Ultimate Guide to Performance Budget Optimization


Are we investing the right amount in the right places to hit our goals?

Every company faces this question when thinking about paid media budgets. This guide will help you answer that question and get the best performance from your digital marketing investment. After reading, you will be able to:

  • Determine the level of investment required to hit your business goals
  • Understand the trade-offs between volume and efficiency
  • Make decisions across marketing channels, not within publisher silos
  • Maximize the return on your marketing investment

Our focus in this document is on digital performance marketing channels, including search (PPC), social, mobile app store, and ecommerce (retail media, shopping) ads.  These ads aim to drive short-term actions such as a website visit, app download, lead form, signup, or purchase.  

For marketers blessed with a blank check to spend as long as they hit their efficiency targets, there is plenty here for you as well.

There are many other levers to optimize your paid digital campaigns, including bidding,  creative, targeting, and more. These are outside the scope of this guide. Here we focus entirely on how much and where to spend. 

Measurement is the Foundation

As the adage goes, “you can’t manage what you can’t measure.” To paraphrase, you can’t maximize the return on your marketing investment unless you measure the results.

 Although measurement is outside the scope of this paper, let’s touch on a few key considerations:

  • Beware of Double Counting: Relying on publisher pixels may lead to multiple publishers reporting ad costs associated with the same Order or Lead.  Enterprise tools, like Google Analytics or Marin Tracker, de-duplicate conversions. Deduplication aligns advertisers’ ROAS metrics with the orders recorded by internal tools. 
  • Attribution: You’re reaching prospects across multiple channels. You need a method to distribute credit. We recommend a single source of truth across all publishers rather than relying on publisher-specific attribution, which is inherently self-serving.
  • Focus on value, not just the conversion: Focus on measuring the actual business value of your conversions. Tracking revenue is better than just a count of conversions, profit is better than revenue, and customer lifetime value helps you maximize the total value of the customer relationship.
“ROAS metrics that better reflect performance lead to superior results because automated tools, based on machine learning, have more accurate input data.”
—Simonas Lisiukas, Director, Customer Intelligence Engineering

The Budgeting Process

Every company has different constraints and considerations when managing budgets. Make sure you understand the rules that dictate your company's process, and think about what you changes you might push to align with this process.

Budget or spend target, e.g. $100,000 for the month.  The marketer (or agency) is tasked with maximizing results while spending to the target. Underspending results in missed opportunities, and over-spending can lead to credits or losing the client.

Efficiency target / unlimited budgets, e.g., Cost per Action, or Revenue over Ad Spend (ROAS).  For some companies, there isn't a traditional “budget” that the marketing team can spend. Instead, they can spend as much as they want if they hit a certain efficiency target. For these, companies maximizing volume to that efficiency target is the primary objective.

Multiple Constraints, We have customers who come to us and say they want to spend $100K for the month but don’t pay over $400 per lead.

First, let’s examine the relationship between Spend and Efficiency. As you spend more, your conversions will generally get more expensive and less efficient.
When you have multiple constraints, the reality is that you will hit one of them first. In this example, the marketer faces the choice of spending no more than $400 per lead or spending less than $100K for the month. They will first hit the $400 CPA and spend less than $100K for the month.

Share of Voice / Impression Share: For some campaigns, you may look to achieve a particular “share of voice” or “impression share.” This objective is common for branded terms in search or if you are trying to make sure you are always present for a given audience.

Planning cycle: What is your budget planning cycle (month, quarter, etc.)? Do you have special promotional periods that overlap with more regular cycles?

Seasonality: Do you have regular patterns that cause spending or efficiency to fluctuate?  Conversion rates may be higher at the end of the month, for example, or volume may change based on holidays.

Agency vs. brand: As a brand, you have more control and freedom to adjust budgets, priorities, and adapting to changes in performance. Ensuring you come in on target is especially important if you’re an agency. Still, you should also stay in close communication with the client so that they can make adjustments and take advantage of new opportunities.

For high-volume agencies that work with many clients (think SMBs, franchises, auto dealers, etc.), a key challenge is efficiently staying on top of a larger number of budgets, accurately hitting the target spend, and providing feedback to the client.

How much should I spend?

Every budget allocation has a defined purpose, and that objective must be measurable.
—Brett Kahnke, Forrester Analyst from our joint webinar on Budget Planning

Clearly state your objective before determining how much to spend. Your program will aim for one or more goals:

Generate Awareness: You want people to hear about your brand. This objective is typically an upper funnel campaign with goals related to reach, frequency, or lift in brand awareness metrics.

Drive Traffic: This objective is about getting clicks to your website as part of mid-funnel campaigns or for sites that monetize traffic.

Promote Engagement: Most relevant for advertising on social networks, this is about getting likes and shares rather than direct engagement on your site or purchases.

Deliver Leads: Your goal is to get the names of new customers, typically for offline sales efforts. Common in B2B, this is relevant for any industry with a longer sales cycle. As cookies disappear, this is an increasingly common strategy for consumer brands who might target users signing up for newsletters. 

Drive Installs: App-oriented businesses focusing on driving downloads of their application.

Generate Sales: The most easily understood of all goals, whether or not the viewer made a purchase. For ecommerce companies, this can happen immediately on the website, but for others, this may need to be an upload of a downstream or offline conversion. The most sophisticated advertisers will look beyond the current purchase, including margin or customer lifetime value.

Managing your budget isn’t just about making sure you spend within limits; it’s about meeting your business goals. For one Marin client with many media sites, their goal is total page views across paid and organic. They needed a system to dynamically shift budget across properties based on the gap between organic performance and overall goals. Marin measured owned and earned traffic relative to the overall target and automatically shifted spend to fill any gaps with the most efficient paid traffic across a broad number of publishers, all while staying in compliance with the total budget. This process repeats for thousands of campaigns daily.

Now, let’s talk about how much to spend. For many companies, there is a top-down approach where finance (or your manager) gives you a number to hit and you must maximize results under that target. Often based on spending a certain percentage of your revenue on marketing, or in bigger organizations, there may be numbers for different geographies, lines of business, or even pet projects. Sometimes allocations are made by channel, and teams must execute to that budget. 

If you’re looking for a benchmark: according to the CMO Survey, companies average 9.5% of revenue on marketing1, with startups generally spending significantly more, up to 20% of revenue.

These figures are for the entire marketing budget, of which paid digital is a subset. However, digital is now capturing the majority of media spend.  According to Business Insider, digital represented nearly 72% of total media spend in 2022, up from 56% in 2019.2

Goal-based budgeting: Another approach is to start with your goals (for example, revenue or conversions) and then determine what you need to spend to achieve those goals based on your expected results. For this to work, you need to understand your customer journey and conversion rates at different stages of that journey.

Imagine a simplified example of a company looking to generate 1000 demo requests. If they know that 15% of people who download an article request a demo, and 5% of people who see an ad download a white paper, they would need to get their message out to 133K people in their audience. At a $7 CPM, this would suggest a budget of $933 for this campaign.

Of course, this is a simplistic example, and we often don’t have a direct connection between all of the campaigns we run and revenue or lead numbers. That is why having a good attribution system is important.

What will I get in return?

Most publishers have tools to help you understand your expected return on a given level of investment. For example, when setting up a campaign on Facebook, you can see the estimated reach and number of clicks for your intended audience. LinkedIn offers similar metrics, including click-through rate (CTR) and cost-per-click (CPC).

Google provides the Performance Planner, which shows the revenue or conversions you will achieve for different levels of investment. 

These tools are helpful but take a siloed view, showing you the results of a campaign or publisher. What you want is an aggregated picture across your program. 

External tools like Marin help you see the relationship between spend and return for your entire program or specific campaigns. This pacing tool also lets you take action, adjusting the campaign budgets and targets to meet your desired level of spend.

So you have a budget, but you think that you should have additional budget. Here are the key points to make when requesting additional funds from your team:

  • Highlight the current spending and performance: Demonstrated results will increase the chances of your request
  • Specifics of the request: Be clear about what you are looking to invest and how it will be used. What do you aim to accomplish (revenue, reach, etc.)?
  • Provide an outline of how the budget will be used
  • Demonstrate you are thinking it through by highlighting what could go wrong 

How should I allocate my budget?

In the previous section, we discussed clarifying your objectives. Does your current budget align with those objectives? If they don’t, that’s a great place to start. We don’t recommend starting from scratch; start with your current program, adjust the allocation, and measure the impact.

Because benchmarks can also be a helpful reference, here’s the average digital budget allocation from our recent survey. 

Different industries and company sizes will stray from these averages; there is not a one-size-fits-all answer, but here are some ways to think your allocation.

Performance-based marketers have seen the publishers grow over the last few years.  As these emerging publishers grow their share of voice and earn a place in your media mix.  The tools you may have become accustomed to with the more established players are unavailable to manage budgets.  

Understanding Marginal Opportunity

The optimal way to allocate your budget is to use the marginal return. In other words, how much will I get in return for the next marketing dollar? We can estimate the trade-off between spend and conversion for the ad auction. The slope of the line is the marginal return and reflects how much more you must spend to get the next conversion. In the example below, increasing spend in the blue auction drives more conversions than the same spend increase in the red auction.

Doing this manually for a mature marketing program spanning many channels and perhaps hundreds or thousands of campaigns is impossible. This process shouldn’t be a quarterly or monthly exercise; it should happen daily. You need an algorithm working for you.

Publisher tools will do this within a campaign or set of campaigns, but if you rely on that, you are missing the opportunity to adjust across channels.

Tools like Marin are the answer.  These tools will estimate the auction landscape of each campaign across publishers and automatically allocate to the best use of that dollar. The result is an adjusted allocation that delivers better results.

Let’s look at a financial services client of Marin’s who sells similar products to different affinity groups. Initially, they were looking at the aggregate performance of Facebook vs. LinkedIn to allocate their budget. However, when they started looking at the marginal opportunity for each product based on its performance, they ended up with a dramatically different allocation by channel. To get the most accurate budget, we mapped the estimated lifetime value of each conversion from their CRM system back to the campaign.


In the webinar we recently did with Forrester, "How to Plan Your 2023 Marketing Budget," they recommended a 70-20-10 rule:

Tried and True
Campaigns and messaging that you know work for a proven audience. This portion is the spend that keeps the lights on and the business moving forward.
New Audiences
In search of growth, we should always look for untapped audiences and markets. Expand your current campaigns to find these overlooked opportunities.
Don’t be afraid to take risks with your messaging or channels. Always be testing. Also, ensure the team knows it's okay for the tests to fail or they won’t take the risk appropriate for this spend category.

Channel Considerations

Here are additional considerations to help map objectives to specific digital marketing channels

Paid Search

Brand: Harvesting demand is a crucial function of paid search, and you want to ensure you are showing up for those brand searches and not allowing the competition to conquest your lower-funnel users. Allocate enough budget to hit your target Impression Share and then use the remaining for upper funnel terms.

Google vs. Microsoft: We believe advertisers should be present on both, but you can also use the marginal opportunity to figure out the exact allocation for your company.

Google vs. Amazon: Similar to the question of Google vs. Bing, the answer here is both.

Additional Marketplaces: As retail media expands, don’t neglect smaller marketplaces where you might find new audiences at a lower cost if the competition is less than you would find on Google and Amazon. If you’re not using them, put them in your experimental budgets.

Social and Display

Impression Caps: Publishers have a default frequency cap because overserving leads to lower conversion rates and poor user experience. However, advertisers may still set more conservative thresholds that reflect their brand strategy, for example, limiting views of Brand Ads to less than 2 per user per 7 days.

Frequency metrics are less important than click-through rate (CTR) and Conversion Rate metrics for budget planning. For example, if an advertiser needs to generate 100 leads and is willing to pay the same cost as a radio ad @ $25 per lead, their budget is $2,500. To hit this goal, CTR and Conv Rate must be steady, whereas ad frequency can fluctuate. 

Frequency vs. CTR: We also recommend analyzing the correlation between frequency metrics and CTR or Conversion Rate to determine if there is a strong relationship and, thus, a reason to cap frequency. 


As we head toward a cookieless world, incrementality testing trumps attribution as the gold standard for understanding the value of a specific marketing channel or campaign. The basic idea is to measure the performance of an audience exposed to the campaign against a control that is as similar in every way possible to the test group. 

One approach is to launch new channels or campaigns in a phased approach. Launch in a set of geographies, and don’t launch in a matched control set. After you have enough data for measurement, you can see the incrementality of that new campaign or channel.

How do I stay on track?

We are big believers in managing by dashboards. You should have a simple view showing what targets you’re trying to hit and whether you are on pace for those targets. Red, yellow, and green indicators can highlight whether you’re on pace, trending fast, or trending slow.

However, if the forecast underlying your dashboard isn’t solid, the dashboard will be of questionable value, so let’s dive into forecasting.


We all know that understanding how your campaigns will perform over the rest of the month isn’t as simple as extrapolating out the performance for the last week. There are several types of seasonality to adjust for when predicting spending.

Day of Week: Almost every business has peaks and valleys throughout the week, but they are different. Every forecast should be aware of day-of-week patterns for your business.

Day of Month: Some businesses have cycles corresponding to specific times of the month, perhaps payday or month-end.

Seasonal Trends: When is your peak or low season? How does that compare to the rest of the year? Is it static every year, or does it expand or contract based on factors like weather or the economy?

Spikes or Events: Some spikes are explainable based on events like sales, new product releases, or other external factors. Others may just be noise in the data. Either way, you need to handle them in your forecast. For specific events like a sale, you should map your promotional calendar to historical spikes to understand the impact of those events. For something publicized, like a sale, there might be an impact before, during, and after the event. You should use outlier detection for one-off spikes so that abnormal volume isn’t factored into the forecast.

Paradigm Shifts: You might experience a structural change to the performance of your campaigns, for example, a competitor leaving or entering a market. If you see a significant volume change, looking at Share of Voice (or Average Position) metrics can help you better understand the change.

A useful dashboard needs to understand how these factors affect your business and accurately forecast your pacing dashboards.

It is important to understand whether changes in your revenue or conversions are volume changes (more clicks and impressions) or conversion rate changes (or possibly both). Volume changes will affect your spending but may not require adjustment to your optimization settings.

Budget Pacing Reports

There are many different ways to build budgeting dashboards. Some publishers offer a version in their platforms but lack a way to see the big picture across publishers. So your options are to build it yourself in Excel or Google Sheets, leverage a BI system, or use a third-party budgeting tool like Marin.

Google Budget Report

Google offers a report for campaigns or Shared Budgets. This report highlights daily spend, cumulative spend (actual and projected), and how your budget settings have changed over the month. It works for individual campaigns as well as shared budgets.

In the example below, the solid gray line is your monthly spend limit, and the blue line shows cumulative spend. The campaign is not projected to reach the monthly spending limit.

However, there are limitations; most notably, it is Google only and doesn’t offer a dashboard view across multiple budgets.

An international software company with publisher accounts for each country was looking to manage PPC budgets by region and product lines. They found that the tools available from the publisher UI were deficient for two primary reasons: they didn’t work across publishers, nor could you add campaigns from separate publisher accounts to a single portfolio. For a budget target for Brand-focused campaigns in EMEA, the available tools would not allow adding Brand campaigns from separate accounts for Germany, Spain, the UK, Africa, and the rest of Europe publisher accounts into a single portfolio.

Marin optimizes across publisher accounts and publishers, giving the company ultimate flexibility on its budgeting needs.

Amazon Budget

Currently in beta, this dashboard shows how your campaigns perform relative to their budget. It estimates the potential impressions, clicks, and sales for each campaign. It also has a recommendation for adjusting your budget to maximize volume.

The Avg. Time in Budget column shows what portion of the day your campaign was on, and Est. Missed Sales shows the revenue you lost if it hadn’t run out of budget.

BI Tools (e.g., Google Data Studio)

Building your dashboard on a BI Platform such as Google Data Studio allows you to bring in additional data sources and report across a broader number of campaigns. Here’s an example of budget pacing built in Google Data Studio.

This type of dashboard is a good start, but it has shortcomings. First, you must get the data into GDS, which is easy for Google Ads but gets harder for non-Google publishers. Third-party solutions like Marin’s BI Connect can help.

The more significant issue is the lack of forecasting. GDS does not understand the seasonality of your business, and with no built-in forecasting tools, it’s hard to predict where spend will be at the end of the planning cycle.

Marin Spend Optimization

Budget pacing is another area where a purpose-built tool is helpful in your budgeting process. Our dashboard addresses the limitations of publisher and BI tools and offers a cross-channel view, sophisticated forecasting, the ability to track multiple budgets, easy-to-read status indicators, and can automatically adjust your campaign budgets and targets to keep you on pace.

Adjusting to Hit Your Goals

Budget Pacing dashboards are great for knowing what’s happening, but you must also hit your targets. You’ve got a few key levers to pull to hit your targets:

Campaign Budget: For most platforms, this will be your primary lever as the platform will try to maximize your revenue or conversion to the budget you provide. 

Campaign Targets: If you don’t use Maximize Conversions/Revenue targets for your campaigns, you may need to change your campaign targets to hit your desired spending goals. Usually, this occurs when your CPA targets are too low (or ROAS targets too high) to allow you to spend your full budget. And if you are managing bids manually, this applies to keyword-level bids instead of campaign targets.

Status If you’re running hot, the only solution to ensure you don’t overspend is to pause your campaigns and restart the next month. Pausing campaigns should be your last resort and only be used if you haven’t successfully managed them throughout your planning cycle. 

New Campaigns/Audiences: If you’re having trouble hitting your budgets, you may need to launch new campaigns or target new audiences to achieve your desired volumes.

Frequency of Changes and Learning Mode 

No one wants to get to the 25th of the month and realize they are pacing way behind budget and then have to scramble to catch up over the last five days. Ideally, you are making small adjustments daily based on the actual performance of the campaigns. 

One thing to be aware of is Learning Mode — a period where Google’s SmartBidding algorithm calibrates itself by exploring various volume buckets and measuring performance impact, which may result in wasted spend. Learning Mode is most commonly triggered when campaigns see a significant change to their SmartBidding targets/ campaign daily budgets. 

The threshold for significance is reported at 20-30% by various sources3,4,5.  On average, it takes 7 days for campaigns to exit Learning Mode. However, depending on campaign volume, it can take between three days and two weeks. Budget-conscious users must ensure changes to targets/budgets are less than 20%.

Changing your Allocation 

One of the big challenges, or opportunities as we like to think, is that many companies don’t change their budget allocation more than once a month or even once a quarter. As the market conditions change and you have a better view of performance, your budget allocation should adjust. In a world where things can go viral overnight, if you can’t jump on trends around your industry and your brand, you’re missing out on huge opportunities, or wasting spend on campaigns that are no longer performing. 

We recommend a tool that’s constantly identifying the biggest opportunities. We’ll talk about how to build that in the automation section below.

How can automation make this easier?

Although the publishers offer great tools, budgeting is too complex to jump between different publishers. They need a unified platform. Furthermore, manually managing your budgets in Excel, Google docs, or via Google Data Studio is missing many advantages of having a system purpose-built for budgeting.

Marin Spend Optimization

The world’s leading marketers are taking advantage of budget management platforms like Marin. These platforms maximize the return on your marketing investment by ensuring optimal allocation and reducing the effort required to stay on target across campaigns, accounts, and channels. 

Marin has workflows to automate the best practices highlighted in this guide, bringing together budget reporting and activation, including: 

What-if Analysis: Determines the right level of spend across the entire account for an unlimited number of Strategies and the impacts on efficiency and volume, should you increase or decrease your budget. 

Dynamic Budget Allocation: Automatically allocates budget based on marginal opportunity and continually readjusts as performance changes to maximize your goals and ensure each dollar gains the highest return.

Budget Management at Scale: The automation in Marin enables account managers at agencies to handle more clients than ever without the risk of overspending. Automated reporting and alerts eliminate manual tracking. 

Pacing Dashboard: Marin’s visual dashboards give advertisers an at-a-glance summary of campaign performance for an instant view of which programs are on track and any that may need attention.

Automated Pacing: Marin continually monitors campaign spending and  automatically adjusts budgets to ensure you hit your plan without worrying about over or underspending.

Sophisticated Forecasting: Advanced machine learning, publisher models, and auction simulations to handle every aspect of seasonality, including day-of-week, month, and year patterns.

Shared Budgets: Allow advertisers to map any number of campaigns from multiple Google and Bing accounts to a single strategy with a share spend goal for ultimate flexibility regardless of account structure or complexity. 

Flexibility: Marin meets the needs of any business with custom planning cycles, rules for pausing and resuming budgets, rollover spend options and mid-period budget updates.  

Revenue Integrations: Make budgeting decisions based on the full business value of your conversions, not just the initial touch point with 3rd party revenue, offline data upload, and CRM integrations.

Ready to learn more, check out a product tour of all Marin has to offer.

Is your PPC Budget working for you?

Maximize the return of every marketing dollar with Marin.