How to Build a Messaging Hierarchy That Sticks
Five Levels, One Approved Wording Each
Short answer
A messaging hierarchy is five levels with one approved wording each: the positioning claim, the value proposition, two to four pillars, two to three proof points per pillar, and five objection answers. New copy quotes the value proposition verbatim and may paraphrase below it. One named owner freezes the wording and dates every change.
1. What is a messaging hierarchy?
A messaging hierarchy is a one-page document holding the approved wording at each level of a company message, from the single positioning claim down to the answers to the objections you hear most.
- Messaging hierarchy
- An ordered set of approved wordings, from a single top-level claim down to proof points and objection answers, so that everyone writing copy starts from the same sentences.
The word hierarchy is doing real work. The levels are not a list of themes, they are a chain of derivation: each level exists to make the level above it believable, and anything that does not support the level above it does not belong in the document at all.
That is also the failure test. If you can delete your top-level claim and every pillar underneath still makes sense on its own, you have written five parallel messages rather than one hierarchy, and the team will keep choosing between them.
2. What are the levels, and what belongs at each one?
Five levels: one positioning claim, one value proposition, two to four pillars, two to three proof points per pillar, and five objection answers. The counts are limits, not suggestions.
| Level | What it is | How many | Where it gets used |
|---|---|---|---|
| 1. Positioning claim | One sentence naming the buyer, the alternative and the difference | Exactly 1 | Internal only. Every level below derives from it |
| 2. Value proposition | The outcome the buyer gets, in the buyer's own words | Exactly 1 | Homepage hero, ad headlines, the first line of any pitch |
| 3. Pillars | The reasons the value proposition is true | 2 to 4 | Section headings, feature pages, deck slides |
| 4. Proof points | Evidence under each pillar: a number, a named customer, one demo step | 2 to 3 per pillar | Body copy, sales answers, case studies |
| 5. Objection answers | The agreed response to the objections you actually hear | Exactly 5 | Pricing page, sales calls, review replies, support macros |
Levels one and two carry one wording each because they are what consistency is measured against. Below that, paraphrasing is fine and even desirable, since a proof point read out in a sales call should not sound like it is being recited.
3. How do I build one in a single session?
Two hours, working top down from a settled positioning claim. Write level two in the buyer's own words, cut to three pillars, attach proof to each, then write the five objection answers verbatim.
Do it in one sitting with two people. Splitting it across a week produces levels written in different voices, which is the thing you are trying to prevent.
- Start from the positioning claim, not a blank page. Paste the existing positioning statement at the top. If there is not one, stop and write that first, because a hierarchy built without it becomes an argument about wording that is really an argument about strategy.
- Write the value proposition in the buyer words. Pull the actual phrasing from lost-deal notes, support tickets and reviews, and use their nouns rather than yours. If buyers say "keeping an eye on rivals" and you say "competitive intelligence workflows", they are searching for the first one.
- Cut the pillars to three. Write as many as you like, then keep three. The exercise of cutting is what surfaces which reasons are load-bearing, and three is the number a reader retains from one page of copy.If two pillars cannot be told apart without reading their proof points, they are one pillar.
- Attach two proof points to every pillar. A number, a named customer, a screenshot or a single demo step. A pillar with no available proof is a wish, and the correct move is to delete it now rather than to publish it and hope nobody asks.
- Write the five objection answers word for word. Take the five objections your last twenty deals actually raised, not the five you expect. Write each answer as a sentence somebody could say out loud, and keep the price objection first because it is always asked.
- Freeze the wording and date it. Mark levels one and two as fixed wording, note the date, and name the person who can change them. Without a date, the next person to disagree assumes the document was a draft, and it becomes one.
4. How do competitor claims change what belongs at each level?
A pillar two or more rivals already publish is parity, not differentiation, so it moves down to proof rather than sitting in a heading. Check every pillar against live competitor copy before freezing it.
The check is mechanical. Open the homepage and the top ad creative of five alternatives, write their claims into a column next to your three pillars, and count the overlaps. A pillar matched by two or more rivals is a claim the reader has already read twice today.
Three rules for what to do with the overlaps:
- Matched by two or more rivals: demote it. Keep it as a proof point, because buyers still need it answered, but do not spend a heading on it.
- Matched by one rival who is stronger than you on it: rewrite it to the part you do differently, or drop it. Competing head-on for a claim somebody else owns is the most expensive way to be second.
- Matched by nobody: promote it and put proof behind it this quarter. Unmatched pillars are where headline attention should go.
The awkward part is that this check expires. Rivals rewrite their homepages, and a pillar that was unclaimed in January can be on four sites by June. Oppira records competitor messaging and ad copy as it changes, so the overlap column can be refreshed in minutes rather than rebuilt from scratch each quarter.
5. How do I keep the team saying the same thing?
Put the hierarchy where copy is written, allow verbatim quoting at levels one and two only, and review new work against the document rather than against taste. Consistency is a location problem before it is a discipline problem.
Four mechanisms, in order of how much drift each one prevents:
- One page, in the tool where copy actually gets written. A hierarchy living in a slide deck nobody opens will be reinvented by whoever is on deadline.
- Copy-pasteable sentences at levels one and two. If the approved wording has to be retyped, it will be improved in transit, and every improvement is a small divergence.
- A single review question: which pillar does this piece serve, and which proof point does it use? Copy that cannot answer either is not off-brand, it is unattached.
- A named owner who is allowed to say no, and who dates every change. Two people with equal authority over the wording produces two wordings within a month.
Notice that none of these is a style guide. Tone, punctuation and voice are a separate document with a separate purpose, and merging them buries the five decisions that matter under thirty pages of preference.
6. How do I know the hierarchy is actually working?
Three tests, each under fifteen minutes: ask three colleagues for the value proposition, map your last ten pieces of copy to pillars, and count how often the five objections still arrive unanswered.
Run all three once a quarter and write the results down:
- The wording test: ask three people to write the value proposition from memory. Different phrasing is fine. A different promise means level two is not being used.
- The mapping test: take the last ten published pieces and mark which pillar each one serves. More than two unmapped pieces out of ten is drift, and it usually means the pillars are too abstract to write from.
- The objection test: count how often each of the five objections came up and how often the agreed answer was used. An objection answered five different ways is a level five that exists on paper only.
The mapping test is the one that produces the most useful surprise. Teams routinely find that their most-published theme is not a pillar at all, which is either a signal to promote it or evidence that the calendar is running on habit.
7. What breaks a messaging hierarchy?
Five things break it: too many pillars, pillars that are really features, missing proof, a second hierarchy invented for a campaign, and storing the document somewhere copy is not written.
| Failure | How it shows up | The fix |
|---|---|---|
| Five or more pillars | A homepage where every section has equal weight and nothing is remembered | Keep three. Demote the rest to proof points under the survivors |
| Pillars that are features | Headings that name parts of the product rather than reasons to believe | Restate each as the outcome it buys, and keep the feature as its proof |
| Proof points missing | Copy that asserts, then changes subject. Sales adds numbers nobody approved | Two proof points per pillar or the pillar is deleted this week |
| A campaign hierarchy invented alongside it | Two value propositions in market at once, usually from paid and organic | Campaigns get their own angle at level three or below, never a new level two |
| Stored where copy is not written | Everyone knows it exists, nobody quotes it, wording diverges monthly | Move it into the doc or tool where drafts are written, and link it in the brief template |
The campaign hierarchy is the most damaging of the five and the hardest to refuse, because it always arrives with a good reason. Hold the line at level two: a campaign may argue a different pillar, but if it needs a different promise, that is a positioning conversation and not a campaign brief.
Key Takeaways
Five levels, one approved wording each
Positioning claim, value proposition, two to four pillars, two to three proof points per pillar, and five objection answers. The counts are limits.
Cut the pillars to three
Three is what a reader retains from one page, and the act of cutting is what reveals which reasons are actually load-bearing.
A pillar with no proof gets deleted, not published
Two proof points per pillar: a number, a named customer, or one demo step. Unproven pillars force sales to invent evidence nobody approved.
Demote pillars two or more rivals already claim
Parity claims still need answering, so keep them as proof points, but do not spend a heading on something the reader has already read twice today.
Consistency is a location problem first
Keep the one page where copy is actually written, with copy-pasteable sentences at levels one and two. A hierarchy in an unopened deck gets reinvented on deadline.
Refuse a second level two
Campaigns may argue a different pillar. A campaign that needs a different promise is a positioning conversation, not a brief.
Frequently Asked Questions
Explore More
Related analyses, benchmarks, and industry insights
Related Guides
Glossary Terms
Turn reading into a reaction
Oppira watches your competitors, keeps your playbook current, and drafts the response. Start free and see your market clearly by tomorrow.