Marketing audit
B2B marketing audit, the whole framework
A B2B marketing audit finds the gap between what you spend and what you get, and says what to fix first. The framework is on this page, questions included, because it is not secret knowledge and gating it behind a form would be slightly absurd.
What is a B2B marketing audit?
A B2B marketing audit is a structured review of whether your marketing is pointed at the right buyer, whether the funnel leaks, and whether the numbers you are making decisions from are true. It produces a prioritised list of what to fix, not a report card.
The useful version answers one question: where is the biggest gap between what you are spending and what you are getting, and what is the cheapest thing that closes it. Everything else in an audit is in service of that.
This page contains the actual framework we use, including the questions, because a marketing audit is not secret knowledge and gating it behind a form would be slightly absurd. If you work through it yourself and find your own problems, that is a good outcome. If you get halfway and realise the data needed to answer half the questions does not exist, that is also a finding, and it is the most common one.
When should you run one?
When something has gone wrong twice, when you are about to spend significantly more, or when somebody new is arriving who needs a baseline. Running one because it is the start of the year is how audits become documents nobody acts on.
Pipeline has missed target for two consecutive quarters
One quarter is noise. Two is a pattern, and by the third the cause is usually several months old and harder to isolate.
Cost of acquisition rose and nobody can explain why
An unexplained rise is almost always a mix shift rather than a channel getting worse. Finding which segment moved is an afternoon's work once the data is joined.
You are about to increase spend significantly
The worst time to discover broken tracking is after you have tripled a budget on the strength of numbers it produced.
A new marketing leader is arriving
An audit gives them a baseline they did not write, which makes the first ninety days about decisions rather than about discovery.
Sales and marketing disagree about lead quality
This argument is never settled by opinion. It is settled by looking at what happened to the last two hundred leads.
You are raising, or being diligenced
Investors will ask for CAC payback by channel and cohort retention. Finding out you cannot produce those during diligence is an avoidable kind of painful.
One timing note worth taking seriously. Do not run an audit in the same month you are launching something significant. The team will be stretched, the data will be unrepresentative, and the recommendations will arrive when nobody has capacity to act on them. Audits produce value when there is room to change something afterwards.
What does an audit actually cover?
Six areas, in a deliberate order: positioning and ideal customer profile, pipeline and funnel, attribution and data, channels and spend, content and assets, then team and process. The order matters because findings in the earlier areas change how you interpret the later ones.
Starting with channels is the common mistake and it produces confident wrong answers. If the ICP is wrong, every channel looks underperforming. If attribution is broken, every channel comparison is fiction. Positioning first, data second, tactics third.
The sections below contain the actual questions. Work through them in order. Anywhere you cannot answer, write down that you cannot, because the gaps are usually more informative than the answers.
Area one: positioning and ideal customer profile
Ask whether three people would describe the buyer the same way, and whether the deals you actually close match the deals you target. Divergence between the stated ICP and the closed won list is the single most common strategic finding, and it invalidates everything downstream.
- Can three different people in sales each explain who you serve and why you are the obvious choice, in thirty seconds, without a deck?
- Can you name the companies you are explicitly not for?
- Can you name what your closest competitor genuinely does better than you?
- Do your last twenty closed won deals match the ideal customer profile your campaigns target?
- Do your last twenty losses share anything in common that the ICP would have predicted?
- Does the language on your homepage match the language buyers use on recorded sales calls?
- If a buyer reads only your homepage, can they tell what category you are in?
The thirty second test is worth doing literally. Ask three people in sales, separately, who you serve and why you are the obvious choice. If you get three answers, positioning is the project and no channel work will compensate.
The wins and losses comparison is the other high value exercise here. Take the last twenty closed won deals and the last twenty losses, and compare them against who your campaigns target. Companies are frequently excellent at selling to a segment they are not marketing to, which is a cheap problem to fix once seen.
Area two: pipeline and funnel
Map the conversion rate at every stage and find the two that leak most. Most companies know their overall conversion and not their stage conversion, which means they are improving the wrong step. The largest leak is frequently between lead arriving and anybody contacting it.
- What is the conversion rate at every stage, from visitor to lead to qualified to opportunity to closed won?
- Which two stages leak the most, and has anyone looked at why rather than assuming?
- How long does each stage take, and which one has got slower in the last six months?
- What percentage of leads are ever contacted, and how quickly?
- What happens to a lead that is not ready to buy, and is anybody measuring whether that works?
- Of the deals lost, how many were lost to a competitor, to no decision, and to budget?
Two things to look at that teams routinely skip. Stage duration, because a stage that has got slower is usually a better signal than one that has got leakier, and it is easier to diagnose. And the fate of leads that are not ready, because in a market with a long buying cycle those are a significant share of eventual revenue and they are frequently abandoned.
On lost deals, the three way split between lost to a competitor, lost to no decision, and lost to budget matters enormously. A loss to no decision is a marketing problem about urgency and problem framing. A loss to a competitor is a positioning or product problem. Treating them the same is how companies spend a year solving the wrong one.
Area three: attribution and data
Check whether closed revenue can be traced back to the activity that produced it, and whether your systems agree with each other. Fix tracking before analysing anything, because broken UTMs and unjoined systems do not produce slightly wrong channel conclusions, they produce confidently wrong ones.
- Can you trace a closed won deal back to the marketing activity that produced it?
- Are your CRM, your analytics and your ad platforms joined, or are they three separate opinions?
- Do your UTMs follow a convention that anybody has written down?
- Does your demand form ask how the buyer heard about you, and does anyone read the answers?
- What attribution model are you using, and does the person reading the report know that?
- If you switched from last touch to first touch tomorrow, which channel's budget would you change?
The question that reveals the most is the last one: if you switched from last touch to first touch tomorrow, whose budget would change. If nobody knows, the reporting is not being used to make decisions. If somebody knows immediately and the answer is substantial, you have found a channel that is either being starved or being overfunded.
The self reported field deserves defending. A single question on the demand form asking how the buyer heard about you is imprecise, unfashionable and frequently more accurate than the analytics stack, because it captures the recommendation in a private Slack group that no tool can see. The failure is rarely that companies do not have the field. It is that nobody reads it.
Area four: channels and spend
Work out cost per qualified opportunity by channel, not cost per lead, and check how concentrated your pipeline is. A channel producing most of your pipeline is a strength and a risk at the same time, and the risk is usually unmanaged because the channel is working.
- Which channel produced your best twenty customers, and what is the evidence rather than the assumption?
- What is cost per qualified opportunity by channel, not cost per lead?
- Does any single channel produce more than about sixty percent of pipeline, and what is the plan if it stops working?
- Which channel has been running longest without anyone asking whether it still works?
- What are you spending on tools, and which paid seats and add ons are genuinely used?
- What did you stop doing in the last year, and did anything get worse as a result?
Concentration is the finding people resist most. When one channel produces the majority of pipeline, the honest reading is that you have one channel and a hobby. That is survivable and common, and it only becomes a crisis when the channel changes, which is why the question is not whether to diversify today but whether anybody has a plan.
The spend review is the least glamorous part and it frequently pays for the engagement. Overlapping tools, seats for people who left, an add on bought for a campaign that ended two years ago. We find this on most audits and it is usually enough to fund the fixes.
Area five: content and assets
Separate the pages that produce pipeline from the pages that produce traffic, and be honest about how much content is aimed at someone who will never buy. Most B2B content programmes have a small number of pages doing the work and a large number of pages doing nothing.
- Which pages produce pipeline, and which produce only traffic?
- How much of your content is aimed at people who will never buy?
- Do you have comparison and alternatives pages, and if not, who is writing them about you?
- When was anything on the site last updated, and does it describe your current pricing and integrations?
- Can a buyer get the answer to their question without filling in a form?
- Is there a single free tool or asset that earns links and captures intent, or only articles?
A content audit for SaaS usually finds the same shape: a handful of bottom of funnel pages producing nearly all the pipeline, and a large body of top of funnel content that attracts readers who are not buyers. The instinct is to produce more. The better move is usually to update the pages already working, because refreshing a page that converts is cheaper and faster than creating one that might.
The comparison question matters more than it looks. If you have no comparison or alternatives content, somebody else is writing it about you, and their framing becomes the default answer a buyer finds. These pages are uncomfortable to write and they convert better than almost anything else on a B2B software site.
On free tools: a calculator, an assessment or a scorecard will out-earn a year of articles, because it ranks, earns links and captures intent that content cannot. Almost nobody builds them, because they are software rather than writing. We built five on this site for exactly that reason.
Area six: team, process and tooling
Ask who owns each part of the funnel and what happens when a lead arrives at four in the afternoon. Process findings are less exciting than strategy findings and they are frequently where the recoverable money is, because they are cheap and immediate to fix.
Four things worth checking.
Ownership of the handoff. Who is responsible for a lead between arriving and being contacted, and what is the agreed time. If the answer is ambiguous, this is almost certainly your largest leak.
What the team spends time on. Not what the plan says. A day of honest accounting usually reveals a large share going to reporting, internal updates and tool wrangling rather than to work that reaches a customer.
Where decisions get made. If every channel decision needs the founder and the founder is in sales calls, marketing moves at the speed of their calendar, which no process fixes.
Whether anybody talks to customers. The strongest predictor of whether marketing works that we have found. Teams that listen to sales calls write different copy from teams that do not, and the difference is visible on the page.
How do you prioritise what the audit finds?
Score every finding on impact, confidence and effort, then do the high impact, high confidence, low effort items first regardless of how uninteresting they are. An audit that produces twenty recommendations and no order is a document rather than a plan.
The scoring is simple and the discipline is in applying it honestly. Impact is how much pipeline or cost it moves. Confidence is how sure you are, based on evidence rather than conviction. Effort is calendar time rather than difficulty, because a two hour fix that waits three weeks for a decision costs three weeks.
Two rules that make the output usable. Nothing goes on the list without a named owner, because unowned recommendations are a to do list for nobody. And the list is capped, usually at five or six items for a quarter, because an audit that recommends twenty changes guarantees none of them get done properly.
The most common sequencing error is starting with the most interesting finding rather than the most tractable one. Positioning work is intellectually satisfying and takes a quarter. Fixing the lead handoff takes a week and frequently moves more revenue. Do the second one while doing the first.
What do you usually find?
Six things, with striking regularity. Broken attribution that nobody mentioned, an ICP that no longer matches the customer base, one channel carrying everything, content volume with no pipeline contribution, an unowned handoff to sales, and tooling spend that nobody has reviewed.
Attribution is broken and nobody said so
By far the most common finding. Not a little broken: unjoined systems, inconsistent UTMs, and a channel report that cannot survive a follow up question. Everything downstream is affected and almost nobody raises it unprompted.
The ICP on the slide is not the ICP in the CRM
The stated target and the actual customer base diverge, usually because the company grew into a segment nobody updated the strategy for. Campaigns then target the old answer.
One channel is carrying everything
Frequently inbound from a founder's network or a single partnership. It works, it is unmanaged, and nobody has a plan for the quarter it stops.
Content volume is high and pipeline contribution is near zero
Published consistently for a year, aimed at a reader who was never going to buy. Expensive, visible, and defended because stopping feels like going backwards.
Nobody owns the handoff to sales
Leads arrive, sit, and age. The conversion rate from lead to contacted is often the single largest leak and the cheapest to fix.
Tooling costs more than anyone realises
Overlapping subscriptions, seats for people who left, an add on bought for a campaign that ended. This usually funds the fixes the audit recommends.
Worth saying plainly: none of these indicate an incompetent team. They are what happens when a company grows faster than its processes, which is the normal condition of a business that is working. The value of an audit is that an outsider will say the thing everybody internally has half noticed and nobody has had cause to raise.
What does a good audit output look like?
Short. Five or six prioritised findings, each with the evidence behind it, an owner and a rough effort estimate, plus a separate list of what you cannot currently measure. If the document runs to forty pages it was written to demonstrate effort rather than to be acted on.
Here is the structure we use, and the one we would suggest if you run this internally.
One page of what is working. Genuinely, with evidence. Audits that open with problems get read defensively, and the things working are also the things most at risk of being changed accidentally.
Five or six findings, ranked. Each one stating the problem, the evidence, the estimated impact and the effort. Ranked by impact divided by effort rather than by how serious they sound.
An owner for each. A name, not a department. Unowned recommendations do not happen, and the act of assigning them surfaces capacity problems immediately.
A list of what you cannot measure. Separate from the findings, because it is a different kind of problem and it usually blocks several of them. This list tends to be the most uncomfortable page and the most useful one.
The underlying analysis, available but not in the document. Spreadsheets, cohort tables, channel breakdowns. Anyone who wants to check the work should be able to, and nobody should have to read it to get the point.
What we deliberately leave out: benchmarks against industry averages, maturity scores out of ten, and anything resembling a grade. They feel rigorous and they drive no decisions, because knowing you are a six out of ten on content maturity tells you nothing about what to do on Monday.
What should you fix in the first thirty days?
The measurement, and the lead handoff. Both are cheap, both are usually broken, and both block everything else. Positioning work is more interesting and takes a quarter, so start it in parallel rather than first.
A sensible first month after an audit, in order.
Week one: fix tracking. UTM convention written down and applied, the self reported source field added to the demand form, and the CRM joined to analytics well enough to ask what a customer cost rather than what a lead cost. Until this is done every channel decision is a guess with a chart attached.
Week one: fix the handoff. Agree who contacts a new lead and within what time, then measure whether it happens. This is frequently the largest single leak in a B2B funnel and the cheapest thing on the list to improve.
Week two: stop something. Every audit finds at least one thing running on inertia. Stopping it frees budget and attention, and it proves the audit produces decisions rather than additions. If nothing gets stopped, the exercise was an expansion plan wearing an audit's clothes.
Weeks two to four: start the positioning work. Customer conversations, win and loss review, the thirty second test with sales. It will not resolve inside the month and it should be underway, because everything downstream waits on it.
What we would not do in the first thirty days: launch a new channel, rebuild the website, or commission a content calendar. All three are reasonable later and all three are expensive guesses before the measurement works.
How do you audit when you barely have any data?
Interview instead of analyse. With fewer than about fifty customers, statistical analysis is not available to you anyway, so the audit becomes qualitative: talk to every customer you can, read every lost deal, and treat patterns across twelve conversations as the evidence base.
Early stage companies frequently assume an audit is not for them because the dashboards are empty. The opposite is closer to true. Small numbers make the qualitative work both necessary and tractable, and the findings at this stage are usually larger and cheaper to act on.
What replaces the data.
Every customer, not a sample. If you have thirty customers, speak to as many as will take the call. At this scale a sample is pointless and the full set is achievable.
Lost deals, specifically. The most informative conversations you can have and the least often requested. People will tell you what your competitor said, which you cannot learn any other way.
The words, written down verbatim. How buyers describe the problem, in their phrasing. At this stage your positioning work is mostly a transcription exercise, and the gap between their words and your homepage is the finding.
What the founder actually did to get the first customers. Usually undocumented and usually the only channel that has demonstrably worked. Writing it down turns a personal network into something repeatable, or proves it is not repeatable, which is equally useful.
One caution specific to this stage. An audit cannot tell you whether you have product market fit, and a company without it will interpret every finding as a marketing problem. If customers do not come back, fix that first. Marketing applied to a product people leave makes the churn more efficient.
Should you run it yourself or hire someone?
Run it yourself first. The questions on this page are the whole framework and working through them costs you a week. Hire someone when you get stuck on the data, when you need an outside voice to say something internally unsayable, or when nobody has the week.
Three honest arguments for doing it internally. You know the business, you have access to everyone, and the act of answering the questions produces most of the value regardless of who writes the report. We would rather you did this and came back for the fixing than paid us to tell you things you could have found.
Three arguments for an outside pair of eyes. Internal audits tend to avoid conclusions that implicate the person conducting them, which is human rather than dishonest. Somebody who has seen thirty funnels recognises patterns faster than somebody who has seen one. And an external finding is easier to act on politically, which matters more than it should.
If you want a starting point that takes ten minutes rather than a week, the free go-to-market scorecard on our homepage covers a subset of these questions and gives you a rough read on where the gaps are. It is free, it requires no call, and it will not tell you anything this page does not.
How often should you repeat it?
Once a year for the full six areas, and a short review of the numbers every quarter. Auditing more often than annually produces diminishing returns, because the strategic findings take two or three quarters to resolve and re-examining them monthly just restates problems you already know about.
The quarterly version is deliberately narrow. Cost per qualified opportunity by channel, payback by segment, the two leakiest funnel stages, and whether the things agreed last quarter actually happened. That last question is the one most often skipped and the one that determines whether any of this was worth doing.
The annual version is the full framework again, and it should be run by somebody who did not run the previous one where possible. An audit repeated by the same person tends to confirm the previous conclusions, which is human and not useful.
One trigger that overrides the calendar: a significant change in the business. A new market, a new pricing model, an acquisition, or a change in who buys. Any of those invalidates the positioning work, and a strategy built for the previous answer will quietly stop working in a way that takes two quarters to show up in revenue.
How do we run a B2B marketing audit?
Two weeks, fixed price, and it ends with a prioritised list of five or six things with owners attached. We look at the same six areas described above, speak to sales as well as marketing, and tell you if the honest answer is that your marketing is fine and the problem is elsewhere.
- 01
Data and access
Days 1 to 2
CRM export, analytics, ad accounts, and the last twenty wins and losses. Where data is missing we note it as a finding rather than working around it quietly.
- 02
Conversations
Days 3 to 6
Two or three people in sales, whoever owns marketing, and recorded customer calls. Sales interviews surface more in an hour than a week of dashboard analysis.
- 03
Analysis
Days 7 to 10
The six areas, worked through in order, with everything we cannot verify flagged rather than assumed.
- 04
Findings and a plan
Days 11 to 14
A prioritised list scored on impact, confidence and effort, capped at five or six items, each with a named owner. Delivered in a working session rather than emailed as a deck.
What you get is short on purpose. A forty page audit is a way of demonstrating effort, and nobody acts on it. Ours runs to a handful of pages plus the underlying analysis if you want it.
What it costs: a fixed 2,500 to 4,000 dollars depending on how many channels and systems are involved, agreed before we start. If we conclude that your marketing is not the problem, we will say so and the audit has still done its job, which is why it is priced as a defined piece of work rather than as the first step of something larger.
If the findings need building afterwards, that is quoted separately and you are under no obligation to use us. B2B SaaS marketing covers the execution side and fractional CMO the leadership side, and plenty of clients take the audit and do the work themselves.
FAQ
Common questions
What is a B2B marketing audit?
A B2B marketing audit is a structured review of whether marketing is pointed at the right buyer, whether the funnel leaks, and whether the numbers behind your decisions are true. The output should be a prioritised list of what to fix with owners attached, not a scored report card.
What questions does a marketing audit ask?
Six areas worth of them, in order: positioning and ICP, pipeline and funnel, attribution and data, channels and spend, content and assets, then team and process. The full question list is published on this page rather than gated, because it is a framework rather than secret knowledge.
How long does a marketing audit take?
Two weeks for a thorough external one, or about a week of your own time if you work through the questions internally. Anything sold as a same day audit is a tool report, and anything taking two months will arrive after the situation that prompted it has changed.
What does a B2B marketing audit cost?
We charge a fixed 2,500 to 4,000 dollars depending on how many channels and systems are involved, agreed before we start. Doing it yourself with the framework on this page costs a week of attention and will find most of the same things.
When should we run one?
When pipeline has missed target for two consecutive quarters, when acquisition cost has risen without explanation, before a significant increase in spend, when a new marketing leader arrives, or when sales and marketing disagree about lead quality. Running one because it is January is how audits become documents nobody acts on.
What is the most common finding?
Broken attribution that nobody raised. Not slightly broken: unjoined systems, inconsistent UTMs, and channel reporting that cannot survive a follow up question. It matters because every downstream conclusion about which channel works inherits the error.
Can we do a content audit on its own?
Yes, and for SaaS companies it is a reasonable standalone piece. The shape is almost always the same: a small number of bottom of funnel pages producing nearly all the pipeline, and a large body of content attracting readers who will never buy. The usual recommendation is to update what works rather than publish more.
We are pre revenue. Is an audit useful yet?
Partly. With almost no customers the statistical work is unavailable, so the audit becomes qualitative: talk to everyone who bought, read every lost deal, and write down the words buyers use verbatim. What an audit cannot tell you at that stage is whether you have product market fit, and a company without it will read every finding as a marketing problem.
Do you do the work you recommend?
Only if you want us to, and it is quoted separately. The audit is priced as a defined piece of work rather than as a qualifying call, and plenty of clients take the findings and execute them with their own team. If we conclude your marketing is not the problem, we will tell you that too.
Keep exploring
Related
- Free go-to-market scorecardTen minutes, no call, covers a subset of this
- Fractional CMOWhen the findings need a leader
- B2B SaaS marketingWhen the findings need executing
- Go-to-market strategyWhen the direction itself is the gap
- Brew, free enrichmentBuild the outbound list without credits
- About VoliomWho does the work
Let's talk
Tell us what you're building.
Get a free 30-minute strategy call. You'll leave with a clear plan, a realistic estimate and our honest opinion, whether or not you work with us.
- Email usmail@voliom.com