Be in the answer when a student asks which course to take.

Increasingly, yes, and the change happened fast: AI use in the college hunt went from 26% of high-school students in spring 2025 to 46% by the end of the year. Among international students the pattern is stronger still, with 96% of AI users saying the guidance matched or beat official sources. Advertising there does not replace your website; it makes sure you are in the comparison being drawn.

By context rather than by list. The placement is matched to the conversation, so we build coverage from the questions your admissions team hears — entry requirements, cost, format, transfer credit, career outcomes — and weight it toward programs that need students. There are no lookalike audiences and no interest targeting yet, which is exactly why your own measurement has to carry the weight.

Plenty, and it is better said plainly. No video placements, no lookalike or interest audiences, thinner reporting than mature platforms, and creative options that are limited compared with paid social. That is why we keep the budget capped, buy on clicks first, and lean on the pixel, the event structure and the destination page for performance.

By separating the stages and being patient about the last one. Inquiry and application events are tracked distinctly, enrollment comes back from the CRM, and the benchmark is your existing cost per enrolled student rather than a platform metric. For context, education search clicks averaged $6.23 in 2026 at an 11.4% conversion rate, so a cheaper inquiry that never enrolls is easy to spot and easy to cut.

Some overlap is expected and worth monitoring rather than fearing. Because one team holds the channels, we can see whether total applications rise against a flat combined budget, which is the only cannibalisation test that means anything. If the new platform simply reshuffles the same students at a higher price, we move the money back and say why.

Small, steady and honest. A modest monthly cap held for a full recruitment cycle produces a usable answer; a large burst that ends before applications close does not. We size it against your current spend in the first session, and the cap never moves without your approval — useful when the budget has to survive a committee.

Handled carefully, yes, and carefully is the operative word. Placements are labelled as advertising, the claims we make are limited to what your prospectus already says, and cost and requirement information stays accurate because a student who feels misled becomes a complaint rather than an enrollment. We keep admissions in the approval loop for every message.

Usually a handful of program pages rather than a rebuild. The click arrives from a conversation where requirements and cost have already been discussed, so the page has to confirm those quickly and offer a short next step. We build those destinations alongside the campaigns and hand them over documented.

Faster than in mature channels. Advertiser numbers went from 2,585 to 3,770 in one week when the minimum spend dropped, and prices follow competition. So we review weekly during a test rather than monthly, and treat the current pricing as a window rather than a permanent state.

Yes, and that is the sensible way to buy it. Our paid search, AI search and analytics teams work from one plan, so an application is attributed once instead of claimed by three vendors. Taking this platform on its own is equally fine, and everything stays in accounts you own.

It matters more, not less. Assistants lean on what they can read about you, so accurate program information, clear cost pages, outcomes data and third party coverage all shape the unpaid part of the answer. The two work together: advertising buys presence now, while our AI search team improves what the model says about you when nobody is paying.

Two lines only: a fixed monthly management fee scoped to the number of programs in play, and the media budget, which stays capped and is paid by your institution directly to OpenAI. There is no percentage of spend, so our incentive stays pointed at efficient enrollment rather than volume. Included is the measurement build, the question map, campaign management, the destination pages and a monthly written summary. Excluded is anything we would have to invent: we do not promise application numbers before a term of real data exists.

Ask what each channel buys. Traditional paid listings buy people already using your program vocabulary; paid social buys attention from an audience that was not looking; conversational advertising inside ChatGPT buys the research stage, where a student is comparing higher education options and forming a shortlist. A sensible mix keeps search funded first, holds a creative testing budget on social, and treats this platform as the cheap early-mover experiment while placements are still underpriced. We will tell you when the marketing mix does not need a fourth channel.

A named senior buyer who knows education marketing, supported by our measurement and creative specialists. You meet them in the first session and they stay on the account — no handover to a junior team once the contract is signed, and no ticket queue. Weekly working sessions are with the people making the changes, so questions get answered by whoever pressed the button.

You get a written verdict with three options: scale, hold at the current cap, or stop — each argued from the numbers, including the ones that do not flatter us. Everything built along the way stays useful regardless: the event structure, the question map and the destination pages all keep working for your other channels.