EVERYTHING YOU
NEED TO KNOW.

For Canadian and US brands in 2026, programmatic media costs typically range from $5–$15 CPM on display, $20–$35 CPM on online video, $30–$50 CPM on CTV, and $8–$15 CPM on digital audio. Most reputable agencies layer a 15–20% management fee on top of media spend. Realistic minimum spend for measurable programmatic results in our experience is $5,000–$10,000/month total — below that, you don't have enough reach or data to optimize. Want our actual 2026 Toronto pricing across all 13 DSPs we run? See our programmatic page or book a free audit.

Programmatic advertising is the automated buying of digital ad inventory in real time across thousands of websites, apps, streaming TV platforms, and audio services — using software (called a demand-side platform, or DSP) instead of a human salesperson. When someone loads a webpage, a millisecond-long auction decides which ad gets shown to them based on who they are, what they're doing, and how much advertisers are willing to pay. Done well, programmatic is the most precise way to reach a specific audience at scale. Done badly, it's how budgets quietly vanish. More on how we run programmatic.

Google Ads is search-intent-based: someone types a query into Google, and you bid to show up. Programmatic is audience-based: you target a person across the open web, CTV, audio, and apps based on who they are or what they're doing, regardless of whether they searched for anything. Google Ads typically captures bottom-of-funnel demand (people ready to buy). Programmatic builds the top and middle of the funnel (awareness, consideration). For most B2C brands the right mix is roughly 30–40% search, 60–70% programmatic + paid social. See how we build the right mix.

Across our client book we actively use 13 DSPs and each has a sweet spot. The Trade Desk leads on premium open web + CTV. Google DV360 wins for YouTube and integrated Google inventory. StackAdapt is strong for Canadian-specific buys and self-serve simplicity. Yahoo DSP is underrated for native and audio. Amazon DSP is essential for retail-adjacent brands. Adobe Advertising Cloud is best when you're already in the Adobe stack. The right DSP depends on your audience, KPIs, and budget tier — there's no universal winner. Get a custom DSP recommendation.

In 2026, expect CTV CPMs of $25–$45 in Canada and $30–$55 in the US, with premium streaming inventory (Disney+, Netflix, Hulu, Crave) running $40–$70 CPM. Completion rates above 90% are achievable when creative and frequency capping are set correctly. The metric that actually matters for CTV isn't CPM though — it's incremental reach against your target audience. We've seen $35 CPMs deliver better ROI than $25 CPMs because the audience match was tighter. See our programmatic and CTV approach.

Linear TV is traditional broadcast and cable: your ad runs at a fixed time to whoever is watching that channel, bought weeks or months in advance with almost no targeting control. CTV (Connected TV) delivers ads through internet-connected televisions — Hulu, Peacock, Disney+, Crave, Tubi — using programmatic buying. The practical differences are significant. With CTV, we target specific households by income, purchase intent, or geography rather than a demographic guess. We track completion rates (typically 92–95% on CTV vs. roughly 70% on linear) and adjust budgets mid-flight based on performance. Linear TV CPMs run $10–$15; CTV CPMs land at $25–$45 in Canada and $30–$55 in the US. For most mid-market brands we run today, CTV generates measurably lower cost-per-reach against a defined audience than linear does. See our full CTV and programmatic approach.

CTV advertising typically runs $20–$40 CPM for standard inventory, climbing to $40–$65 CPM for premium placements on platforms like Netflix or Hulu. In practical terms, a $5,000/month budget reaches roughly 125,000–250,000 targeted households. What makes CTV worth that premium over traditional TV is precision — we layer behavioral data, income targeting, and first-party audience segments on top of the buy, so you're reaching the right people, not just people watching TV. We run CTV across Samsung Ads, The Trade Desk, and Amazon Advertising, giving us access to inventory that single-platform setups can't touch. If you're spending $3,000+/month on digital, CTV is almost always worth testing. Learn more about our programmatic & CTV programs →

Retail media is advertising you buy inside a retailer's ecosystem — Amazon Ads, Walmart Connect, Instacart, Target Roundel — where shoppers are already in buying mode. US advertisers will put about $69.3 billion into it in 2026, up from $58.8 billion in 2025, and Amazon alone takes roughly 80% of that. It converts well because intent is high, but it's a walled garden: you're renting the retailer's data and you never own the audience. The mistake I see most is a brand shifting its whole budget to Amazon, then wondering six months later why new-customer growth flatlined. Retail media harvests demand — it doesn't create it. The mix that works is retail media on-site for conversion, plus open-web programmatic and CTV off-site to build the demand that shows up on the digital shelf later. We judge both on blended CAC, not platform ROAS. See how we run programmatic →

Digital out-of-home is billboards, mall screens, gym displays and transit boards bought the same way you buy display: through a DSP, on a CPM, with audience and daypart targeting. We run it through Vistar Media inside our 13-DSP stack. Expect $4–$50 CPM depending on the screen — roadside digital boards across the GTA usually land $8–$15, while airport and premium urban panels push $25–$50. The real shift is flexibility. You used to sign four-week static contracts; now you can run a $10K two-week test, trigger creative by weather or daypart, and retarget the mobile devices that actually walked past the screen. That last part is what makes DOOH measurable — we tie exposure to site visits and conversions instead of guessing at impressions. See how we plan it inside a full media mix on our programmatic advertising page.

A PMP is an invite-only auction. A publisher opens a specific slice of inventory to a short list of buyers, you get a deal ID, and you bid on it through your own DSP exactly like open exchange — except you’re competing against a handful of advertisers instead of the entire internet. The premium is real: PMP inventory clears around $12.40 CPM against roughly $5.85 on the open exchange, call it 2x. What you’re buying for that is quality. Viewability runs in the low 90s versus about 71% open, and invalid traffic sits near 1.2% versus 8.7%. Run the math on working impressions, not headline CPM. Paying double for inventory twice as likely to be seen by an actual human is a wash — paying double for premium CTV or a publisher your competitor can’t buy is not. Full side-by-side numbers in our open exchange vs. PMP benchmark breakdown.

For most local service businesses, $1,500–$5,000/month is enough to start gathering meaningful conversion data. For e-commerce or lead-gen brands competing in moderately competitive verticals, $8,000–$25,000/month is realistic to see scalable returns. Anything under $1,000/month per campaign usually doesn't generate enough clicks for Google's algorithm to optimize. The right number is whichever lets you collect 30+ conversions/month per campaign — that's the threshold for Smart Bidding to actually work. More on Google Ads budgeting.

It depends entirely on your margin structure. For e-commerce with 40–50% margins, target a 3.5–5x ROAS minimum. For high-margin SaaS or service businesses, anything above 2x can be profitable. For luxury retail or considered-purchase B2B, ROAS isn't the right metric — track pipeline value instead. The trap most advertisers fall into is chasing ROAS at the expense of growth: a 6x ROAS on $5,000 spend is great, but a 3x ROAS on $50,000 spend often produces more profit. Get a ROAS audit on your account.

Three highest-leverage moves: (1) Fix your conversion tracking first — about 40% of accounts we audit have broken or incomplete conversion data, which sabotages all bidding. (2) Audit search terms weekly and add negative keywords aggressively; most accounts waste 15–25% of spend on irrelevant queries. (3) Restructure into 5–10 tight ad groups by intent stage rather than 50+ keyword-stuffed ad groups — the algorithm performs better with concentrated data signals. Most accounts can drop CPA 20–35% in 60 days with just these three. Book a free Google Ads audit.

Yes, but not as your only campaign. Performance Max works well as a layer on top of strong Search and Shopping campaigns — it fills coverage gaps and finds incremental conversions. It works poorly when used as a primary strategy because you lose granular control over placements and search terms. Best results we see are with this stack: branded Search + non-brand Search + Shopping + Performance Max, with Performance Max getting 30–40% of total Google Ads budget. Always exclude branded terms from PMax. Our full PPC approach.

AI Max is Google’s newest campaign-level setting — it reached general availability in early 2026 and is replacing Dynamic Search Ads (full sunset planned for February 2027). When enabled, AI Max uses Google’s AI to expand keyword matching beyond what you’ve set, generate ad headlines and descriptions from your site content, and dynamically adjust landing page targeting based on user intent. The average uplift Google reports is about 7% more conversions at a similar CPA. The risk: without guardrails, AI Max matches to irrelevant queries, generates copy that doesn’t reflect your brand voice, and sends traffic to suboptimal pages. Every account we manage gets AI Max evaluated against tight negative keywords, brand exclusions, and URL restrictions before we turn it on. It’s a powerful lever used carefully — not a fire-and-forget setting. More on how we manage Google Ads campaigns.

You're not imagining it. Average Search CPC hit $2.96 in Q1 2026, up 12% year over year — the steepest annual jump since 2021. Three things are driving it. AI Overviews cut organic click supply by roughly 15–20%, so demand that used to land on free results now lands in the auction. Performance Max and Demand Gen widened the auction across Shopping, YouTube and Display, so you're bidding against advertisers you never used to see. And Smart Bidding optimizes for conversion value, not cost control — it will happily pay $6 a click if the model likes the conversion. What we actually do about it: brand negatives on PMax, asset groups split by margin instead of product category, portfolio tROAS ceilings, and a weekly search terms audit. On most accounts that claws back 15–25% of wasted spend inside 60 days. See how we manage paid search →

You’re buying views, not clicks, so cost per view is the number that matters. In 2026 skippable in-stream runs $0.02–$0.10 CPV for most advertisers, climbing to $0.10–$0.30 in competitive verticals like legal, finance and B2B SaaS. On a CPM basis it’s roughly $6–$15 for standard video and $10–$25 for non-skippable. The spread by industry is wide: CPG sits near $0.018 CPV while legal runs closer to $0.058 — 3x, on the same platform. Budget-wise, $3,000–$5,000/month is the floor where the algorithm gets enough data to optimize; below that you’re paying for learning you never get to use. Two things move CPV more than bid strategy: a hook that survives the first five seconds, and audience signals built from your own first-party lists rather than Google’s broad affinity segments. We build and run YouTube alongside Search and Demand Gen — see our paid search management approach.

Depends entirely on your industry, but the cross-industry Search average now lands somewhere between 3.5% and 6.6% depending on whose account pool you’re reading, with several 2026 benchmark sets putting the overall figure near 6.6%. It has climbed three years running — responsive search ads and AI-generated assets did that. The spread underneath the average is what actually matters. Arts & entertainment clears 12.75%, finance and insurance 9.83%, travel 9.32%. At the other end, automotive repair sits at 5.56% and dental around 5.66%. Note that the low-CTR verticals are also the expensive ones: weak CTR drags Quality Score, Quality Score pushes CPCs up, and the cycle compounds. Don’t optimise CTR in isolation, though. A high CTR on broad, unqualified queries just buys expensive bounces. Judge it against conversion rate and cost per lead together. Our 10-step Google Ads audit checklist walks through where to look first.

Meta (Facebook + Instagram) is the workhorse for B2C: broad reach, low CPMs ($8–$15), strong conversion tracking. TikTok is essential for brands targeting under-35 audiences and is the cheapest scale platform right now ($5–$11 CPM) but demands a high creative volume. LinkedIn is mandatory for B2B with deal sizes above $10K — high CPMs ($30–$50) but the targeting (job title, company, seniority) is unmatched. Most balanced paid social mixes split 60/25/15 between Meta, TikTok, and LinkedIn for B2C; 20/15/65 for B2B. See our full paid social approach.

The 2026 playbook is different from what most guides still teach. Stop layering audiences and let Advantage+ Shopping or broad targeting run. Fix your Conversions API setup if you haven't — iOS 14.5 broke pixel-only attribution and most accounts still haven't moved to CAPI. Run a creative volume strategy: 8–15 fresh variants per ad set every 2 weeks, not 2 variants every 2 months. And consolidate ad sets — the algorithm needs $50–$100/day per ad set minimum to exit learning. Most accounts drop CPA 25–40% in 6 weeks with this stack. Talk to our paid social team.

TikTok CPMs in 2026 range from $5–$11 for broad targeting, $11–$18 for narrow interest stacks, and $18–$28 for retargeting. Engagement rates are 2–5x higher than Meta on cold traffic when creative matches platform norms (vertical, native, hook in first 1.5 seconds). The trap: cheap CPMs mean nothing if creative refresh cadence is slow — ad fatigue on TikTok hits at roughly 4–6 impressions per user, half what Meta tolerates. Plan for 8–12 creative variants per month minimum. Our TikTok creative approach.

LinkedIn CPMs in 2026 range from $30 (broad B2B targeting) to $80 (senior decision-makers in tech/finance). CPC averages $6–$14. Minimum realistic monthly spend is $5,000 — below that, ad sets can't exit learning. Lead Gen Forms typically deliver MQLs at $50–$200 in mid-market B2B; $250–$700 for enterprise. LinkedIn is expensive but conversion quality is unmatched if your average deal size is above $15K. For deal sizes under $5K, you're usually better on Meta or programmatic ABM. More on LinkedIn for B2B.

Neither platform universally wins — but after managing millions in spend across both, here's how we think about it. Meta's conversion tracking is more mature, its creative requirements are more flexible, and it typically delivers stronger ROAS for brands with established audiences (average CPM $7–$15 vs. TikTok's $3.50–$10). TikTok's lower CPM makes it ideal for top-of-funnel awareness and reaching 18–34 demographics, especially in fashion, beauty, lifestyle, and CPG. Our usual recommendation: start on Meta to build converting campaigns, then layer TikTok in for top-of-funnel expansion once you know your unit economics. Brands running coordinated campaigns on both platforms consistently outperform single-platform strategies by 25–35%. The answer is rarely either/or. Explore our paid social services →

Because the auction got more crowded and your signal probably got worse. Meta CPMs climbed roughly 20% year over year and the blended average now sits near $13.48, with industry ranges from about $2.82 in low-competition verticals up to $42 in finance and insurance. Two things drive most of the pain we see. First, budget is rotating out of Google Search into Meta as AI Overviews eat click volume, so more advertisers bid on the same impressions. Second, creative fatigue: Meta's delivery system quietly charges you more once frequency climbs past 3-4 and CTR decays. Before you blame the platform, confirm your Conversions API is passing clean, deduplicated events. Accounts with degraded pixel signal routinely pay 20-30% more for the same audience. We refresh creative every 2-3 weeks on client accounts for exactly this reason. Here's how we structure it on paid social.

Cheaper than most people expect, which is exactly why it’s getting crowded. In 2026 Reddit CPCs run $0.50–$3.50 with the median around $1.25–$1.85, and CPMs land between $3 and $12 — roughly half what you’ll pay on LinkedIn for a comparable B2B audience. The platform minimum is $5/day, but that’s a floor, not a plan. For conversion campaigns budget $75–$150/day: Reddit needs about 50 conversion events to exit the learning phase, so at a $50 CPA you’re looking at roughly $2,500 in spend before the algorithm is genuinely optimized. Expect 20–40% higher costs in Q4. The catch is creative — Reddit users punish anything that reads like an ad, and a downvoted post gets buried fast. Plain-spoken copy and subreddit-level targeting beat broad interest categories every time. More on how we run it on our paid social page.

GA4 attribution is more complex than Universal Analytics was, but it's also more accurate when set up right. Three musts: (1) Enhanced conversions enabled with first-party data passed via gtag or a server-side tag. (2) Conversions configured as Events with the right counting method (once per session vs every event). (3) Server-side GTM running in parallel to client-side so iOS and ad-blocked traffic gets captured. Most accounts under-report conversions 20–40% because they skip server-side tagging. Get a GA4 audit.

Multi-channel attribution is how you assign credit to each touchpoint in a customer's journey before they convert. If your customer journey spans more than one channel (it does), and you're spending more than $20K/month total (you probably are), then yes — you need it. Last-click attribution dramatically under-credits awareness channels like programmatic, CTV, and paid social, which is why those budgets often get cut prematurely. We typically recommend data-driven attribution in GA4 plus a quarterly media mix model for clients above $50K/month. More on attribution.

For Google Ads with a healthy conversion stream, you should see directional signal in 2–3 weeks and stable performance in 6–8 weeks. For programmatic and CTV, expect 8–12 weeks before optimization compounds — these channels need data volume that takes longer to accumulate. For paid social, 4–6 weeks if creative volume is sufficient; longer if you're refreshing slowly. If you're 12+ weeks in and not seeing clear improvement, something structural is broken (tracking, audience, creative, or bidding) — not a patience problem. Get a diagnostic audit.

Honest answer: it depends on spend level and channel complexity. Below $50K/month total media spend, an agency almost always wins on cost — an experienced senior in-house buyer alone costs $120K+ fully loaded. Above $200K/month, a hybrid model usually wins: in-house strategy + agency execution. Single-channel businesses (only Google Ads, only Meta) can go in-house earlier than multi-channel brands. We've run both models for clients before. More about our agency model.

First-party data is any data you collect directly from your own customers: email lists, CRM records, purchase history, site behaviour, loyalty signups. It’s deterministic, privacy-compliant by design, and compounds in value the longer you collect it. It matters in 2026 because third-party cookie-based targeting continues to erode and platforms like Meta and Google reward advertisers who bring their own data signals. Practically: Customer Match campaigns using a solid first-party list routinely deliver 50–70% higher conversion rates than cold audience targeting. Server-side Conversions APIs — CAPI on Meta, Enhanced Conversions on Google — that pass first-party signals directly to the platforms typically recover 15–25% of conversions that pixel-only setups miss. If you haven’t connected your CRM and consent management to your media activation stack, that’s the highest-leverage fix available in 2026. See how we use first-party data in targeting.

Attribution modeling determines which touchpoints in a customer journey get credit for a conversion. The default — last-click attribution — is free, simple, and almost always wrong. It systematically overvalues brand keywords and retargeting while ignoring the display ad, YouTube view, or TikTok scroll that triggered awareness weeks earlier. Google's data-driven attribution (DDA) is better, but it requires a minimum of 300–500 monthly conversions to function properly. For most of our clients, we run a hybrid approach: data-driven attribution inside Google Ads combined with a third-party cross-channel view. This lets us tell you whether your programmatic spend is genuinely moving the needle — or just claiming credit for conversions that would have happened anyway. See how we approach paid media strategy →

Browser pixels now miss somewhere between 30% and 50% of conversions. Ad blockers, Safari's ITP, consent banners and iOS all strip the signal before it reaches Google or Meta. The fix isn't a better pixel — it's moving measurement server-side. We run server-side Google Tag Manager as the hub: the conversion fires from your server, gets matched to the click using hashed first-party identifiers like email and phone, then fans out to GA4, Google Ads Enhanced Conversions, Meta CAPI and the TikTok Events API at once. A clean Conversions API build typically recovers 40–60% of what the pixel was losing, and most accounts see reported conversion rates improve 15–25% in the first quarter — not because you sold more, but because you're finally counting what you sold. That cleaner data also feeds the bidding algorithms, so performance genuinely improves. See how we approach measurement →

Incrementality testing answers the only question that really matters: would this sale have happened anyway? You hold media back in a matched set of geos or audiences, run it everywhere else, and measure the gap. That gap is your real lift. It matters because platform-reported ROAS misleads in both directions. We've seen retargeting report 6x while delivering almost no incremental revenue, and prospecting report 2x while driving most of the net-new demand. With multi-touch attribution coverage down to roughly 30-60% after third-party cookies, tests are now the ground truth that keeps your attribution model and MMM honest. Practically: hold out at least 10% of the market (20% is cleaner), run 2-4 weeks to cover conversion lag, and discard any result whose confidence interval crosses zero. We run geo tests quarterly on the largest channel in each account. Start with marketing strategy.

Write the answer first, then the article. Answer engines extract the opening one or two sentences under each heading to decide whether you resolve the query, so every section needs a self-contained direct answer with a concrete number in it — not a wind-up paragraph. Four things carry most of the weight. FAQ schema and answer-first formatting score highest in published AEO studies. Statistical density matters, because specific figures get quoted and adjectives don’t. Brand mentions across third-party sites are the single strongest predictor of a ChatGPT citation. And check your robots.txt — blocking AI crawlers in 2026 removes you from the index entirely. Worth knowing: only about 11% of cited domains show up in both ChatGPT and Perplexity, and ChatGPT leans heavily on Bing’s top 10, so your Bing rankings deserve separate attention from your Google ones. We cover the full method in our guide to generative engine optimization.

We're a Toronto-based paid media agency that runs campaigns across programmatic, Google Ads, paid social, CTV, and digital audio for brands in Canada and the United States. We operate inside 13 DSPs and every major paid social platform. Most of our clients are mid-market B2C and B2B brands spending $20K–$500K/month on paid media who want senior strategist attention rather than being handed to a junior account manager. See all our services.

We're integrated with 13 DSPs including The Trade Desk, DV360, Yahoo DSP, Xandr, Amazon Advertising, Vistar Media, Samsung Ads, Adelphic, Spotify Ads, Adobe Advertising Cloud, Infillion, Media.Aisle, and illumin. This gives us access to inventory no single-platform setup can match. See our programmatic capabilities.

Yes — always. We provide real dashboards with actual campaign data, not vanity metrics. You see exactly where every dollar goes, what's performing, and what we're optimizing. No black boxes. Book a free 30-minute audit.

A real one-on-one strategy call with our founder Ryan Roberts. We review your current paid media accounts, identify the two or three biggest leaks in your media mix or attribution setup, and tell you honestly whether your current spend level is sufficient to move the metrics you care about. No sales pitch, no obligation. About a third of brands who book it walk away with a plan they can execute themselves; the rest end up working with us. Book your audit.

Yes. About 60% of our client book is in Toronto and the GTA, with the remainder split across Vancouver, Calgary, Montreal, Ottawa, and major US metros (New York, LA, Chicago, Austin). Paid media is location-agnostic on our end; we work in your time zone and report on your cadence. Meet our founder Ryan Roberts.

Two models dominate. Percentage of ad spend usually lands at 10-20%, with the higher end applied to smaller budgets; the wider market band runs 15-30%. Flat retainers typically run $2,500-$15,000 per month and are more common under about $10K in monthly spend. A useful sanity check: management fees should stay under roughly 20% of spend, and under 15% of the revenue the agency is responsible for driving. If someone wants 25% to run an $8K/month Google Ads account, you're funding their overhead, not strategy. Watch what's excluded, too. Creative production, landing pages, and CTV ad serving are often billed separately. And remember the retainer is not your media budget: if you spend $10K on Google Ads, that $10K goes to Google. We'll tell you what a realistic fee looks like at your spend level on a free 30-minute audit.