Google Ads for Education Consultancies: Winning Students Without Buying Unqualified Enquiries

The short answer

Overseas-study consultancy is a crowded, heavily contested advertising category, and it has a defining structural problem that most other verticals do not: enquiries are easy and enrollments are hard. A consultancy can generate an enormous number of form submissions and phone calls from students who will never meet an entry requirement, will never be able to fund an application, or are three years away from applying and are simply reading. Every one of those looks identical in the Google Ads conversion column.

So the account cannot be built to maximize enquiries. It has to be built around qualification — which queries you buy, what the page asks before it accepts a submission, and which conversion action the bidding algorithm is actually optimizing toward. That is what this post is about. It assumes you already know how to run a Search campaign; if you want the mechanics of running an account from Nepal at all — payment routes, feature availability, the regulatory position — that is a separate and prerequisite topic covered in the complete guide to Google Ads in Nepal, and this post does not repeat it.

A note on what I am and am not bringing to this. My operating experience is cross-market: I have managed more than USD 170,000 in Google and Meta ad spend across seven-plus countries, and the large majority of that was outside Nepal. So I am not going to tell you what a study-abroad keyword costs in Kathmandu, because I do not have that figure and nobody publishing one has it either. What I can do is describe how this category behaves, which parts of the platform Nepal removes, and how to build an account whose incentives point at enrollments rather than at form fills.

Why this vertical breaks the normal lead-gen playbook

Most lead-generation advice assumes a rough proportionality: more enquiries, more customers. It holds well enough for a plumber or a dentist, where the enquiry itself is close to the transaction and the qualification test is short. It fails badly in student recruitment, because three things are all true at once and each of them is unusual.

• The decision is enormous and irreversible in a way almost no other advertised purchase is. A student is choosing a country, spending a family's savings or a loan, and reorganizing years of their life. Nobody makes that choice from an ad.
• The eligibility gate is real and external. Academic record, English test scores, funding evidence and a visa decision all sit outside the consultancy's control, and any one of them can end an application that was otherwise perfect.
• The interested population is far larger than the eligible population, and both are far larger than the population who will apply this cycle. Advertising reaches all three groups with the same ad and the same page.

That third point is the whole problem in one line. In most verticals, interest is a reasonable proxy for intent. Here, interest is abundant and cheap, and it correlates weakly with anything that produces revenue. An account that gets rewarded for interest will find more of it, indefinitely, at a falling cost per enquiry — and the cost per enquiry falling is precisely the signal that everyone reads as improvement.

The consideration cycle, and what it does to your attribution

A study-abroad decision does not have a funnel so much as a long, looping, non-linear consideration period. A student searches about Australia in one month, sits an English test months later, compares the United Kingdom and Canada after that, discovers a funding problem, disappears for a semester, returns, asks a cousin who studied in Japan, and eventually searches for a consultancy in their own city. Between the first search and an enrollment there can be months, and frequently longer than a year.

Why last-click will systematically mislead you

Last-click attribution assigns the entire outcome to the final click before conversion. In a short cycle that is a tolerable simplification. In this cycle it is not a simplification, it is a specific and predictable distortion, and it always distorts in the same direction. The final click before an enrollment is nearly always a branded query, a consultancy-selection query, or a direct visit. The research queries that put the student on the path — the destination comparisons, the eligibility questions, the course searches — will have happened far earlier, often outside the conversion window entirely, and will show as having produced nothing.

Follow that logic one step further and you get the failure. Reporting says the destination-research campaign produced no enrollments. Somebody pauses it. Six months later branded and consultancy-selection volume declines, because the thing feeding it stopped, and nobody connects the two events because they are separated by more time than any reporting period covers. This is the single most common way a study-abroad account gets quietly dismantled by a competent analyst doing exactly what the data appeared to say.

What to do instead

• Extend the conversion window well beyond the default. If your real cycle is many months, a thirty-day window is measuring a different business than the one you run. Set it against the observed gap between first touch and enrollment in your own CRM, not against a platform default.
• Use a data-driven attribution model rather than last click, and accept that the credit it distributes will be fractional and untidy. Untidy and roughly right beats clean and structurally wrong.
• Read assisted conversions and path reports as a first-class metric for upper-tier campaigns, not as a curiosity. For destination-research campaigns, assist volume is the primary measure and direct conversions are the secondary one.
• Judge research-tier campaigns on remarketing audience growth and on later return rate, which are observable within weeks, rather than on enrollments, which are not.
• Import the enrollment itself from the CRM as an offline conversion so that the platform eventually sees the outcome that matters, even if it sees it late.

All of that depends on the measurement layer being sound in the first place, which in this vertical it usually is not. Conversion actions counting the same submission twice, a thank-you page firing on a browser back-button, form and phone conversions weighted identically when they are worth completely different amounts — these are the errors that make every argument in this post unprovable. The conversion tracking implementation guide covers the setup end to end, and the four-step tracking audit is the faster way to find out whether what you already have is telling the truth. Do that before acting on anything below.

Intent tiering by query — the core of the account

This is the section that does the most work. Most consultancy accounts are structured by destination — an Australia campaign, a UK campaign, a Canada campaign — which is intuitive and almost useless, because it groups together queries that sit at wildly different distances from an enrollment and then judges them on a single blended cost per enquiry. Structure by intent tier first, and by destination inside it.

There are five tiers that matter in this vertical. They are not stages a student moves through in order — the cycle is not linear — but they are reliable indicators of what the searcher is doing at the moment they type, and that is what you are buying.

Tier one — destination-country research

Queries about studying in a country generally: cost of living, whether a country is a good option, post-study work rules, comparisons between two destinations. The student is orienting. They may be years from applying, they may not have sat an English test, and they very often do not yet know whether they are eligible for anything.

This traffic is cheap, plentiful, and will produce enquiries all day if you point a form at it. Those enquiries will be the ones your counseling team quietly stops calling back. But the tier is not worthless — it is where the relationship starts, and a consultancy that is absent here is absent from the beginning of every journey. The treatment is to fund it deliberately and modestly, send it to content rather than to a hard enquiry form, and measure it on audience growth and assisted conversions rather than on enrollments. Give it a capped budget and hold the cap.

Tier two — test preparation

English test preparation queries — classes, practice material, score requirements. This tier is different in kind from the others, because for many consultancies test prep is a service in its own right with its own revenue, not merely a signal of study-abroad intent.

That distinction has to be reflected in the account or the reporting becomes incoherent. If you sell test preparation, run it as its own campaign against its own conversion action and judge it on enrollments into classes. If you do not sell it and are bidding on these queries purely as an intent signal, be honest that you are buying a signal, not a customer, and treat the tier the way you treat destination research. What you must not do is pool test-prep enquiries with consultancy enquiries and report a blended number, because the two convert to completely different things on completely different timelines.

Tier three — visa and eligibility

Questions about student visa requirements, financial documentation, academic gaps, refusal history, dependents, or whether a particular profile qualifies for a particular country. This is the most valuable tier that most consultancies under-invest in, and the reason is instructive: these queries look informational, so they get classified as content topics rather than as commercial ones.

They are commercial. A student searching whether a two-year academic gap is acceptable has a specific problem, is far enough along to have hit it, and is looking for someone who can tell them what to do about it. That is exactly the moment a consultancy is useful. It is also the moment where the temptation to over-promise is strongest, which the section on claims deals with directly — the correct answer to a hard eligibility question is an honest assessment, and honesty here converts better than reassurance because the student has usually already been told what they want to hear by someone else.

Tier four — scholarship and funding

Scholarship searches, tuition-fee queries, education loan questions, cheapest-country comparisons. Treat this tier with more care than any other, because it has a specific pathology: it is highly motivated and frequently unfundable. The student is engaged, will fill in your form, will take the call, and may be searching precisely because the money is not there.

This is the tier that produces the enquiries that consume the most counseling time per enrollment. It is not a tier to exclude — funding questions are normal and plenty of funded students ask them — but it is the tier where landing-page qualification earns its keep most visibly. A funding-source question on the form, asked plainly and without judgment, sorts this traffic before a human is spent on it. Measure this tier on enrollment rate specifically, not on enquiry volume, and expect the two numbers to diverge more here than anywhere else in the account.

Tier five — consultancy selection

Queries where the student is choosing a provider rather than researching a destination: best education consultancy in a named city, consultancy reviews, competitor names, and your own brand. This is the tier closest to an enrollment and it should be the first claim on budget, ahead of everything above it.

Two things go wrong here regularly. The first is under-funding brand defense on the grounds that people searching your name would have found you anyway — in a vertical where competitors bid aggressively on rival brand terms and where a student comparing three consultancies is genuinely undecided, that assumption is expensive. The second is competitor bidding done carelessly: it is legitimate to bid on a competitor's name, but the ad and the page have to give the searcher a reason of their own rather than a comparison that reads as an attack, and the trademark rules on what may appear in ad text are worth checking before writing the copy rather than after a disapproval.

One structural note on running this: the tiers need genuinely separate campaigns with separate budgets, not ad groups inside a shared campaign. Shared budgets flow toward whatever converts cheapest, and what converts cheapest is always the research tier. Separating the budgets is the mechanism that stops the account from re-optimizing itself back into buying interest.

Seasonality: budget should follow deadlines, not the calendar month

Study-abroad demand is not evenly distributed and neither is the value of an enquiry. Intakes have application deadlines, offer windows and course start dates set by universities and destination-country processes, and an enquiry's worth depends almost entirely on where it lands relative to those dates. The same enquiry that can still be turned into an application for the next intake is worth many times the identical enquiry that arrives after the window closed and has to be held for months.

A flat monthly budget ignores that completely. It spends the same money in the weeks where an enquiry has a live path to an application and in the weeks where it does not, which means a meaningful share of the annual budget is systematically bought at the wrong time. And because the enquiries arriving in dead weeks still look like conversions, nothing in the account reports a problem.

Building the calendar

1. Write down the actual deadline calendar for the destinations and partner universities you work with — application deadlines, offer decision periods, visa lodgement lead times and course start dates. Use published dates, not received wisdom.
2. Work backwards from each deadline by the length of your own process: counseling, document collection, test sitting if required, application preparation and submission. That backward-extended period is your productive advertising window for that intake.
3. Overlay the windows. In a multi-destination consultancy they will overlap, and the overlaps are where budget should concentrate.
4. Identify the genuinely dead weeks — the periods where an enquiry cannot reach the next deadline and must wait for the one after. Reduce spend there rather than eliminating it, because research-tier presence still has value and restarting a campaign from zero costs learning.
5. Shift tier mix rather than only total spend. Close to a deadline, consultancy-selection and eligibility tiers deserve nearly everything. Far from one, research and test-prep tiers are the sensible place for a smaller budget to sit.

The second half of that last point is the part most accounts miss. Seasonality in this vertical is not only a volume question, it is a mix question. The tier that deserves budget changes with the calendar, because what a student can usefully do changes with the calendar, and an account that only turns a single budget dial up and down is expressing a much cruder version of the same idea.

Qualification before the sales team

Here is where I part company with standard conversion-rate optimization advice, and I want to be explicit about it because the standard advice is not wrong in general — it is wrong here.

The general rule is that fewer form fields produce more submissions. That is true, well-established, and it is exactly why it gets applied thoughtlessly in a vertical where it does damage. The rule optimizes for submissions. It is the correct rule when a submission is cheap to process and roughly equivalent to every other submission. In study-abroad advertising a submission is neither: it costs a counselor a real conversation, frequently a long one, before anyone learns whether the student could have enrolled at all. The scarce resource is not traffic and it is not form completions. It is counseling hours.

The three questions worth losing volume for

• Intended intake. When does the student actually want to start? This single field separates a live applicant from someone browsing two years ahead, and those two people need completely different treatment, different follow-up cadence and different priority. Asking it costs a small share of submissions and immediately re-sorts the entire queue.
• Funding source. How is the study being funded — family funds, education loan, scholarship being sought, employer, undecided? Asked plainly and without judgment, this is the highest-value question on the form, because funding is the constraint that ends the most applications and it is discoverable in one field rather than in one phone call.
• Academic background and test status. Highest qualification completed, broad grade band, and whether an English test has been sat or booked. Eligibility is an external gate and this is the cheapest possible look at it before a human is spent.

Add those and submitted volume will fall. That is not a side effect to be minimized, it is the mechanism working. What should rise is the share of enquiries a counselor can progress, the speed at which the queue is triaged, and the number of enrollments produced per counseling hour — which is the only efficiency number in this business that actually matters.

How to ask without losing the students you want

Qualification questions fail when they read as gatekeeping. A few practical rules make the difference. Ask them in the student's language of decision rather than in yours — intended start is a friendlier question than intake cycle. Use structured options rather than free text so the answers are usable in routing and scoring. Never make an eligibility question feel like a rejection: an undecided funding option must be present, because a student who is genuinely undecided is not the same as one who cannot pay, and forcing a false answer destroys the value of the field. And put the qualification questions after the contact fields, so a partial submission still leaves you with a contactable person.

The mechanical part is in the same family as what I have written about elsewhere: once the qualification fields exist, they become the inputs to routing and scoring, so the enquiries that meet a live-intake, funded, eligible profile reach a senior counselor quickly while the rest go into a nurture track. The guide to building lead scoring automation covers how to build that scoring model without inventing a points system nobody validated, and the lead routing system that does not drop leads covers the fallback paths, which matter more in a vertical where the enquiry that gets missed on Friday has often chosen a competitor by Monday.

What a close rate looks like, and what it does not tell you about Nepal

The only performance figures I will put in this post are my own, and they need their denominators attached. Across the accounts I have managed — largely outside Nepal, across seven-plus countries — consultancy-type businesses have run at a lead-to-customer close rate of around 8 percent, meaning roughly eight of every hundred leads generated went on to become paying clients; the cross-market figure across all the niches I have worked in is around 12 percent on the same denominator. Those are downstream business outcomes rather than ad-platform conversion rates, which is precisely why they are worth stating: most advertising reporting never sees them at all.

Two things follow, and the second matters more. First, consultancy sits below the cross-market typical, which is what you would expect from a category where eligibility and funding gates sit outside the seller's control — a lower close rate is the structure of the vertical, not evidence of a bad process. Second, and to be completely explicit: these are cross-market figures from accounts largely outside Nepal. They are not a Nepali benchmark, they are not a target for your consultancy, and anyone quoting them as either has misread them. Use them as a demonstration that close rate is the number worth measuring, then go and measure your own.

What Nepal specifically takes away from this playbook

Three Google Ads features that this vertical would ordinarily lean on are not available for Nepal. They are absences from Google's own published country lists rather than opinions, and each one changes what a Nepali consultancy has to build.

No lead form assets — the landing page carries everything

Google states that lead forms are only eligible to serve in some countries, and publishes the list. It runs to 85 countries and includes Bangladesh, India, Pakistan and Sri Lanka. Nepal is not on it. So in-ad lead capture is unavailable to a Nepali consultancy, and every enquiry has to be captured on your own page. (Even if that changed, a second gate would bind: the same documentation sets lifetime Google Ads spend requirements for lead forms on video, display and Search headlines that most Nepali accounts will never approach.)

For this vertical specifically, that constraint is closer to a gift than a handicap. In-ad forms are optimized for frictionless capture, which is the exact opposite of what a study-abroad account needs. Being forced onto your own page means the qualification questions live where you control them, the data lands in your own systems rather than in a platform export, and the student sees the consultancy before they see a form. Build accordingly: the page has to carry the credibility, the destination detail and the qualification, because the ad unit can carry none of it. The general version of that work is in where paid traffic actually leaks on landing pages, and every point in it applies with more force when the page is the only conversion surface you have.

No call reporting — and most enquiries arrive by phone

Google publishes the list of countries eligible for call reporting and Google forwarding numbers — thirty markets including India and Indonesia — and Nepal is not among them. You can still attach a call asset with your own number and it will serve. What does not exist is the forwarding number, the resulting call counts, campaign and keyword attribution for calls, and call-duration conversions.

In a category where a large share of student enquiries arrive as a phone call, this is a serious gap, and it interacts badly with everything above. If form fills are visible and calls are not, smart bidding will optimize toward whichever queries produce form fills — which, in this vertical, skew toward the lower-intent research tiers, because those are the students most willing to type into a form rather than call. The account will steadily buy more of the traffic that produces the enquiries it can see, and the enrollment number will not move.

• Use a third-party call tracking provider with dynamic number insertion where one supports Nepali numbers, and import calls as offline conversions. This is the only route that restores keyword-level attribution.
• Failing that, run a distinct number for paid traffic. Channel-level truth with no keyword granularity is far better than nothing and takes an afternoon.
• Ask and record the enquiry source at first human contact, every time, as one required field. Unglamorous, and often the only ground truth available.
• Import enrollments from the CRM as offline conversions, so that the platform eventually optimizes toward the outcome rather than toward the observable proxy.

Nepali is not a targetable language

Google's published language-targeting table lists 51 languages and Nepali is not one of them, while Hindi, Bengali, Urdu, Tamil and several other South Asian languages are. Google also states that ads created in a language not supported for ads language targeting will be disapproved. Practitioners target Nepal geographically and set language targeting to English — that is common practice and inference, not documented Google guidance, and it should be tested in your own account with disapprovals watched.

For a study-abroad consultancy the practical effect is smaller than it would be for many other categories, because the audience is by definition preparing to study in English and is disproportionately likely to be searching in English already. But it does mean the segment of prospective students who research primarily in Nepali is harder to reach deliberately through this channel, and it is a limitation to acknowledge rather than a solved problem.

What Nepal does give you: geography

Google's official geo-targets file contains 114 active Nepal targets — the country, five provinces, districts, cities and around eighty municipalities, including Kathmandu, Lalitpur, Biratnagar, Butwal, Hetauda, Damak, Nepalgunj, Rajbiraj and Birendranagar. For a consultancy with physical offices, that granularity is the most useful structural lever in the account. Consultancy-selection queries are inherently local — a student choosing a provider is choosing one they can walk into — so the near-enrollment tier should be targeted tightly around the offices, while research-tier campaigns can reasonably run wider.

Advertising claims and student trust — the non-negotiable section

This section is not a compliance footnote and it is not here to be balanced. Study-abroad advertising in this region has a reputation problem, and that reputation problem is the single largest constraint on conversion for every honest consultancy in the market. Students and their families arrive at your page already braced to be misled, because many of them have been, or know somebody who was.

The three claims to never make

• Guaranteed visa approval, or any visa success framing that implies a controlled outcome. A visa is decided by a foreign government against criteria and discretion that no consultancy influences. Presenting it as guaranteed is a promise about someone else's decision.
• Guaranteed admission. Admission is an institutional decision made against academic criteria. A consultancy can prepare an application well; it cannot commit an admissions committee to accept it.
• Guaranteed jobs, placement or permanent residency after study. These depend on labor markets and immigration policy that shift independently of anything either party does, and they are the promises that cause the most severe harm because they are the ones students borrow against.

Two separate reasons make these non-negotiable, and the second is the one I want to leave you with. The first is straightforward: misleading claims are an advertising-policy risk, and an account built on them is one review away from stopping. Google's policies on misrepresentation are worth reading directly rather than paraphrased from a blog post, and if you are unsure whether a specific claim crosses a line, the honest test is whether you could deliver on it if the student held you to it.

The second reason is that these are life-altering, expensive, largely irreversible decisions made by young people and financed by families who often cannot absorb the loss. A student who enrols on the strength of a guaranteed-visa promise and is refused has lost money that was borrowed, a year of their life, and in many cases a family's confidence in them. That harm is not an abstraction, and no reasonable amount of enquiry volume justifies participating in it. I am not going to cite specific regulations or advertising codes here, because the position does not need one — the argument stands on what the claim does to the person who believes it.

Why restraint is also the commercial position

The uncomfortable objection is obvious: competitors making these promises will out-convert you at the click. Sometimes they will. But look at what the promises actually cost the category. The reason a prospective student reads a consultancy page with suspicion, the reason they ask three friends before they call, the reason the sales cycle is as long as it is — a large part of that friction is inherited distrust manufactured by exactly these claims. In a market where skepticism is the default, the differentiated position is not a bigger promise, it is a verifiable one.

Which is a commercial argument, not only an ethical one. What actually converts a wary student is specificity about things you control and can evidence: which universities you have working relationships with, what your process is step by step, what documentation you prepare, what happens if an application is unsuccessful, what your fees are and when they are charged. An honest eligibility assessment that tells a student their profile is weak for one destination and strong for another is more persuasive than any guarantee, because it is the first thing they have heard that could not have been said to everyone. Advertise the process, not the outcome.

Remarketing across a long cycle without becoming intrusive

Remarketing is genuinely well-suited to this vertical, because the cycle is long and the reason someone did not convert is usually timing rather than disinterest. It is also the mechanism most likely to make a consultancy feel like it is following people around, which in a category already carrying a trust deficit is an expensive impression to create.

Segment by what they were doing, not by how long ago

Most remarketing setups segment by recency alone, which is close to meaningless here. A student who read a destination guide four months ago and a student who abandoned a counseling booking four days ago are at opposite ends of the process. Build audiences from behavior and let membership duration reflect the tier: research-page visitors deserve long membership durations and light frequency, while abandoned-form and abandoned-booking audiences deserve short, urgent, tightly capped ones.

• Cap frequency deliberately and low, and review it as a real setting rather than leaving it to default. In a cycle measured in months, an uncapped campaign accumulates an enormous number of impressions against the same person.
• Exclude converters and current applicants across every campaign. A student who has already enrolled seeing your acquisition ads for months is a support problem and a visible sign that nobody is managing the account.
• Rotate the message across the cycle. Serving the same enrollment prompt for six months teaches the audience to ignore the brand. Move from destination content, to eligibility guidance, to intake deadline reminders as the calendar advances.
• Use intake deadlines as the natural intensity trigger. Increasing pressure because a genuine deadline is approaching is useful information; increasing pressure for no external reason is nagging, and students can tell the difference.
• Suppress the family-sensitive placements. This decision is discussed at home, and an ad that turns up in front of a household at the wrong moment does more harm than the incremental reach is worth.

The judgment call underneath all of this is simple: remarketing should feel like the consultancy remembered the student, not like it is tracking them. If the ad is offering something the student would plausibly want right now — a deadline they might have missed, guidance on a question they were clearly researching — it reads as helpful. If it is the same enrollment prompt for the fourth month running, it reads as pursuit.

Search terms, negatives, and the specific waste in this vertical

Every account needs negative keyword discipline, and the general list for service businesses applies here too. What is worth calling out is that this vertical has categories of waste that are unusual, high-volume, and invisible unless someone reads the search terms report deliberately.

• Job and employment intent. Queries about working abroad, work visas and job placement look adjacent to study migration and are a completely different person with a completely different need.
• Direct university intent. Students searching a specific university's own admissions pages want the university, not an intermediary. Some of that traffic is workable and much of it is not — read it before deciding, and split rather than blanket-blocking.
• Government scholarship and exam names that you have no relationship with. High volume, high enquiry rate, near-zero enrollment, and a reliable source of enquiries that consume counseling time and end in an apology.
• Free, sample, PDF, syllabus and past-paper modifiers. Research and study-material intent, not consultancy intent, and it arrives in enormous volume against test-prep keywords.
• Other countries and other services you do not handle. Obvious, and consistently present in accounts that added destinations to the website but never to the negative list.
• Competitor and franchise-enquiry terms you did not intend to buy — people looking to open a consultancy rather than to use one.

The cadence matters as much as the list. In a market where call data is missing, the search terms report is one of the few uncontaminated signals in the account, so read it weekly rather than monthly during an active intake window. The negative keyword list every service business needs is a reasonable starting frame to build the vertical-specific list on top of.

The metric that lies

Enquiry volume up, enrollments flat. That is the signature failure of this vertical and it is worth stating precisely, because it does not present as a failure. It presents as a good month.

The mechanism is not mysterious. If the conversion action in the account is a form submission, then smart bidding will find the cheapest form submissions available, and the cheapest form submissions in this category come from research-tier queries and scholarship-seeking traffic. Cost per enquiry falls. Enquiry count rises. Every headline number in the reporting improves. Meanwhile the counseling team is working harder for the same enrollments, and the conversation about why marketing is not delivering happens six months later with no data connecting the two.

The second numbers that catch it

• Qualified-enquiry rate: the share of enquiries that meet a defined viability bar — live intake, plausible funding, plausible eligibility. If enquiries rise and this rate falls, the extra volume is not worth having.
• Enrollments per counseling hour. The only genuine efficiency measure in the business, and the one that moves in the opposite direction to cost per enquiry when the account is drifting.
• Cost per enrollment, not cost per enquiry. Late, noisy, and the only cost figure that describes reality.
• Enquiry-to-appointment rate by intent tier. This is the earliest available warning, because appointment booking is observable within days while enrollment is not.

The structural fix, once those numbers exist, is to move the conversion action further down the funnel. Optimize toward a booked counseling appointment rather than a form submission, or import qualified-enquiry and enrollment events from the CRM as offline conversions. Bidding will always chase whatever it has been pointed at, so the only durable solution is to point it at something closer to revenue.

Where this breaks in practice

• The CRM cannot report enrollment by source, so none of the tier-level measurement above can be produced. This is the most common blocker and it is not a marketing problem — the fix is upstream, in how enquiries are recorded and how outcomes are written back.
• Counselors do not record why an enquiry died. Without a reason code, you can see that scholarship-tier enquiries do not enroll but never learn whether it was funding, eligibility or timing, which are three completely different account changes.
• Response time. In a market where a student is comparing three consultancies simultaneously, an enquiry that waits a day is often already lost. Response time destroys more budget in this vertical than bidding strategy ever will, and no account structure compensates for it.
• Qualification fields get quietly removed. Somebody notices submissions fell after they were added, reverts the change, and volume returns along with the workload. Document why the fields exist, next to the form, or this will happen roughly every eight months.
• Seasonal budget shifts get approved once and then forgotten. The calendar moves, the campaign settings do not, and the account spends peak-window budget in a dead window the following cycle.
• Attribution windows left at default, which quietly recreates the last-click distortion the account was restructured to avoid.

Honest scope — when this is the wrong channel

Every post here says no to something, and here it is. Google Ads is a genuinely strong channel for this vertical because the intent is searchable and the customer value is high. It is not the right first move for every consultancy, and there are four situations where I would say so plainly.

1. If nobody can answer enquiries quickly. A consultancy where enquiries wait a day is not ready to buy more of them. Fix response time first; it costs nothing and it is worth more than any campaign change.
2. If there is no system recording outcomes. Without a CRM that can tie an enrollment back to a source, you will be optimizing on form submissions indefinitely, which this entire post argues against. Build the recording layer before the media.
3. If the counseling capacity is already saturated. Buying more enquiries into a team that cannot work the ones it has produces slower responses to everybody, including the students who would have enrolled. Capacity is the constraint; advertising will not relieve it.
4. If the business model depends on volume rather than fit. If the plan is to generate as many enquiries as possible and convert whoever happens to be convertible, this channel will serve that plan efficiently and the outcome will be a large CRM, an exhausted team, and a reputation you will spend years repairing. That is a business-model decision, not a marketing one, and no account structure will fix it.

There is also a cheaper unbuilt asset for most consultancies, and it costs no foreign currency at all: a complete Google Business Profile with current information and genuine reviews. Google Business Profile is available in Nepal, consultancy-selection searches are inherently local, and for a firm with physical offices this is frequently the highest-return thing that is not yet done.

A realistic starting sequence

This is the order I would work in for a consultancy starting or restructuring an account. The first campaign does not appear until step six, which is deliberate — everything before it determines whether the campaign can be evaluated at all.

1. Write down the intake calendar and the internal process length. Deadlines, offer windows, course start dates, and how long your own counseling-to-application process actually takes. Everything about budget pacing depends on this document and it usually does not exist yet.
2. Define what a qualified enquiry is, in writing, agreed with the counseling team. Intended intake within a defined horizon, a plausible funding position, a plausible eligibility position. If the team cannot agree the definition, that disagreement is the finding.
3. Fix the measurement layer. Conversion actions deduplicated, form and phone treated distinctly, attribution windows set against your real cycle length, data-driven attribution enabled, and CRM write-back planned so enrollment can eventually be imported as an offline conversion.
4. Close the call gap as far as you can. Third-party call tracking if a provider supports your numbers; a distinct paid-traffic number if not; source logged at first contact regardless. Nepal has no native call reporting, so this is a build, not a setting.
5. Rebuild the landing pages with qualification in them. One page per destination and per major eligibility question, mobile-first, with intended intake, funding source and academic background asked after the contact fields.
6. Launch the consultancy-selection tier first, alone. Brand defense plus provider-choice queries, exact and phrase match, tight geography around the offices, with a negative list from day one. This tier is closest to an enrollment and the fastest to evaluate.
7. Add the eligibility and course-specific tiers next, each as its own campaign with its own budget. Point them at the matching pages rather than at the homepage.
8. Add the research and test-prep tiers last, with capped budgets and with their success measured on assisted conversions and audience growth rather than on enrollments. Hold those caps when someone points out how cheap the enquiries are.
9. Build remarketing from behavior, segmented by tier, with frequency capped and converters excluded, and rotate the message as the intake calendar advances.
10. Reconcile monthly against the CRM. Enquiries by tier against qualified enquiries against enrollments. That reconciliation is the entire management system for this account, and no in-platform report substitutes for it.

What this post is, and what to verify yourself

The mechanics here are cross-market: they describe how a high-consideration, externally-gated, long-cycle lead-generation category behaves, and they come from managing accounts across seven-plus countries rather than from Nepali student-recruitment data specifically. The Nepal-specific parts — no lead form assets, no call reporting, Nepali absent from the language-targeting table, 114 geo targets down to municipality level — come from Google's own published country lists, retrieved and read, and you can check each of them yourself in an afternoon.

What you will not find in this post is a cost per enquiry, an enrollment rate, a market size or a claim about what share of Nepali students do anything. Those numbers do not exist in any published, checkable form for this market, and in a category where families make borrowed-money decisions, publishing invented ones would be worse than publishing nothing. If you see them elsewhere, ask where they came from before planning around them.

If you want the method underneath rather than the vertical specifics, the Signal-to-Revenue Framework documents how I sequence measurement, structure and bidding in any market, and everything in this post is that sequence applied to a category where the signal is unusually noisy. I run that practice through Arcetis, and the reason this post spends more time on qualification than on campaign settings is simply that in student recruitment the settings are the easy part.

Related reading