Sample shortlist · candidates written by us, real screener output
PPC Specialist
This is what every search delivers: finalists screened against the brief, each with a score, screening notes, structured answers, and a graded work sample. These sample candidates were written by us and run through the real screener — the scores and notes are its unedited output, not a mock-up.
The brief they were screened against
Role: PPC Specialist · Band: $1,500–2,550/mo
Must-haves: runs Google Ads to a target cost per acquisition, a diagnostic order when CPA misses, conversion tracking you would trust with the budget, willing to recommend pausing spend
Nice to have: Meta Ads, GA4 and Google Tag Manager, Performance Max, landing-page testing
Scored by the same rubric the production screener runs on real applicants (paid-ads, claude-sonnet-4-6), on 2026-09-23. Nothing below was edited afterwards — including where it went against a candidate. The rubric and weights are published.
Felipe A.
SCREEN 84/100Screening notes
Strongest evidence is the roofing account narrative (Search Partners at 31% spend / 2 conversions turned off before touching bids, $214→$108 CPA over three months) and the tracking-first diagnostic order that explicitly cross-checks Google's conversion count against CRM/call log before changing anything — both details the questions did not supply. The work sample is unusually disciplined: splitting repair vs. maintenance vs. brand into separate campaigns with budget rationale, deferring broad match until conversion history exists, and naming cost-per-qualified-call at 60 seconds as the only Week 1 metric shows genuine structural thinking rather than platform-UI familiarity. The one material gap is Meta Ads depth — described as 'smaller' with no CPA or diagnostic story attached, which matters only modestly given the brief weights Google heavily. At $2,300 he sits near the top of band but the roofing turnaround and the tracking-before-bids discipline are specific enough to justify presentation to the client.
Experience
6 years in paid search, the last 4 at a Medellín agency that runs Google Ads for US home-services companies: HVAC, roofing, plumbing, garage doors. I manage 9 accounts spending $4K–$38K a month, all bid to cost per lead, and I report weekly from the client's own lead records, not from Google Ads alone. One example: a roofing account paying $214 a lead against a $120 target. Search Partners and one broad-match keyword pulling in job seekers were most of it; three months later it was $108. My Meta Ads experience is smaller — lead ads for two clients.
Scenario answer
First, the arithmetic. $4,000 at $180 is about 22 conversions. To hit $90 on the same spend the campaign needs about 44, so twice the conversion rate, half the cost per click, or some of both. A 6% CTR says the ads get clicked. The problem is which searches trigger them, or what happens after the click. 1. Tracking, before I change anything. I compare Google's 22 with what the client actually got from paid search this month: CRM leads, call log, form emails. If they got 35 real leads, the true CPA is about $114 and the job is fixing tracking, not bids. If they got 12, it's $333 and Google is counting something that isn't a lead. Then the conversion actions: which are primary, whether leads count once per click, whether any stopped recording after a site change, whether calls count at all. Conversions are also reported against the click date, so the last week always looks worse than it will end up. 2. Search terms for the month, sorted by cost. Negatives go in. On match types: in my accounts, broad match without conversion history behind it is where the money usually leaks. Any keyword that has spent more than $180 with no conversion gets paused or cut back to exact. I also split out brand terms: if they're in this campaign, they flatter the 6%, and non-brand CPA is worse than $180. 3. Settings. Search Partners and Display on or off, location set to 'Presence' rather than 'Presence or interest', and where the clicks actually came from. On a roofing account last year, Search Partners was 31% of spend for 2 conversions. I turned it off before anything else. 4. The landing page. I fill in the form on my phone, check that the ad's offer is actually on the page, and compare conversion rate by device. 5. Bidding, last. On Maximize Clicks it's buying clicks, not leads. With a $90 target on Maximize Conversions, it's being asked for half this month's CPA; in my experience that mostly stops it spending. I'd set the target near the real number and bring it down in steps. When I'd say pause: if tracking is broken, pause all but exact-match brand and core terms until it's fixed. We'd be spending blind. Otherwise the fixes get two weeks, which at the current budget is about $1,800, roughly 20 conversions at target. If CPA isn't under about $120 and falling by then, I tell the client to pause non-brand search and fix the page or the offer first. I say it in the same report as the number, not a week later.
Work sample
PRODUCT: residential AC repair for an HVAC company in the Phoenix metro. Repair calls only, not new systems. Budget $2,000 a month; Google plans a month as 30.4 days, so the daily budgets below add up to $65.79. CAMPAIGN 1: PHX | Search | AC Repair | Exact+Phrase. $1,400/month ($46.05/day) One ad group per problem, each with its own ads and a landing page that matches: - AC repair: [ac repair], [ac repair near me], "ac repair service" - Not cooling: [ac not cooling], "ac blowing warm air", "ac running but not cooling" - Emergency: [emergency ac repair], "same day ac repair" - Failed parts: "ac capacitor replacement", "ac compressor repair" Exact and phrase only at launch. Broad match needs conversion history to steer it, and a new account has none. Location: 25-mile radius around the office, Presence only. Schedule: all hours, because in July the calls come at night too. CAMPAIGN 2: PHX | Search | Brand | Exact. $100/month ($3.29/day) [company name], [company name ac repair]. Cheap, and it keeps the business above any competitor bidding on its name. CAMPAIGN 3: PHX | Search | AC Maintenance | Phrase. $500/month ($16.45/day) - Tune-up: "ac tune up", "ac maintenance" - Pre-season: "ac service before summer" Lower urgency and cheaper clicks, and a tune-up visit often turns into a repair. Left out on purpose: new-system searches like "ac replacement cost". They're the most expensive clicks with the longest decisions, and $2,000 won't buy enough of them to learn anything. Shared negative list on all three: jobs, salary, training, school, diy, how to, parts, manual, car, auto, window unit. Bidding: Maximize Clicks for two weeks, capped at Keyword Planner's high top-of-page bid, then Maximize Conversions once calls are coming in. Calls count as conversions at 60 seconds or longer; forms count once per click. THE ONE METRIC FOR THE FIRST TWO WEEKS: cost per qualified call, meaning a call of 60 seconds or more that the office confirms was a real job in the service area. Two weeks is about $920 of spend. That's too little to judge a final CPA, but enough to see whether the calls are real. If half are job seekers or people outside the area, I fix targeting before I touch bids.
Cristina F.
SCREEN 84/100Screening notes
Strongest concrete evidence: the B2B client story (form-fill to SQL bidding, CPA $610→$390 in four months) is specific, directionally plausible, and names the exact mechanic (offline conversion import from CRM) — the kind of detail that is hard to fabricate and directly maps to the role's must-haves; the scenario answer demonstrates a genuine diagnostic order (tracking integrity before bids) with named steps (Tag Assistant, CRM reconciliation, lag awareness) rather than generic advice; the work sample is unusually clean — three campaigns with a logical budget split, exact/phrase rationale explained against a real failure mode ('dental software' drifting to patients), the 'login' negative call-out, and a two-week decision rule tied to break-even math rather than a vanity metric. Biggest concern: the salary ask of $2,450 sits at the very top of the band and is hard to justify on the profile alone without a live skills test or client reference — the 14-account/$180K spend claim is unverified, and the scenario answer does reuse the $180/$90 figures supplied in the question, so the diagnostic framing is evidenced but the numbers themselves are not candidate-generated. Verdict: advances to final interview; confirm the SQL/CPA story with a reference or a brief live audit before approving top-band compensation.
Experience
8 years in paid media, the last 4 as PPC lead at a Manila agency working for US e-commerce and B2B software companies: 14 accounts, about $180K a month in combined spend. I'm the person the team sends tracking problems to — Tag Manager, GA4, enhanced conversions, offline conversion imports from HubSpot and Salesforce. Last year I moved a B2B client from bidding on form fills to bidding on sales-qualified leads imported from their CRM. Cost per form fill went up; cost per SQL went from $610 to $390 in four months. Meta is my second platform, not my first.
Scenario answer
Two questions before any optimising: is the $180 real, and where did the $90 come from? Is the $180 real? Four checks, in this order, because each can move the number more than any bid change could. 1. What's being counted. In the conversion summary, which actions are primary. I've seen a page view of the contact page counted next to the real form submit, and I've seen the real form not counted at all after a redesign. 2. Does the tag fire? I submit the form myself with Tag Assistant open, then check the conversion arrives in the account. 3. Does Google agree with the business? This month's conversions in the account against the CRM for the same dates, paid search only. If the CRM has 30 and Ads has 22, the real CPA is about $133 and the fix is tracking, not bids. 4. Lag. Conversions are recorded against the click date, so the last days of any month come in short. Where did the $90 come from? If this campaign has never had a month under $150, $90 is a hope, and I'd rather say so now than after two months of optimising toward it. If it was at $90 in the spring, something changed, and the change history usually shows what. Then the account, in this order: search terms by cost and by match type (a 6% CTR with a bad CPA is usually the right clicks on the wrong searches, or brand terms mixed in), then networks and location settings, then the landing page on a phone, and bidding last, because Smart Bidding can only chase what the conversion data tells it. When I'd recommend pausing. I ask what a customer is worth. Say a sale brings $300 of margin and one lead in three buys: break-even is $100 a lead, so at $180 every lead loses $80. - If tracking is broken: pause everything but brand until it's fixed, even if that takes a week. - If tracking is fine and the fixes can't get CPA under break-even within the next $1,500 of spend, about 11 days at this budget: pause, and say what has to change first — the offer, the page or the price. I write that rule down and send it before the spend starts, so we've agreed what 'not working' means before the numbers arrive.
Work sample
PRODUCT: appointment scheduling and reminder software for US dental practices, $249 a month per location, sold after a 20-minute demo. Budget $2,000 a month; at Google's 30.4 days a month the daily budgets below add up to $65.79. CAMPAIGN 1: Search | Non-brand | Dental Scheduling | Exact+Phrase. $1,450/month ($47.70/day) - Scheduling software: [dental scheduling software], "dental appointment software", "dental practice scheduling" - Online booking: "online booking for dentists", [dental online booking] - Reminders and no-shows: "dental appointment reminders", "reduce dental no shows" Exact and phrase only. On broad match, 'dental software' drifts toward patients looking for a dentist, and there's no conversion history yet to steer it. CAMPAIGN 2: Search | Competitors | Exact. $450/month ($14.80/day) One ad group per competitor, for the three tools the sales team says it loses deals to: [competitor name], [competitor name pricing], [competitor name alternative]. The ads compare contract length and price, and don't use the competitor's name in the ad text. CAMPAIGN 3: Search | Brand | Exact. $100/month ($3.29/day) Negatives on all three: dentist near me, dental insurance, cleaning, jobs, salary, school, free, template, login. The last one matters here, because patients search 'dental portal login' to reach their own dentist's system. Conversions: 'demo booked' is the primary action, counted once per click. Pricing-page visits and form starts are secondary, so they're visible but don't steer bidding. From day one, sales-qualified demos are imported from the CRM, so in a month or two we can bid on the demos sales actually wanted. Bidding: Maximize Conversions without a target for the first four weeks, then a target CPA set near what the campaign has actually achieved. THE ONE METRIC FOR THE FIRST TWO WEEKS: cost per booked demo, read against the client's ceiling of $400. About one demo in four becomes a customer, and they'll pay up to $1,600 to win one. Two weeks is about $920, so I'm looking for two or more demos, not a precise CPA. If $920 buys none, I go back to the search terms and the demo form before the rest of the month is spent.
Thabo N.
SCREEN 52/100Screening notes
The candidate's strongest concrete evidence is the tool list (GA4, GTM, Meta, Looker Studio) and the 4-year agency tenure with named verticals, but no numbers, client names, or real results appear anywhere — every claim is adjective-driven ('passionate,' 'always looking to improve') with zero specificity the questions did not supply. The scenario answer contains a critical CPA-thinking failure: the candidate explicitly refuses to recommend pausing spend, frames a 2× CPA miss as needing 'two to three more weeks,' and treats a 6% CTR as proof the ads are working rather than checking whether that CTR reflects high funnel waste — the brief specifically required willingness to recommend stopping spend and a diagnostic order, both of which are absent or inverted. The work sample is the most damaging section: broad match on a $1,200/month budget for a $149 product with no stated target CPA, Maximize Clicks as the opening bid strategy, and CTR named as the two-week success metric — this is a textbook vanity-metric orientation the brief explicitly flags as disqualifying, and the budget allocation ($600/month to remarketing before the funnel is validated) shows no payback thinking. Verdict: competent on platform mechanics at a surface level, but the core commercial judgment the role requires — CPA ownership, willingness to pull spend, attribution honesty — is either absent or actively contradicted by the answers; do not advance.
Experience
4 years in digital marketing at a Johannesburg agency, the last 2 focused on Google Ads for local and international clients in e-commerce, education and professional services. I build campaigns, write ads, do keyword research, manage budgets and prepare the monthly performance reports. I hold the Google Ads Search and Display certifications and I am passionate about data-driven marketing and always looking for ways to improve results.
Scenario answer
A CPA of $180 against a target of $90 means the campaign is not converting well enough, even though the 6% CTR shows that the ads are relevant and people are interested in them. First I would check the keywords, looking at which ones use the most budget without converting, and pause the weakest. I would also go through the search terms report and add negative keywords for irrelevant searches. Next I would review the landing page. Since the CTR is good, the ads are working, so the problem is most likely after the click. I would check that the page loads quickly, works well on mobile and has a clear call to action, and suggest improvements to the client. I would also make sure the conversion tracking is set up correctly. Then I would look at the bidding. If the campaign is not already using a CPA target, I would switch it to Maximize Conversions with a target CPA of $90 so that Google optimises towards the client's goal. I would not recommend pausing the campaign at this stage. The changes need time to show results, and I believe it is better to keep optimising than to stop. Instead I would lower the daily budget while the changes take effect, give it two to three more weeks, and send weekly reports so the client can see the progress.
Work sample
PRODUCT: an online Excel course for beginners, $149. Campaign 1: Search - Excel Course ($1,200/month) Ad group 1, Excel course: excel course, excel training, learn excel (broad match) Ad group 2, Excel certification: "excel certification", "excel certificate online" (phrase match) Campaign 2: Search - Brand ($200/month) Brand name (exact match) Campaign 3: Display Remarketing ($600/month) People who visited the course page but did not buy. Total: $2,000/month. I would use broad match in the main ad group to reach more people and discover new keywords, and phrase match where the intent is more specific. I would add negative keywords such as 'free' and 'jobs'. The bidding would start on Maximize Clicks to build traffic, then move to Maximize Conversions. The one metric I would judge the first two weeks on is CTR, because it shows whether the ads and keywords are relevant to the audience before there is enough conversion data.
This is the artifact every search delivers
Post the role and the same screener runs on your real applicants — the questions arrive already written, and you read every finalist in full before any fee is due.
The same screener, other roles
- Customer Support VA
- Executive Assistant
- Bookkeeper
- Operations Manager
- Social Media Manager
- Data Entry Specialist
- Content Marketer
- Junior Developer
- Graphic Designer
- Sales Development Rep
- Real Estate VA
- E-commerce VA
- Bookkeeping Assistant
- Virtual Assistant
- Video Editor
- Copywriter
- SEO Specialist
- WordPress Developer
- Shopify Developer
- Accountant
- Project Manager
- Lead Generation Specialist
- Web Designer