead times. Creative is briefed once, executed for both surfaces, and landed synchronously. The calendar friction disappears.

One accountability metric. Contribution margin per customer over a defined window becomes the headline, with acquisition and retention metrics sitting underneath as supporting tactical views. The leadership conversation moves from "did Meta hit ROAS" to "is the engine producing profitable customers".

One commercial partner. When something is not working, there is one team to call. The "is this a creative problem, a retention problem, or a measurement problem" question gets answered inside the same conversation. The seam disappears.

The objection: is this just bundling?

A fair pushback. Bundling for the sake of bundling is rarely better than specialism. Two mediocre service lines under one roof is worse than two excellent ones under different roofs.

The argument is not that bundling is better. The argument is that the seam between acquisition and retention has become the dominant cost in the system, and the only way to remove the seam is to put both functions on the same commercial model. That is the structural point. A connected agency that ran retention badly would not be better than two specialists. A connected agency that runs both well is structurally superior, because the seam is gone.

The test for whether any agency model is "just bundling" is whether the operations underneath are actually integrated. One reporting view. One calendar. One accountability metric. Connected briefings on every campaign. If those things are real, it is not bundling. It is the connected model. If those things are absent and the agency just happens to offer both services, that is bundling, and the structural advantage is not there.

What to ask of your current setup

Three diagnostic questions for the next leadership meeting.

When you look at last quarter's marketing performance, can you pull one report that shows acquisition cost, retention engagement, and contribution margin by cohort in the same view? If yes, your setup is integrated. If you have to reconcile across two reports, you are paying for the seam.

When you ran your last major promotion or product launch, did the paid media team and the email team work from the same brief, on the same calendar, with the same creative? If yes, your calendar is integrated. If they worked separately and you stitched it together internally, you are paying for the seam.

When something is not working, do you call one team or two? If one, you have a connected partner. If two, you have a fragmented system that the agencies have asked you to manage on their behalf.

If most of the answers are "two", the seam is costing you more than the agencies are.

Where this leaves us

The connected agency model is not the right answer for every brand. Brands at $1M to $3M revenue can usually run with a single specialist on each lane and stitch the seam themselves. Brands above $5M, in our experience, cannot. The complexity is too high, the leadership bandwidth too constrained, and the cost of the seam too compounding.

The agencies that win at the $5M to $30M tier in the back half of 2026 will be the ones that have moved to one commercial offer covering acquisition and retention, with one reporting view, one accountability metric, and one team. The ones that have kept acquisition-only as the operating model will increasingly be selling against a structural disadvantage they did not have to carry.

Where to go next

Webtopia and Oaks run acquisition and retention as one connected agency, with one reporting view and one accountability model. If you want a view on whether your current setup is integrated or fragmented, book a call and we will walk through your reporting, your calendar, and your performance with you.

For the broader Full Picture argument, The Leaky Bucket Audit is the diagnostic piece that quantifies the cost of running the seam.

Frequently asked questions

Should I use one agency or two for DTC marketing?

For DTC brands above $5M revenue, one connected agency running both acquisition and retention is structurally superior to two specialists. The seam between acquisition and retention has become the dominant cost in DTC marketing economics in 2026, driven by flatter audience targeting, noisier attribution, and rising CAC. A connected agency removes the seam. Two specialists, however good individually, leave the brand to reconcile across two reporting views, two calendars, and two accountability models.

What is the difference between a specialist agency and a connected agency?

A specialist agency does one thing (paid media, or email, or SEO) very well, with deep operators and platform expertise. A connected agency runs multiple connected functions under one operating model, with one reporting view and one accountability metric. The argument for connected is not that bundling is better than specialism. It is that the seam between acquisition and retention has become the most expensive part of the system, and only a connected operating model can remove it.

Why is the acquisition-only agency model becoming outdated?

Three structural shifts. The audience layer has flattened as Advantage+ has eaten manual targeting, removing the specialism that justified acquisition-only agencies in 2018. Attribution has degraded as iOS, Advantage+, and multi-touch journeys have eroded the precision of platform ROAS. And the cost of customer leakage has become the dominant economic problem, with CAC up 25 to 60 percent since 2022, making retention quality the lever that decides whether a DTC business compounds.

How do I know if my current agency setup is connected or fragmented?

Three questions. Can you pull one report showing acquisition cost, retention engagement, and contribution margin by cohort in the same view? Did your last promotion run from a single brief on a single calendar across paid media and email? When something is not working, do you call one team or two? If most of the answers are "two", you have a fragmented setup, and the seam between your agencies is costing more than the agencies themselves.