Claim, evidence, clarification and language anchors: meaning and application

You have phrased a promise, but sales, the website and clients each read something different into it, and the evidence does not quite fit. Neurofactor uses a four-part structure for this: claim, evidence, clarification and language anchors. This article explains the parts, their link to the pillars of a proposition and how to write out one pillar.

Claim, evidence, clarification and language anchors is a Neurofactor working method for writing out one promise or pillar in four parts: what you promise (claim), what you show (evidence), what it is not (clarification) and which fixed words consistently accompany it (language anchors).

It is a practical method, not a validated scientific model. Toulmin's model of argument and research on evidence, repetition and processing ease provide context, but do not show that the structure works.

Neurofactor working definition: claim, evidence, clarification and language anchors is a four-part structure for writing out one promise or pillar so that everyone who communicates about it promises, shows, limits and words it the same way. It is an editorial tool, not researched as a model, and it predicts no effect.

The four parts:

1. Claim: what you promise, in one testable sentence from the recipient's perspective.
2. Evidence: what you show to make the claim plausible, such as a way of working, an example, a commitment, a figure with a source or an independent judgement. The evidence concerns the same claim.
3. Clarification: what the promise explicitly is not, so that the recipient does not read the claim more broadly or differently than intended.
4. Language anchors: the fixed words that belong to this pillar and recur consistently, from website to conversation.

Pillars. A proposition usually rests on a limited number of load-bearing promises: pillars, or proof pillars. How many are needed depends on the offer and target group; there is no fixed number. Each pillar gets its own claim, evidence, clarification and language anchors.

Separate articles, all in preparation, cover what you promise and why (associative value proposition), what makes evidence credible (evidence anchor and proof asset) and how to organise messages by target group (message matrix). The core message and communication lexicon will follow.

Neurofactor put the structure together; it has no origin in the literature. Several lines of research provide context, not validation.

Toulmin: claim, data and warrant

In The Uses of Argument (1958), philosopher Stephen Toulmin described a layout of arguments. He distinguishes the claim (the conclusion to be supported) from the data (the facts appealed to) and the warrant (the general rule connecting data and claim), plus backing (support for the warrant), qualifier (how strongly the claim is put) and rebuttal (circumstances in which the claim does not hold) (Verheij, 2005). Toulmin wrote this as a contribution to epistemology; its later popularity as the 'Toulmin model' in rhetoric and writing instruction was not his original aim (Toulmin, 1958/2003).

Claim and evidence resemble claim and data. Toulmin also shows what the four-part structure leaves implicit: the warrant, the reason this evidence supports this claim. Rebuttal is close to clarification, but describes when a claim does not apply; clarification mainly describes how a promise should not be read.

Research on evidence in persuasive communication

Reviewing fifty years of research, Reinard (1988) argued that evidence effects are more consistent than previously assumed, with the Elaboration Likelihood Model explaining when evidence matters most (Reinard, 1988). In an experiment with 324 participants, Hoeken (2001) supported a claim about a profitable cultural centre with anecdotal, statistical or causal evidence. Statistical evidence was the most convincing, and what participants considered strong evidence did not automatically convince them most (Hoeken, 2001). Hoeken cautions that the study used one topic and one version per evidence type. The lesson: your own sense of what convinces does not replace testing.

Repetition, fluency and their limits

Zajonc (1968) found that repeated exposure to unfamiliar words, characters and faces increased how favourably they were rated (mere exposure) (Zajonc, 1968). Alter and Oppenheimer (2009) describe processing fluency, the subjective ease of processing information, as a cue that influences judgements such as truth and liking. Context and people's beliefs about what ease means shape that influence (Alter & Oppenheimer, 2009). A meta-analysis of 51 studies shows that repeated statements are judged as more true (the truth effect) (Dechêne et al., 2010).

That fixed words make a message easier to recognise is, for language anchors, a hypothesis. The truth effect also shows the downside: repetition can make a statement seem more credible whether or not it is correct. So only repeat claims you can substantiate.

Message house and message pillars

Practitioners use a 'message house': a core message as the roof, supporting messages as pillars. It has no fixed scientific definition. The four-part structure adds clarification and language anchors to each pillar.

Claim. Write one sentence from the recipient's point of view: what changes for them? Avoid superlatives you cannot prove. Can a client tell whether you deliver?

Evidence. Choose evidence about exactly this claim. Make the 'warrant' explicit: why does this evidence make the claim plausible? Only use evidence that exists; for a new service that is often a way of working, a commitment or an example. See also what evidence means in a target group profile.

Clarification. Write two to four sentences that start with 'This is not…' or 'This does not mean that…'. Choose the most likely misunderstandings, for example from objections or client conversations. That clarification prevents misinterpretation is editorial advice, not researched for this method. It belongs in the message, not the small print.

Language anchors. Record a small number of load-bearing words, with a short explanation for each. Preferably choose words the target group uses itself; see also industry jargon and abbreviations and tone of voice. Note words to avoid because they suggest the wrong promise.

See also what a product promise is.

1. Choose the pillar. Start with the pillar that causes the most misunderstandings.
2. Phrase the claim. One sentence, from the recipient's perspective, verifiable.
3. Gather existing evidence. Note what each piece shows and why it supports this claim. Drop evidence about something else.
4. Write the clarification. Write one sentence per expected misunderstanding about what the promise is not.
5. Record language anchors. Choose the load-bearing words, describe their meaning and note words you avoid.
6. Check coherence. Does the evidence support the claim, does the clarification leave it intact and do the anchors fit?
7. Test with the target group. Have target group members retell the pillar in their own words; decide in advance what to look for. See how to turn target group insights into testable messages.

This example is fictional. No research outcomes are attached to it.

Situation. An accountancy firm is developing a new service for start-ups. The proposition currently rests on a few pillars, such as a fixed monthly price and one point of contact. The team first writes out the pillar that raised the most questions: staying on top of tax.

Question. How do website, advisers and quotes promise the same thing without promising more than the firm delivers?

Available information. There are no client results yet, so the evidence consists of what can already be checked: an example, a documented commitment and an explanation of the method.

Approach. Advisers receive the written-out pillar as a reference. Before publication, a few founders describe in their own words what they expect after reading it. The team decides in advance what to look for: does anyone read the claim as a tax saving, and does 'estimate' come across?

Possible interpretation. If participants read the claim as a saving, sharpen the clarification or anchors; it is not evidence that the service will fail.

Next step. The team writes out the next pillar and checks that the pillars' language anchors do not conflict.

• Evidence about something else. A general satisfaction score under a specific claim suggests support that is absent.
• Clarification that hollows out the claim. So many reservations that no promise remains. Clarification sharpens a promise; it does not withdraw it.
• Language anchors as a trick. Language anchors are meant for consistency, not to make a weak claim look stronger through repetition.
• Treating the method as evidence. A well-written pillar has not yet been tested with the target group.

The method makes messages consistent and verifiable, but cannot tell you whether a promise is relevant to the target group; that requires research.

With claim, evidence, clarification and language anchors, you write out one pillar of a proposition in full: what you promise, what you show, what it is not and which words recur. Argumentation theory and research on evidence and repetition provide context. Whether your target group understands and believes the pillar, you test separately.