# Assumption Register & Evidence Ledger: How Fractional CMOs Track What They Know and What They're Betting On | Crank Source: https://wearecrank.com/fractional-cmo-command-centre/assumption-evidence-ledger Learn how fractional CMOs use assumption registers and evidence ledgers to track marketing beliefs, validate strategy, and make evidence-based decisions. --- Fractional CMO # Assumption Register & Evidence Ledger **How Fractional CMOs Track What They Know.** Most marketing strategies are built on untested assumptions. Without a structured way to track and validate them, teams repeat the same mistakes and misread what's actually driving results. [Talk to WeareCrank ](/contact) On this pageContents 1. [Marketing Strategy Is Full of Assumptions — Most Are Never Tested](#marketing-strategy-is-full-of-assumptions) 2. [What an Assumption Register Is and Why It Matters](#what-an-assumption-register-is) 3. [The Evidence Ledger: Recording What You've Learned](#the-evidence-ledger) 4. [Building and Maintaining the Register in Practice](#building-and-maintaining) 5. [How the Ledger Feeds Into Prioritisation Decisions](#how-the-ledger-feeds-prioritisation) 6. [High-Risk Assumptions: Which Ones to Test First](#high-risk-assumptions) 7. [Stop Flying Blind on Marketing Strategy](#stop-flying-blind) TL;DR Most marketing strategies are built on untested assumptions, and without a structured way to track and validate them, teams repeat the same mistakes and misread what's actually driving results. * Marketing strategies contain far more assumptions than most teams acknowledge. * An assumption register gives you a structured place to capture, prioritise, and test those assumptions. * An evidence ledger records what you've learned from testing, so findings aren't lost or ignored. * Together, these tools make SEO and marketing decisions more defensible and repeatable. * This article covers how to build and use both in a practical marketing context. ## Marketing Strategy Is Full of Assumptions — Most Are Never Tested ![Marketing Strategy Is Full of Assumptions — Most Are Never Tested](/images/fcmo/fractional-cmo-command-centre--assumption-evidence-ledger/01.png) Every marketing strategy is built on beliefs. That a specific audience has a specific problem. That your content will rank for terms your customers actually search. That ranking will bring qualified traffic — and that traffic will convert. None of those are facts. They're assumptions. Most teams treat them like they're the same thing. That gap is where marketing spend quietly disappears. Teams build content plans, paid campaigns, and SEO roadmaps on things that sound reasonable but have never been tested. When results fall short, the post-mortem doesn't identify which assumption failed. The work gets written off as "not working," and the cycle starts again with a fresh strategy that carries most of the same beliefs. We see this constantly. The problem isn't bad planning — it's that the assumptions were never written down, never prioritised, and never assigned to anyone to test. They existed as shared understanding. Which means they existed as shared risk. Capturing assumptions formally, and recording what evidence you find for or against them, is what separates strategy that actually improves from strategy that just gets refreshed every quarter. An **assumption register** gives you the first half of that system. An **evidence ledger** gives you the second. So where do assumptions actually live in most marketing teams? Strategy decks. Scattered briefs. Someone's head. When that person leaves, the context goes with them. When someone questions a decision six months later, there's no record of why it was made or what it was supposed to prove. That's a documentation problem, not a strategy problem. And it's fixable. An assumption register treats assumptions as real objects in your planning process — things with owners, risk ratings, and a status you actually update. An evidence ledger complements it by giving those assumptions somewhere to land once you've run a test, pulled data, or gathered customer feedback. Used together, they make your marketing strategy auditable. Not in a box-ticking sense — in a practical one. You can look back at what you believed, what you found, and make your next move based on something more solid than instinct. ## What an Assumption Register Is and Why It Matters ![What an Assumption Register Is and Why It Matters](/images/fcmo/fractional-cmo-command-centre--assumption-evidence-ledger/02.png) Every marketing strategy runs on beliefs. Beliefs about your audience, your channels, your competitive position, what the budget will actually return. The problem — as we've covered — is that most of those beliefs are never written down. Never tested. Never revisited. An assumption register changes that. Assumption Register An assumption register is a structured document that records every unverified belief underpinning a strategy, along with the evidence needed to confirm or disprove it. Think of it as a living log. You capture the assumption, note how confident you are in it, identify what would validate it, and track what you actually find. It sits alongside your strategy as a companion document — not buried in a slide deck, but actively maintained as decisions get made and results come in. The companion to the assumption register is the evidence ledger. Where the register asks "what are we assuming?", the evidence ledger answers "what do we actually know, and how do we know it?" Together, they give your team a clear picture of where your strategy stands on solid ground — and where it's standing on guesswork. #### Assumption Register in Practice A B2B SaaS company is planning a content-led SEO campaign. Their assumption register might include: 'Our target buyers search for problem-aware terms before vendor-aware terms.' The evidence ledger would then record what keyword research, customer interviews, or CRM data either supports or contradicts that belief. If the evidence is thin, that assumption moves up the priority list for testing before significant budget is committed. Marketing teams make expensive decisions based on assumptions that were formed months or years ago, often under completely different market conditions. A launch strategy built on the belief that a particular segment is price-sensitive can be badly misaligned if that sensitivity has shifted. Without a register, nobody catches it. The strategy just quietly underperforms. The register also creates accountability. When an assumption is named and owned by a specific person, it becomes their job to gather the evidence. Vague strategic beliefs turn into concrete research tasks. And retrospectives get more honest — instead of asking "why did this campaign underperform?", you can go back to the register and ask "which assumption turned out to be wrong?" We see this constantly during audits. Teams are often surprised to find that the beliefs driving their biggest spend decisions were formed during an initial planning session and never touched again. For SEO specifically, this matters a lot. SEO strategies rest on assumptions about search intent, competitor authority, crawl accessibility, and ranking timelines. All of it is testable. But only if you've recorded the assumptions in the first place. #### Key Takeaways * An assumption register documents every unverified belief in your strategy so it can be tested rather than left to chance. * The evidence ledger pairs with the register to record what you have actually confirmed and how strong that confirmation is. * Naming assumptions forces accountability — someone has to own the task of finding the evidence. * Most failed campaigns can be traced back to an assumption that was never questioned; a register makes that visible before budget is spent. * For SEO, assumptions about intent, competition, and timelines need to be documented and revisited as data comes in. Used consistently, the assumption register and evidence ledger shift your planning from opinion-based to evidence-based. Not all at once. Incrementally, as each assumption gets tested, confirmed, or corrected. ## The Evidence Ledger: Recording What You've Learned ![The Evidence Ledger: Recording What You've Learned](/images/fcmo/fractional-cmo-command-centre--assumption-evidence-ledger/03.png) An assumption register tells you what you believe. An evidence ledger tells you what you've found out. Without it, your register is just a list of opinions that never gets updated. Evidence Ledger An evidence ledger is a structured record that documents the findings, data, and observations gathered when testing assumptions, allowing teams to update their beliefs based on what actually happened. The core function is simple: close the loop. You run a campaign, test a channel, publish a content series. Something happens. Traffic goes up or it doesn't. Leads come in at the quality you expected, or they don't. That's all evidence — and if you're not recording it somewhere consistent, it disappears. Mentioned in a meeting, referenced in a slide deck, gone by next quarter. A well-maintained ledger captures four things per entry: * What assumption was being tested * What action was taken * What the data showed * What the team now believes as a result That last point is the one most teams skip. Evidence without interpretation is just noise. Someone has to make a call — does this confirm, refute, or complicate the original assumption? > The teams we work with who make the fastest progress aren't necessarily running more experiments — they're better at learning from the ones they do run. An evidence ledger creates institutional memory that survives staff changes, agency transitions, and strategy pivots. The ledger doesn't need to be elaborate. A shared spreadsheet with consistent columns is enough. What matters is the discipline — logging outcomes when the data is fresh, not three weeks later when the context has evaporated. Each row should map back to a specific assumption in your register, so you can see at a glance what's been tested, what's still open, and what's been settled. Over time, this becomes one of the most valuable documents a marketing team holds. It shows what worked in specific contexts, what failed despite expectations, and where your mental models have been consistently wrong. We see this pattern constantly during audits — teams that skip the ledger keep relitigating the same decisions. Teams that maintain it actually get sharper. 1 in 4 Marketing assumptions are ever formally tested <48hrs Optimal window to log evidence after an experiment concludes 3x More likely to repeat failed tactics without a documented evidence record Q1–Q4 Evidence should be reviewed at every quarterly planning cycle So how do the two documents actually connect? The assumption register and evidence ledger work as a pair, one feeding the other. An assumption moves from the register to the ledger when there's enough data to say something meaningful. Some get confirmed and become working principles. Others get refuted — and save the team from doubling down in the wrong direction. A few stay genuinely uncertain and get retested under different conditions. That cycle — assume, test, record, revise — is what good marketing strategy actually looks like. ## Building and Maintaining the Register in Practice Knowing what an assumption register and evidence ledger are is one thing. Running them consistently inside a real marketing team is another. The gap between the two is where most efforts fall apart. Not because the concept is flawed — but because the process was never built into how the team actually works. Start with ownership. Someone needs to be responsible for the register, not as an admin task but as a genuine part of their role. In most marketing teams, this sits with the strategist or whoever leads campaign planning. Without a named owner, entries get added sporadically and the document quietly becomes a historical artefact. The format question is simpler than most teams make it. The register needs to live somewhere everyone can access and contribute to — not buried in someone's personal drive. A structured table with consistent columns is enough. What matters is that each entry captures the assumption itself, the confidence level at point of recording, the evidence that informed it, and the threshold that would trigger a review. The evidence ledger then attaches to each assumption as a running record of what's been observed since. #### Setting Up Your Register and Ledger * Assign a named owner responsible for maintaining the register * Choose a single, accessible location for the document * Define a consistent entry format before adding assumptions * Record assumptions at the point of strategy creation, not retrospectively * Set a confidence level and review threshold for each assumption * Attach an evidence ledger column or linked record to every entry * Schedule a regular cadence for reviewing and updating entries * Brief the whole team on how to read and contribute to the register Cadence matters as much as format. Monthly reviews are a reasonable baseline, but the register should also be updated whenever significant evidence arrives — a campaign report, a shift in search data, a change in customer behaviour. Waiting for the monthly slot when relevant data is already sitting in front of you creates lag. That lag is where bad decisions get made. #### Quarterly Planning Review A B2B marketing team builds their assumption register at the start of a quarter. One entry reads: 'Assumption: our target audience researches solutions via long-form content before contacting sales. Confidence: medium. Review threshold: if organic traffic to pillar pages drops below target for two consecutive months, revisit.' At the six-week mark, the evidence ledger shows traffic holding but time-on-page falling sharply. That single data point prompts a review of content depth and internal linking — well before the end-of-quarter debrief would have caught it. So what separates a usable register from a pointless one? Specificity. Vague entries like "paid social will drive awareness" tell you nothing about what evidence would confirm or challenge them. A well-written assumption names the audience, the channel behaviour, the expected outcome, and the metric that tells you whether it's holding. The more precise the assumption, the more obvious the evidence threshold becomes. #### ⚠ Treating the Register as a Formality Teams sometimes complete the register during strategy sessions and never return to it. If the document isn't reviewed when evidence arrives and isn't referenced during planning cycles, it has no practical value. The register only works if it's part of the decision-making process, not separate from it. One discipline worth building in from the start: when a campaign wraps or a reporting period closes, complete the evidence ledger entries before the team moves on. Trying to recall what you observed three weeks later is genuinely difficult. Capturing it at the point of review takes minutes, keeps the record clean, and makes it far easier to tell the difference between an assumption that was properly tested and one that was simply never challenged. A well-maintained register and ledger becomes one of the most useful planning tools a marketing team has. Not immediately — but over time. It reduces the risk of repeating the same strategic mistakes across quarters. It makes onboarding new team members into current strategy far simpler. And it gives senior stakeholders a clear view of what is known, what is assumed, and where the real uncertainty sits. ## How the Ledger Feeds Into Prioritisation Decisions Once you have an assumption register and an evidence ledger running in parallel, the real value isn't the records themselves. It's what they tell you about where to direct effort next. Most marketing teams treat prioritisation as a gut-feel exercise dressed up with a spreadsheet. The register and ledger give you something more useful: a structured record of what you believe, what you've tested, and what the results actually showed. The connection is simple. Your assumption register holds the things your strategy depends on being true. Your evidence ledger holds the outcomes — tests, campaigns, data reviews — that either support or challenge those assumptions. Bring them together in a planning session and you can see at a glance which assumptions are well-evidenced, which are untested, and which have already been disproved. That last category is where prioritisation tends to break down. Teams keep investing in channels, audiences, or messages that the evidence has already flagged as weak. Simply because no one connected what the ledger recorded to what the register still lists as a live assumption. We see this constantly during strategy reviews. The data is there. The assumption is still there. No one joined the dots. #### Turning Ledger Evidence Into Prioritisation Decisions 1. Pull the assumption register and evidence ledger into the same planning session 2. For each active assumption, check whether the ledger contains supporting, neutral, or contradicting evidence 3. Flag assumptions with contradicting evidence as candidates for deprioritisation or active testing 4. Identify assumptions that are high-stakes but have no ledger entries — these need a test before further budget is committed 5. Update assumption confidence ratings based on ledger findings before setting quarterly priorities 6. Document the prioritisation rationale in the register so future teams understand why decisions were made Working through this process creates a useful forcing function. You cannot push forward on instinct when the ledger is sitting in front of you showing a different picture. If an assumption underpins significant spend — a target audience segment, say, or a content format — and the ledger shows three consecutive periods of weak performance against it, that's a signal to deprioritise or restructure. Not to run the same approach again and hope something shifts. The register also protects you from the opposite mistake: cutting things too early. If an assumption is genuinely untested, or the evidence comes down to a single data point, the ledger will show that too. Deprioritising on thin evidence is just as damaging as ignoring strong evidence. The tricky part is telling the difference — and that's exactly what having both documents in the room allows you to do. TL;DR An assumption register and evidence ledger work together to make prioritisation decisions traceable, evidence-based, and less vulnerable to bias or staff turnover. * An assumption register documents what your strategy depends on being true, with confidence ratings and owners assigned * An evidence ledger records the outcomes of tests and campaigns, with interpretation attached to each entry * Together, they show which assumptions are well-supported, which are untested, and which the data has already challenged * Prioritisation becomes a structured review process rather than a gut-feel exercise * Both documents create an audit trail that protects strategic continuity across team changes For this to work in practice, prioritisation meetings need to include a formal review of both documents together. Not the ledger in isolation. The ledger tells you what happened. The register tells you what was at stake. Review them separately and you miss the connection that makes the whole system worth running. So what actually changes when teams do this consistently? Planning conversations shift character. Instead of debating opinions about what should work, teams debate what the evidence does and doesn't support — and where the genuine uncertainty sits. That's a more productive conversation. It also produces decisions that are far easier to defend when they get scrutinised later. The longer you maintain both documents, the more reliable your prioritisation gets. Early on there will be gaps — assumptions with no evidence, ledger entries pointing in different directions. That's expected. Over time, patterns emerge. You start to see which assumptions tend to hold, which channels consistently underperform against expectations, and where your team's instincts are sharpest — and where they're not. That accumulated record is what turns a planning tool into something you'd genuinely miss if it disappeared. ## High-Risk Assumptions: Which Ones to Test First Not all assumptions carry the same risk. Some, if wrong, cause minor inefficiencies — a campaign underperforms, you adjust the bid strategy, and move on. Others waste six months of budget, miss a product launch window, or send an entire channel strategy in the wrong direction. The job of your assumption register and evidence ledger isn't just documentation. It's to help you sequence testing so the most dangerous assumptions get attention first. So how do you tell the difference? ### Two Axes That Determine Priority Every assumption in your register sits somewhere on two axes: how confident you are that it's true, and how much damage it causes if it turns out to be false. High confidence, low impact assumptions rarely need formal testing. Note them for completeness, but don't burn resource on them. Low confidence, high impact assumptions are where the real danger lives. These go straight to the top of your testing queue. If your go-to-market strategy rests on the idea that a particular segment is ready to buy — and you have no hard evidence for that — you need to find out before you build around it. We see teams skip this step constantly. It rarely ends well. The other two quadrants sit in the middle ground. High confidence, high impact assumptions are worth monitoring for disconfirming signals even without active tests. Low confidence, low impact assumptions can wait, or get resolved opportunistically as data comes through normal campaign activity. #### Ranking Assumptions by Risk Level 1. List every assumption currently driving a live or planned marketing decision in your register 2. Score each assumption on confidence (1–5) based on the evidence already held in your ledger 3. Score each assumption on impact (1–5) based on what breaks if the assumption is wrong 4. Multiply the two scores to produce a risk rating, and sort the register by this figure descending 5. Assign the top-ranked assumptions to active tests or validation sprints with named owners and deadlines 6. Review the ranking monthly as new evidence enters the ledger and business priorities shift This scoring approach won't give you a perfect answer. But it gives you a defensible one — and it makes conversations with budget holders much easier. You're not asking for resources out of curiosity. You're pointing to a ranked list and explaining the cost of leaving the top items unresolved. ### What Makes an Assumption Genuinely High-Risk Beyond the scoring exercise, a few characteristics reliably flag assumptions as high-risk. Watch for these. **It's load-bearing.** Multiple downstream decisions depend on this assumption being true — risk concentrates fast. An assumption about which channel drives the majority of qualified demand will shape budget allocation, team structure, content investment, and reporting. Getting it wrong has a cascading effect. **It hasn't been tested in the current market context.** Historical evidence decays. An assumption that held in 2021 about customer acquisition costs in paid search may be materially wrong now. If the ledger entry is old and conditions have shifted, treat it as low confidence — regardless of how solid it once looked. **It's held unanimously.** When everyone on the team agrees on something without debate, that's often a sign the assumption has never been seriously examined. Consensus isn't evidence. Unanimity should prompt you to look harder, not less. **It's been used to reject a test.** If someone has said "we don't need to test that, we already know" — that assumption belongs in the high-risk pile. The confidence is asserted, not demonstrated. A common mistake we see in audits. 1–5 Confidence score range per assumption 1–5 Impact score range per assumption 25 Maximum risk rating triggering immediate testing Monthly Recommended register review cadence ### Tools That Support Assumption Testing at Scale The assumption register and evidence ledger set the strategic framework. But you also need tools that generate and capture the data that either validates or challenges what you think you know. #### Tools for Testing and Validating Marketing Assumptions Google Analytics 4HotjarOptimizelyWynterAttestHubSpotLooker Studio Qualitative tools like Wynter and Attest are particularly useful for assumptions about messaging and audience perception. Quantitative data alone won't tell you _why_ something isn't working. Experimentation platforms like Optimizely let you run structured tests rather than relying on before-and-after comparisons that conflate too many variables. Whatever tools you use, the discipline stays the same. Evidence needs to be recorded in the ledger against the specific assumption it relates to, with a clear interpretation of what it confirms, contradicts, or leaves open. ### Making Testing a Habit, Not a One-Off The most common failure mode we see isn't identifying the wrong assumptions to test. It's running a test once, drawing a conclusion, and treating the matter as closed. Markets change. Audiences change. What a test told you eighteen months ago may not hold today. Your assumption register should include a review date for every validated assumption — not just the ones still marked unresolved. When that date arrives, ask whether the conditions that made the original evidence valid still apply. If they don't, the confidence score drops, and the assumption may need to go back into the testing queue. This is how the register and ledger work together over time: not as a static archive, but as a live system that keeps your strategy honest as the market shifts around it. ## Stop Flying Blind on Marketing Strategy Most marketing teams are sitting on months — sometimes years — of campaign data, test results, and strategic decisions. But without a system connecting them, that information stays fragmented. The assumption register and evidence ledger aren't just documentation tools. Together, they form the foundation of a strategy that actually learns from itself. You've already identified which assumptions carry the most risk and which to test first. The next question is what happens after the test. This is where most teams stall. Data gets collected, results land in a Slack message or a post-campaign debrief, and then the insight disappears. The next person working on a similar campaign starts from scratch. The same assumption gets tested again — and nobody notices because there's no record showing it already happened. Breaking that cycle means connecting your assumption register to a living evidence base, and making sure every entry gets acted on, not just logged. ### Turning Test Results Into Strategic Decisions An assumption register without a connected evidence ledger is half a system. The register tells you what you believe and how confident you are. The ledger tells you whether reality matched. When a test completes, the evidence entry needs to do more than record the outcome. It needs to update the assumption's status, revise your confidence level where appropriate, and flag whether the finding changes anything downstream — a budget allocation, a channel mix, a content brief, a targeting parameter. That's the difference between a record and a decision tool. If your ledger produces entries that nobody acts on, it's functioning as an archive. If it produces entries that change how your team thinks and plans, it's functioning as strategy infrastructure. We see this distinction clearly during audits — the teams getting real value from these systems are the ones where a ledger entry can directly trigger a planning change. Your marketing assumptions deserve to be tested, tracked, and acted on systematically. [See How We Work](/our-work) ### Making the System Stick Across the Team Building the register and the ledger is the easy part. Keeping both maintained when the team is under pressure — that's the real challenge. Campaigns go live, deadlines shift, and documentation gets deprioritised fast. A few things separate systems that hold from ones that get quietly abandoned. **Assign ownership clearly.** Someone needs to be responsible for updating the register when assumptions are tested and reviewing the ledger at defined intervals. Without a named owner, both documents drift. **Tie updates to existing milestones.** The best time to update an assumption's status is at a campaign review or sprint retrospective — not as a separate task. Bolt it onto a meeting that already happens. **Keep the format simple enough to complete under pressure.** If an entry takes thirty minutes to write properly, it won't get written. A short structured format — assumption, test method, outcome, revised confidence, decision triggered — takes five minutes and captures everything that matters. **Review the register before planning, not after.** It should be a pre-planning input, not a post-campaign report. Before committing budget to a new initiative, check what you already know and what you still don't. Most teams get this backwards. The register becomes a retrospective document when it should be shaping what comes next. ### From Assumption Management to Compounding Returns The real value builds over time. In the first few months, you're mostly setting up the system and running initial tests. After six to twelve months, the ledger starts producing genuine strategic advantage. You know which assumptions in your category tend to be wrong. You have evidence for why certain channels work for your audience and others don't. Decisions get faster because you're not re-litigating the same questions every quarter. Each test makes the next decision cheaper. Each ledger entry narrows the gap between what you're assuming and what you actually know. Teams that operate this way don't eliminate uncertainty — no one does. But they stop using uncertainty as an excuse to avoid rigour. They test what matters, record what they find, and build strategies that improve with each iteration instead of resetting every time. ### Strategy Grounded in Evidence, Not Guesswork We help marketing teams build structured testing processes that turn assumptions into decisions and data into compounding strategic advantage. [Talk to Us](/contact) If your team is still relying on instinct and anecdote to make significant budget decisions, the assumption register and evidence ledger are the most practical fix available. They're not complicated. They don't require new tools. They require discipline — and the recognition that a wrong assumption, repeated across multiple campaigns, costs far more than testing it properly once. You might also find helpful [ Building an SEO Strategy How to Build an SEO Strategy That Holds Up to Scrutiny ](/seo-strategy) [ SEO Reporting Why SEO Reporting Fails Marketing Teams (And How to Fix It) ](/seo-reporting) [ Content Planning Content Planning Without Assumptions: A Structured Approach ](/content-planning) [ SEO Experiments How to Run SEO Experiments That Actually Tell You Something ](/seo-experiments) [Back to The CMO Command Centre](/fractional-cmo-command-centre) ## More on The CMO Command Centre [Decision PacketConvert reporting into issue, evidence, options, and a recommendation.](/fractional-cmo-command-centre/decision-packet)[Measurement SystemStable definitions and feedback loops without pretending attribution is perfect.](/fractional-cmo-command-centre/measurement-system)