Optimisation is not growth. Silos are the side effect.
Dragos Marica Associate Director, Performance Marketing, Directive
Interviewed by John Horsley
Published
Dragos Marica is Associate Director of Performance Marketing at Directive, the largest independent B2B agency in the world. He has been in performance marketing officially for seven years, with two or three additional freelance years before that. His agency career spans Publicis, OMD, and now Directive, where the focus is on turning ad spend into revenue and profit. He started by freelancing on social media activity (2018), then moved into paid media as the industry developed, and left his hometown for London for the first opportunity to make an impact across big advertisers. In this conversation he sets out the silos as the cause of efficiency-confused-for-growth problem; the one platform generating 80% of revenue is a risk diversification principle; the traditional SEO plus an AI layer discipline for AI-driven discovery; the Reddit and LinkedIn citations are the new authority observation; the refresh content as the quick win tactic for both SEO and AI; the LTV-to-CAC dropping from 3:1 to 2:1 signal of collapsing paid-media efficiency; the creative is at the epicentre of performance now observation; the test, test, test career advice; and the closing instruction: follow the people and you'll find your spot.
Silos cause efficiency to be confused for growth
On the trap.
Companies say they want growth and then optimise for efficiency (cost-per-click, cost-per-lead, short-term metrics). The side effect of silos. Departments are set with narrow KPIs or OKRs. Set the marketing team against leads and they'll do everything to get leads, without caring about what's down the funnel: are they qualified, are they opportunities, are they revenue? They won't care unless you push them to.
The change required: incorporate marketing, sales, and product so they work more together. Efficiency will be earned if companies make that change. Not everyone is there. The more enterprise, the more they lean into the old siloed structure.
On platform dependence.
If 80% of revenue comes from one platform, that's significant risk. Diversify. Any company that relies on a significant piece of revenue from one platform or one source only should rethink the strategy.
Traditional SEO plus an AI layer for the zero-click world
On the discipline.
Everyone wants to optimise for AI search and citations now. The reality: it's traditional SEO with contextual items overlaid to make content visible to AI Overviews and to LLMs like Perplexity, ChatGPT, and Claude.
Plenty of companies are now generating leads and revenue on the back of it, particularly in B2B, but they won't necessarily see the traffic coming in. You don't see the clicks any more, but you see the outcomes of the business with a good strategy on the back of it. Many times it's traditional SEO done a bit better.
On the new authority.
Citations are now coming from Reddit (not a focus prior to 2025-2026) and from LinkedIn. What? From a social media platform? Yes. The easiest quick win: refresh content. Put it on the agenda. Update and refresh so it's viable from both an SEO and an AI perspective.
LTV-to-CAC from three-to-one to two-to-one: the collapse signal
On the moment.
Traditional Google traffic is either too expensive or not delivering the return you want. Clients have migrated their search behaviour to other platforms (ChatGPT, Claude), which reduces query volume and spikes competition.
Our LTV-to-CAC was three-to-one last year and is two-to-one this year. A significant drop. You have to assess and look into other areas. Many people are now looking at LinkedIn ads or Meta ads on the B2B side and realising they've been under-investing in the area that creates the demand.
On the consequence for creative.
Creative is at the epicentre of the change. Everyone should have a strong performance-creative strategy to capture interest and drive demand to the area they want.
Brand at the epicentre when low-hanging fruit disappears
On the split view.
Depends on where the company is. Profitability-wise, they can invest more or less. On digital media, companies should invest more into brand than they did prior, because all the low-hanging fruit known previously is disappearing. It's tough out there. Discoverability is the right word. Companies aren't investing enough in it. Long-term revenue engines.
On community.
The tech brands have built community around the product. Most powerful way of staying relevant. Real people's stories rather than icons or invented stories. Creating those communities on the back of real people's stories has a huge impact. LinkedIn is picking up from an SEO perspective. Reddit the same. People-centred assets are generating growth.
Test, test, test, and follow the people
On iteration.
Investments should be based on what is working and what you have data on. You can take risks. Not big leaps; depends on business maturity. Step by step. 20% additional investment to content and SEO, then up it again as the results improve. Iteration linked with experimentation.
On entering the field now.
Right now is the most difficult moment to enter the market because as a new specialist you're not learning how things were done; you're developing the structure and the revenue engine for the future. Go in, break things, test, test, test, see what works, and continue with what works.
Follow the people and you'll find the spot for you. I asked businesses if they wanted me to post their shoes on Facebook. Yeah, do it. That was my first contract. Then I noticed Meta offered paid ads; I asked clients if I could boost reach and engagement. From there it became revenue and profit.
The question for the board
If LTV-to-CAC has fallen from three-to-one to two-to-one as paid efficiency collapses, what share of digital spend builds brand at the top of the funnel?
