Brand vs performance marketing: what the data says, and what 91 industrial websites show
Brand vs performance marketing is the choice of how to split a marketing budget between brand building, which creates future demand by making a business known and remembered, and performance marketing (the research calls it sales activation), which converts demand that already exists into a click, an enquiry or a sale. The modern evidence on that split comes mostly from Les Binet and Peter Field, who analysed the IPA Effectiveness Databank, a store of campaign results entered into the UK advertising industry's effectiveness awards. Their 2013 study, The Long and the Short of It, covered 996 campaigns for 700 brands in 83 categories.[1]
This essay brings together four bodies of evidence. The first is the IPA work on how brand and activation pay back over different time spans. The second is the follow-up manual, Effectiveness in Context, which shows how the right split moves with a brand's situation. The third is the first B2B cut of the same data, published with the LinkedIn B2B Institute, plus John Dawes's arithmetic on how few buyers are in the market at any moment. The fourth is our own scan of 91 US mid-cap manufacturing and supply-chain companies, which shows what industrial websites are set up to do. We close with a framework, the Balance Dial, and a method any team can use to set and test its own split.
Idea in brief
The problem. Performance marketing reports back in days. Brand building pays back over years. So when budgets are tight, brand gets cut, and the dashboard says nothing went wrong.
Why it happens. Most buyers are not shopping at any given moment. If a business changes a key supplier about once every five years, roughly 5% of buyers are in the market in a given quarter.[4] Performance marketing reaches the 5%. Brand building reaches the rest, before they start looking.
The gap (our data). Among 67 industrial homepages we could read, 55% carried at least one ad-platform conversion tag, the plumbing of performance marketing. Only 9% offered a newsletter sign-up, and 45% had any video on the homepage.
The answer. Start from the evidence for your context, then adjust. The IPA data puts the average optimum at 62% brand and 38% activation for consumer brands, and around 46:54 for B2B.[2][3] Entering a new category or market pushes the brand share up, not down.
Key findings
The essay in 3 minutes- Finding 11.5
Brand and activation work best together.
In the IPA data, campaigns that used both brand and activation channels reported 1.5 very large business effects on average, against 1.3 for brand channels alone and 0.7 for activation alone.[1]
- Finding 243%
Brand effects build for years.
Among emotional, brand-led campaigns, 13% reported very large profit effects after one year and 43% after three or more years. For rational campaigns the figures were 10% and 23%.[1]
- Finding 362:38
The right split depends on context.
The all-context optimum in Effectiveness in Context is 62:38 brand to activation. It ranges from 35:65 for brands in their first one to two years after launch to 84:16 for a brand entering a new category.[2]
- Finding 446%
B2B leans slightly more to activation, but not by much.
In fewer than 50 B2B cases, efficiency peaked at around 46% brand and 54% activation. The authors warn the ratio "should not be followed too precisely".[3]
- Finding 51.6
Reach beats loyalty in B2B.
B2B campaigns aimed at loyalty averaged 0.0 very large business effects. Acquisition campaigns averaged 1.0 and broad reach campaigns 1.6.[3]
- Finding 619%
Industrial websites are wired for harvesting.
In our scan, 55% of readable homepages carried an ad-platform conversion tag and 46% carried LinkedIn's Insight Tag. Only 19% put a quote, demo, sales or buy prompt in the homepage text, and 9% offered a newsletter.
Chapter 1 · 1 min
What brand and performance marketing are
Brand marketing aims to make more people know, remember and prefer a business, so they choose it later. Performance marketing aims to get a measurable action now. The two are not rivals. In the words of the B2B report, businesses "need brand activity to create demand (filling the top of the sales funnel), and they need activation to convert that demand efficiently into revenue".[3]
Terms used in this essay
| Term | Meaning |
|---|---|
| Brand building | Marketing that builds memory and preference over time, usually through broad reach and emotional or distinctive creative. |
| Sales activation (performance marketing) | Marketing that seeks an immediate response from people ready to buy: search ads, offers, retargeting, lead forms. |
| Share of voice (SOV) | A brand's share of all advertising spend in its category. |
| Share of market (SOM) | A brand's share of category sales. |
| Extra share of voice (ESOV) | SOV minus SOM. Brands with positive ESOV tend to grow; see the Hive entry on excess share of voice. |
| Very large business effect | The IPA's measure of a campaign reporting a very large improvement in a business outcome such as sales, share, penetration, price or profit. |
| Share of search | Searches for a brand divided by searches for all brands in its category, a cheap brand-health measure introduced by Les Binet.[5] |
Byron Sharp's How Brands Grow gives the shared logic a name: brands grow through mental and physical availability, being easy to think of and easy to buy. Brand work builds the first. Performance marketing mostly cashes it in. The Hive's notes on How Brands Grow cover the argument in more depth.
Chapter 2 · 2 min
The evidence, part 1: two clocks
The Long and the Short of It is a meta-analysis of the IPA Databank: 996 campaigns, 700 brands and 83 categories.[1] Its central claim is that marketing works on two time scales. Activation produces sales spikes that fade fast. Brand building produces smaller effects at first that grow and last. The two are complements, and the strongest results came from campaigns that used both.
Figure 1. Brand and activation together beat either alone
Very large business effects (average number), 0 to 2
Average number of very large business effects reported, by the kind of channels a campaign used. Source: Binet and Field, The Long and the Short of It, IPA, 2013, slide "Integrate brand and activation".
Activation on its own delivered about half the business effects of the combined approach. The same slide shows brand effects following the same pattern (1.6 for combined campaigns, 1.2 for brand only, 0.5 for activation only), and ESOV efficiency doubling from 0.3 to 0.6 when activation was added to brand.[1] The authors' summary on that slide is plain: integration can double efficiency.
The second clock shows up when you track results over time.
Figure 2. Brand effects keep growing after activation effects level off
Campaign length; share of campaigns (%), 0 to 50
Share of campaigns reporting very large profit effects, by how long the campaign ran. Source: Binet and Field, The Long and the Short of It, IPA, 2013, slide "Emotional effects build over years".
After one year, emotional and rational campaigns look almost the same: 13% against 10%. After three years the emotional campaigns are almost twice as likely to report very large profit effects, 43% against 23%.[1] This is the measurement trap in one table. A team that judges every campaign on a one-year window will see little difference and fund whatever reports fastest.
The same deck found that campaigns aimed at the whole market reported 6.7 very large effects on average, against 2.2 for campaigns aimed only at existing customers.[1]
Limits of the IPA evidence
- The cases are entries to an awards scheme, so they lean toward campaigns that worked. The B2B authors say comparisons with other research suggest the bias is modest.[3]
- The cases skew toward large UK advertisers with relatively big budgets.[3]
- These are associations across many campaigns, not controlled experiments.
The Hive entry on The Long and the Short of It walks through the original study in plain language.
Chapter 3 · 1 min
The evidence, part 2: the split moves with context
The 60:40 figure is the most quoted number in this debate, and the least read. In the 2013 deck it appears as a curve: the number of very large effects peaks when activation takes around 40% of the budget.[1] In the later manual, Effectiveness in Context, the authors report that "the all-context '60:40' rule sweet-spot is in fact 62:38 brand:activation", and that the right balance shifts with how hard brand building or activation is in a given situation.[2]
Figure 3. The optimum brand share ranges from 35% to 84%
All campaigns (average): 62%
Brand share of budget (%), 0 to 100
Optimum share of the budget for brand building, the rest going to activation. Source: Binet and Field, Effectiveness in Context, IPA, Table S3 "Brand:activation balance guidelines", IPA Databank 1998 to 2016 for-profit cases.
Three rows matter most for growth-minded firms. A brand entering a new category does best at 84:16, because nobody in that category knows it yet. A small brand does best at 75:25. And a launch in its first one to two years does best at 35:65, leaning on activation while it builds a base, before shifting back toward brand.[2]
The manual also explains why small brands get less room to coast. Big brands can set share of voice a little below market share and still hold steady. Small brands, in the authors' words, "need to keep SOV much more in line with SOM if they are to survive".[2]
A manufacturer that leads its home market is a small brand, or an unknown one, the day it enters a new country or sector. The data says that is the moment to lean harder on brand.
Tip: If you are entering a new market, budget it as a launch into a new category, not as an extension of your home market's mix.
If you want a second pair of eyes on how your marketing reads to a buyer in a market where nobody knows your name yet, the Brave New audit starts there.
Chapter 4 · 2 min
The evidence, part 3: B2B is closer to B2C than most marketers think
In 2019 the LinkedIn B2B Institute published the first B2B cut of the IPA data, written by Binet and Field. The authors call their findings "tentative": the sample was fewer than 50 B2B cases, entered for awards and skewed toward the UK.[3] With that warning, three results stand out.
Figure 4. B2B's optimum split leans only slightly toward activation
Share of budget (%), 0 to 100
Optimum share of the budget for brand building, the rest going to activation. Source: Binet and Field, The 5 Principles of Growth in B2B Marketing, LinkedIn B2B Institute, Figure 6, IPA Databank 1998 to 2018 cases.
B2B efficiency peaked at around 46% brand. That is lower than the B2C average of 62% and below every consumer sector shown, but only five points below other services at 51%. The headline of that section of the report recommends a 50/50 balance for B2B.[3]
The second result concerns share of voice. For consumer brands, 10% extra share of voice was associated with market share growth of 0.6 percentage points a year. For B2B the figure was 0.7.[3] B2B brands responded to share of voice almost exactly as consumer brands did.
Figure 5. In B2B, broad reach beats loyalty
Very large business effects (average number), 0 to 2
Average number of very large business effects by campaign strategy. Source: Binet and Field, The 5 Principles of Growth in B2B Marketing, LinkedIn B2B Institute, Figure 7, 2008 to 2018 B2B IPA cases.
"In fact, loyalty campaigns had a zero success rate on this portmanteau measure," the authors write.[3] Yet in a LinkedIn survey quoted in the report's foreword, over 64% of B2B marketers believed businesses grow by increasing loyalty, and only 4% measured impact beyond six months.[3]
Why most buyers are out of reach of performance ads
John Dawes of the Ehrenberg-Bass Institute put the logic in arithmetic. Corporations change service providers such as their main bank or law firm about once every five years. "That means only 20% of business buyers are 'in the market' over the course of an entire year; something like 5% in a quarter."[4] Dawes is clear that 95% "is not meant to be a precise rule", and he shows how to work out your own figure from your category's buying cycle.
Applying his method to typical industrial buying cycles gives the table below. The arithmetic is ours; the method is his.
| Typical time between supplier changes | Buyers in the market per year (%) | Buyers in the market per quarter (%) |
|---|---|---|
| 1 year | 100 | 25 |
| 2 years | 50 | 12.5 |
| 5 years | 20 | 5 |
| 10 years | 10 | 2.5 |
Calculated as 1 divided by the years between purchases, and a quarter of that per quarter, following Dawes's worked examples. Dawes rounds the two-year quarterly figure to around 13%.
If a plant changes a key supplier every ten years, about one buyer in forty is in the market in a given quarter. The rest decide later, from memory. The Hive entry on the 95-5 rule covers the idea and its limits.
Chapter 5 · 3 min
Original research: what 91 industrial websites are built to do
The research above is mostly about consumer brands. We wanted to see where industrial companies, the sector we work in most, put their effort. A homepage is a fair record: it shows which tracking a company runs, which actions it asks for, and whether it gives a stranger anything to remember.
On 3 November 2026 we scanned the homepages of 91 US mid-cap manufacturing and supply-chain companies: every company in the S&P MidCap 400 and S&P SmallCap 600 classed under eleven manufacturing and freight sub-industries. We also measured each company's share of attention in its sub-industry using Wikipedia pageviews. The full method is at the end of this essay.
Finding 1: the harvesting machinery is in place, the brand layer is thin
67 of the 91 homepages (74%) returned a readable page to our script. 17 (19%) refused it with a 403 error and 7 (8%) returned another response. The figures below are based on the 67.
Figure 6. Industrial homepages are wired for ad tracking more than for memory
Ad tags
Brand signals
Prompts
Share of readable homepages (%), 0 to 100
Share of the 67 readable homepages showing each signal. Ad tags were detected in the homepage HTML and in any public Google Tag Manager container it loaded. Source: Brave New Media scan, 3 November 2026.
37 of 67 homepages (55%) loaded at least one ad-platform conversion tag. LinkedIn's Insight Tag was the most common, on 31 (46%), and Google Ads tags appeared on 23 (34%). Google Analytics 4 was on 55 (82%), but it is an analytics tool, so we did not count it.
The brand layer was patchier. About half linked to stories, insights or case studies (36, 54%) or to a YouTube channel (35, 52%), and 30 (45%) had video on the homepage. Only 6 (9%) offered a newsletter, the cheapest way to stay in front of buyers who are not shopping yet.
One result surprised us. Only 13 of 67 homepages (19%) put a quote, demo, sales or buy prompt in their readable text. So the tracking for performance marketing is widespread, but the homepage itself often does neither job well: it neither asks the in-market buyer to act, nor gives the out-of-market buyer a reason to come back.
Finding 2: one in five runs neither
Figure 7. How the 67 readable homepages combine the two
Share of readable homepages (%), 0 to 100
"Ad tags" means at least one ad-platform conversion tag. "Brand content" means video on the homepage or a link to stories, insights or case studies. Source: Brave New Media scan, 3 November 2026.
32 companies (48%) showed both, 18 (27%) brand content only, 5 (7%) ad tags only, and 12 (18%) neither. A homepage scan cannot show budgets. It does show that nearly half have the machinery for both clocks in place, and about one in five has neither in view.
Finding 3: share of attention is concentrated and slow to move
Share of search, as Binet presented it, uses Google Trends data.[5] For mid-cap industrial names, search volumes are often too low to compare cleanly, so we used a related public signal: views of each company's English Wikipedia article from November 2025 to October 2026, through the Wikimedia Pageviews API. It is a proxy for attention, not a measure of buyers or sales, and much of it will come from investors, job seekers and students.
Figure 8. In most industrial sub-industries, one name takes a large share of attention
Share of sub-industry pageviews (%), 0 to 100
Share of the sub-industry's total English Wikipedia pageviews, November 2025 to October 2026, sub-industries with five or more companies in the sample. "Equal share" is what each company would get if attention were spread evenly. Source: Brave New Media analysis of Wikimedia Pageviews API data, 3 November 2026.
In six of the seven sub-industries, the most-viewed company took between 2.2 and 4 times an equal share; in electrical components it took 1.4 times. In construction machinery and heavy trucks, one company took half of all attention among eight.
The shares also barely moved. Across the 891 month-to-month changes we could measure, the median change in a company's share of attention was 0.48 percentage points. Ranking companies by share in November 2025 and again in October 2026 gave rank correlations of 0.83 to 1.0 in six of seven sub-industries, and 0.5 in electrical components. Attention, like memory, is a slow variable. That is why it is built over years, and why a few months of brand work rarely shows up.
Chapter 6 · 1 min
A framework: the Balance Dial
The research points to a simple order of decisions. Start from the evidence for your type of business, then turn the dial for your situation. The adjustments below are the offsets Binet and Field publish in Effectiveness in Context, Table S3, from consumer cases. Applying them to a B2B starting point is our judgment, so treat the result as a hypothesis to test.
| Step | Question | Adjustment to brand share | Evidence |
|---|---|---|---|
| 1. Start point | B2B or B2C? | Start near 46% (B2B) or 62% (B2C average) | LinkedIn B2B report, Figure 6;[3] Effectiveness in Context[2] |
| 2. Stage | Launch in its first 1 to 2 years? | Down 27 points while launching | Table S3, launch offset[2] |
| 3. Size | Small brand? | Up 13 points | Table S3; small brands must keep SOV in line with SOM[2] |
| 4. Expansion | Entering a new category or market? | Up 22 points | Table S3, entry into new category[2] |
| 5. Channel | Bought online or by subscription? | Up 12 points (offline: down 7) | Table S3, how purchased[2] |
| 6. Check | Is ESOV positive in the market you want to grow in? | Raise reach before raising the ratio | 10% ESOV linked to 0.6 to 0.7 points of share growth a year[3] |
The dial will not give a precise number. It will stop two common errors: running a new market on a home-market mix, and treating brand as whatever is left after performance is funded.
Chapter 7 · 1 min
How to set your brand and performance split
Step 1: Label last quarter's spend
Open last quarter's marketing costs and mark every line "brand" or "activation". Brand reaches people who are not shopping yet. Activation asks for a response now. Write down the brand share as a single percentage.
Step 2: Protect the harvest floor
Fund first the activation that captures people already looking for you: search ads on your name and core service, a clear quote or contact path, and email to existing customers. Only 19% of the homepages we scanned asked the visitor to act in their readable text.
Step 3: Set a starting split with the Balance Dial
Pick your start point, apply the adjustments that fit, and write the result next to your current share. If you are far below it, move in steps.
Step 4: Choose brand work you can sustain for a year
For an industrial company, brand can be a video series on how your products are made, a trade-show presence, a newsletter for engineers and buyers, or broad-reach ads with no form attached. Choose one or two you can keep going, with a consistent look and voice.
This is where we build AI flows. One approved piece, such as an engineer's explanation of a tolerance problem, becomes a newsletter item, three posts, a short video script and a distributor update, adapted for each market and language, with a person checking every version. That is how a lean team goes broad, into more channels and more markets, without hiring for each.
Step 5: Measure brand on a long clock
Judge activation monthly. Judge brand over at least a year, using signals that move slowly:
| Signal | Where to find it | What it tells you |
|---|---|---|
| Branded search | Google Search Console, queries containing your name | How many people look you up directly |
| Share of search | Google Trends, your name against the main names in your category | Your share of category interest |
| "How did you hear about us?" | A free-text field on forms | Which brand efforts buyers remember |
| Price resistance | Sales notes and discount rates | Whether buyers value you enough to pay full price |
Binet tested share of search in three categories, cars, energy and mobile phones, and found it predicted market share, with a lead of up to a year for cars. The IPA's summary adds that it is "not a perfect predictor", because price and other factors affect conversion.[5]
Which to do first
| Action | Effort | Likely impact | When |
|---|---|---|---|
| Label last quarter's spend | Low | High, it sets the baseline | This week |
| Fix the homepage action for in-market buyers | Low | Medium | This month |
| Add a "how did you hear about us?" field | Low | Medium | This month |
| Start one sustained brand channel | Medium | High over a year | This quarter |
| Set up share of search tracking | Low | Medium | This quarter |
| Rebalance the split with the Balance Dial | Medium | High | Next budget cycle |
Chapter 8 · 1 min
What this means for leaders
For a manufacturer or a supply-chain business, the split between brand and performance is really a decision about where next year's customers will come from. Performance marketing is efficient at converting the few buyers in the market today. It cannot reach the buyer who will start a supplier search next spring, and in long-cycle industrial categories that is most buyers.
The evidence is most pointed for growth. A business entering a new country, sector or customer segment starts with almost no share of attention there, and our data shows attention is concentrated and slow to shift. Effectiveness in Context puts the optimum for a brand entering a new category at 84% brand.[2] Expansion plans that fund only lead generation in the new market are asking activation to harvest demand that does not exist yet.
That makes the brand budget part of the growth plan. Our work sits there: helping companies take one strong idea to more markets, channels and buyers, with AI flows that keep a lean team present in all of them. If that is your next move, start with the Brave New audit.
Chapter 9 · 1 min
Common mistakes
- Treating 60:40 as a law. It is an average from award entries. The researchers' own range runs from 35:65 to 84:16.[2]
- Judging brand on a 30-day window. After one year, emotional and rational campaigns looked almost the same; the gap opened in years two and three.[1]
- Spending on loyalty instead of reach. In the B2B data, loyalty campaigns averaged zero very large business effects.[3]
- Tracking without asking. Ad tags on the homepage are no help if the page gives a ready buyer no clear next step.
- Running a new market on a home-market mix. In a new market you are a small or new brand, and the data says to lean toward brand.
- Starting and stopping. Attention shares in our sample barely moved month to month, so short bursts rarely register.
Chapter 10 · 1 min
Methodology
Sample. The 91 companies listed on Wikipedia's S&P MidCap 400 and S&P SmallCap 600 constituent pages under the GICS Industrials sector in eleven manufacturing and freight sub-industries, the same sample as our AI-readiness study. Official websites came from Wikidata (property P856), corrected by hand for five companies.
Homepage scan. On 3 November 2026 a script requested each homepage once, identifying itself with the user agent "BraveNewResearch/1.0" and a link to brave-new.com. It read raw HTML only, did not run JavaScript, and did not retry or get around any block. Where a homepage referenced a Google Tag Manager container, the script fetched that public container file from googletagmanager.com. Ad tags counted were Google Ads (a conversion ID in the form AW- followed by nine to eleven digits), the Meta Pixel, LinkedIn's Insight Tag and Microsoft Ads (UET). Text prompts were matched with fixed phrase lists such as "request a quote", "contact sales" and "where to buy". Brand content meant an embedded video or a link to a blog, insights, stories, case studies, resources or articles section.
Attention. For each company we resolved the Wikipedia title to its current article, then requested monthly English Wikipedia pageviews (all access, human users only) from November 2025 to October 2026 through the Wikimedia Pageviews API. 89 of 91 articles had all twelve months. Shares were computed within each sub-industry; Figure 8 shows sub-industries with five or more companies. Stability used the companies with all twelve months.
Limits. Homepage signals are a one-day snapshot of one page and say nothing about budgets. Sites that render text with JavaScript are undercounted for text prompts, and tags fired by other means are missed. 24 of 91 homepages did not return a readable page. Wikipedia pageviews measure attention from all audiences, not buyers.
The data behind every figure, and the scripts that produced our own, are published with this essay as CSV files.
Chapter 11 · 1 min
Frequently asked questions
What is the difference between brand marketing and performance marketing?
Performance marketing seeks an immediate, measurable action, such as a click, an enquiry or a sale, from people already looking to buy. Brand marketing builds memory and preference among everyone who might buy later. Performance converts existing demand; brand creates the demand that performance converts next year. The IPA data shows campaigns using both outperform either alone.
What is the 60:40 rule in marketing?
The 60:40 rule comes from Les Binet and Peter Field's 2013 IPA study of 996 campaigns, which found business effects peaked when about 40% of the budget went to activation and 60% to brand. Their later work put the average at 62:38 and showed a range from 35:65 for new launches to 84:16 for brands entering a new category.
What is the right brand vs performance split for B2B?
In the first B2B analysis of the IPA data, efficiency peaked at around 46% brand and 54% activation, and the report's headline recommends roughly 50/50. The sample was under 50 cases, and the authors say the ratio should not be followed too precisely. Adjust it for your stage, size and whether you are entering new markets.
How do you measure brand awareness on a small budget?
Track branded searches in Google Search Console, your share of search against rivals in Google Trends, and direct traffic. Add a free-text "how did you hear about us?" field to forms. If your company and its rivals have Wikipedia articles, their pageviews give a free attention proxy. Judge these over a year, not a month.
Your next step
This week, open last quarter's marketing spend, label every line "brand" or "activation", and write down the brand share. Then put your situation through the Balance Dial and write the dial's number next to it. The gap between the two is your plan for next quarter. If you have questions about what to measure, the Brave New Q&A answers the ones we hear most.
If you want a team to build the brand side across every market you sell to, start with the Brave New audit.
Brave New Media is a marketing agency that builds AI flows to help companies scale their ideas, grow their customer base and expand into new markets, with experience across supply chain and manufacturing.
See also
- The Long and the Short of It, Brave New Hive
- Excess share of voice, Brave New Hive
- The 95-5 rule, Brave New Hive
- Mental and physical availability, Brave New Hive
- How Brands Grow, Brave New Hive
- The Brave New Business Journal, weekly thinking on business, money and AI
References
1. Binet, L. and Field, P. "The Long and the Short of It: 10 key principles of success." IPA, 2013. ipa.co.uk
2. Binet, L. and Field, P. "Effectiveness in Context: A manual for brand building." IPA, Marketing Effectiveness in the Digital Era series. Free download via Thinkbox. thinkbox.tv
3. Binet, L. and Field, P. "The 5 Principles of Growth in B2B Marketing: Empirical Observations on B2B Effectiveness." The B2B Institute at LinkedIn, 2019. news.linkedin.com
4. Dawes, J. "Advertising effectiveness and the 95-5 rule: most B2B buyers are not in the market right now." Ehrenberg-Bass Institute for Marketing Science, B2B report, May 2021. marketingscience.info
5. IPA. "Binet presents fast, cheap and predictive Share of Search metric." 14 October 2020. ipa.co.uk
6. Wikimedia Foundation. Pageviews API. wikimedia.org
7. Brave New Media. Brand and performance signals on 91 US mid-cap industrial company homepages, and Wikipedia share of attention, 3 November 2026. Data published with this essay.