How to Build a Paid Media Plan From Your Own Ad Data

A paid media plan is only as good as the numbers it starts from. Too many plans begin with a target, a total budget and a spreadsheet that splits it by habit: a bit more to Meta, the usual share to Google, something left over for a new channel. This guide shows a different route. You build the paid media plan from your own ad data, so every line in it can be traced back to what your accounts have already done, and every assumption is written down where someone can challenge it.

The method works for a DTC brand planning a quarter, an agency planning for a client or a media buyer planning a launch. If you run several platforms, our guide to managing ads across multiple platforms covers the day-to-day side; this article is about the plan that sits above it.

Wevion connects Meta, Google, TikTok, Taboola, Snapchat and Outbrain through their official APIs and syncs them into one view about every 15 minutes, with profit from Shopify or WooCommerce (in beta) alongside spend. That shared baseline is what a data-led media plan needs before any budget is split between channels.

Start with goals and constraints

A paid media plan begins with the business outcome it has to deliver, written as a number and a date, before anyone opens an ad account. "Grow sales" is not a goal. "Reach a set number of first orders in Q1 at a cost per acquisition below a set ceiling" is one, because it tells you what to measure and when to stop.

Write down the constraints next to the goal, because they shape the plan as much as the target does:

• Budget ceiling: the most you can spend in the period, and whether it is fixed or can grow with results.
• Margin floor: the lowest return you can accept once product costs, shipping and fees are counted. A ROAS target means little until you know the break-even point behind it.
• Cash flow: how quickly you need spend to pay back.
• Capacity: how many creatives the team can produce, and how many campaigns it can run well.
• Stock and offers: what you can sell, when, and at what discount.

A plan that ignores capacity is the most common failure here. If the team can produce eight new creatives a month, a plan that needs thirty will fail in week two, whatever the budget says.

Build a baseline from your own data

Your historical performance is the most reliable input a media plan can have, because it reflects your product, your prices and your audience rather than an industry average. Pull at least 90 days of data by platform, and the same period from last year if you sell into a strong season.

For each platform, collect:

Bring the numbers into one view and one currency before comparing them. Platform dashboards report on their own terms, so two "ROAS" figures side by side are rarely measuring the same thing. If you sell on Shopify or WooCommerce, judge each channel on profit as well as revenue, because a channel with a strong ROAS can still lose money on low-margin products.

Read the saturation signs

Saturation is the point where each extra unit of spend buys less than the one before. Look for these signals in your baseline: CPA rising while spend rises, frequency climbing on the same audience, and results that stay flat when budget goes up. A channel showing all three is unlikely to absorb a big increase efficiently, however good its average looks. The cross-channel budget reallocation framework explains why marginal returns, not averages, should drive a budget decision.

Give every channel a role

A channel role is the job a platform does in the funnel, and it decides how you judge it. Without roles, every platform gets measured on last-click CPA, and the channels that create demand always look worse than the ones that capture it.

A simple split works for most accounts:

• Prospecting: reaching people who do not know you yet. Meta, TikTok and Snapchat often play here, as do native platforms such as Taboola and Outbrain for content-led offers.
• Demand capture: reaching people already searching. Google Search usually leads this role.
• Retargeting: bringing back people who visited or engaged. Keep it small and capped, because it can only harvest demand that prospecting created. Our retargeting strategy guide covers audience windows and frequency.

Judge each role on a metric that fits it. Prospecting can carry a higher CPA if it feeds the retargeting and search pool. Retargeting should look efficient, and if it does not, that is a warning. For the most common split decision, the Google vs Meta budget framework walks through the questions in order.

Allocate the budget in scenarios

Budget allocation in a media plan works best as two or three scenarios rather than one fixed number, because the future rarely matches the forecast. Each scenario states the total spend, the split by channel and the result you expect, with its assumptions written beside it.

Here is a hypothetical example for a DTC brand planning one quarter. The figures are illustrative only and do not come from any real account; total spend is shown as an index, with the conservative scenario set to 100:

Agree in advance what moves you from one scenario to the next: for example, two consecutive weeks inside the CPA ceiling before stepping up. That turns "should we spend more?" into a check against a rule everyone signed off. If you want a formula for the split itself, the DTC cross-channel spend allocation formula gives one built on marginal returns.

Ring-fence a testing budget

A testing budget is a fixed share of spend reserved for new channels, audiences, offers and creatives, tracked apart from the core plan. Without it, tests either never happen or quietly eat into the budget that is meant to hit the target.

Set the share at a level you can afford to lose without missing the plan. Give each test a hypothesis, a success metric, a minimum spend and an end date. When a test wins, move it into the core plan in the next review; when it loses, write down what you learnt and close it. The creative testing framework for Meta ads shows how to structure the creative side of this.

Plan flighting and seasonality

Flighting is the pattern of spend over time: steady, pulsed around key dates or weighted towards peak periods. Use last year's data to see when costs rose and when conversion rates improved, and plan spend around both.

For peak periods, the plan should state when prospecting starts building audiences, when budgets step up, when offers begin and end, and when spend comes back down. Auction costs often rise around big retail dates, so audiences built beforehand tend to be cheaper to reach. Our Black Friday and Cyber Monday 2026 paid social checklist has a week-by-week timeline you can drop into the flighting section of your plan.

Agree KPIs and guardrails

KPIs tell you whether the plan is working; guardrails tell you when to act. Agree both before launch, so nobody redefines success halfway through the quarter.

Keep the KPI list short: one primary metric tied to the goal (such as CPA or POAS), a few secondary metrics per channel role, and volume metrics that confirm the numbers are big enough to trust. Then set guardrails as simple if-then statements:

• If a channel's CPA stays above the ceiling for a full week, cut its budget by a set step and review.
• If frequency on a prospecting audience passes your fatigue threshold, refresh the creative before adding budget.
• If a test reaches its minimum spend without a result, stop it.
• If profit per order falls below the margin floor, pause scaling until the cause is clear.

Guardrails can later become automation rules, but write them in plain language first. A rule nobody can explain is hard to trust when it starts moving money.

Set a reporting cadence

A reporting cadence decides who looks at which numbers and how often, and it keeps the plan connected to what the accounts are doing. A simple rhythm works for most teams:

• Daily: a quick check for broken tracking, disapproved ads and runaway spend.
• Weekly: performance against KPIs and guardrails, small budget moves, test results.
• Monthly: progress towards the goal, scenario changes, what to stop and start.
• Quarterly: rebuild the plan from fresh data.

Weekly is where most of the value sits. Moving budget weekly in small, reviewed steps keeps learning phases stable and stops the plan drifting away from reality.

A simple media plan template

A media plan template turns all of the above into one page a team or client can read in two minutes. Copy this structure and fill it with your own numbers; the values in brackets are placeholders.

This is also a workable digital media plan example for a client deck: each row is one slide, and the baseline row is the one clients ask about first.

Common mistakes to avoid

Most weak media plans fail in the same handful of ways, and nearly all of them come from planning without enough data or without written assumptions.

• Planning from benchmarks instead of your own data. Industry averages describe someone else's product and prices.
• Judging every channel on last-click CPA. Prospecting will always lose that contest.
• Ignoring saturation. Doubling spend on a channel rarely doubles results.
• No testing budget. The plan can only repeat what already works, until it stops working.
• Optimising for ROAS when margins vary. Revenue is not profit.
• A plan nobody reviews. If no one owns the weekly check, the plan is a document, not a plan.

Where AI media planning fits

AI media planning can take the slowest parts of the work off your desk: pulling the baseline, researching the market and drafting a first budget split. The judgement on goals, risk and creative still belongs to you.

In Wevion, Strategic Plan is in beta. You describe your goal in plain words and confirm which connected accounts it may read. Wavo, the AI copilot, then reads spend, ROAS, conversions and top campaigns, adds market research with cited sources, recommends a budget from your past spend split by channel with its assumptions stated, and ends with a paid plan covering funnel, campaigns, creative, KPIs and timeline. Each step waits for your approval, and nothing is published to your ad platforms. See how Strategic Plan works, or compare platforms side by side in cross-channel analytics.

You can try it on your own data with a 14-day free trial on the Pro plan, no card needed. Start your free trial.

Key takeaways

• Build the paid media plan from at least 90 days of your own data, not from benchmarks.
• Give each channel a funnel role and judge it on a metric that fits that role.
• Plan the budget in scenarios, with the triggers for moving between them agreed in advance.
• Ring-fence a testing budget and track it apart from the core plan.
• Agree KPIs and guardrails before launch, and review weekly in small, reviewed steps.