Positioning is the shortest piece of work most companies ever buy and the one with the longest consequences, because everything downstream, the messaging, the site, the sales deck, the pricing conversation, inherits whatever the positioning decided. It is also the easiest work to fake, since the output is a document and a document can be written confidently about nothing. A useful positioning engagement produces a statement your engineers and your salespeople both recognise as true, backed by evidence from customers rather than from a workshop. A weak one produces adjectives. This page sets out what a brand positioning agency should actually deliver, how to test it before it is rolled out, and which questions expose a template.
What positioning is, stated narrowly
Positioning answers four questions in a form a stranger could repeat: who the product is for, what problem it solves, what the customer would use instead, and why they should switch. Everything else, the values, the personality, the tone of voice, hangs off those four and is worth nothing until they are settled. Note the third question, because it is the one most engagements skip. The real alternative is usually not the competitor named in the deck; it is a spreadsheet, an internal team, or doing nothing at all, and a positioning that argues against the wrong alternative fails silently in every sales conversation afterwards. Ask a candidate agency how they establish the alternative and whether they will talk to customers who chose someone else. Those lost deals contain more information than any number of satisfied customers will volunteer.
The process that separates research from a workshop
A credible engagement includes interviews: with your customers, your recent losses, your sales team and your product team, conducted by someone who can ask a follow-up question rather than reading a script. Twelve honest conversations will surface the language customers actually use and the objection that keeps recurring, and that language is the raw material for everything that follows. A workshop with your executives is useful for building agreement, not for discovering anything, because the people in the room already believe what they believe. If a proposal is mostly workshops with no external interviews, you are buying facilitation. Ask how many interviews are included, who conducts them, and whether you will see the transcripts or only the conclusions. Access to the raw material is what lets you check the analysis rather than trust it.
Testing the output before you roll it out
Positioning is testable, cheaply, and almost nobody does it. Take the statement to three salespeople and ask them to use it on live calls for a fortnight, then ask what landed and what they quietly dropped. Put the new value proposition on one landing page against the old one and watch the difference in enquiry quality, not just volume. Read the statement to a customer and ask whether it describes what they bought. If the positioning survives those three tests it is probably true. If the response is polite agreement and nothing changes in how anyone speaks, it is decoration. Build the tests into the engagement from the start, with the agency present for the results, because a provider confident in their process will welcome the check and one selling a document will explain why testing is premature.
Commercials, ownership and the trademark check
Agree the deliverables explicitly: the positioning statement, the messaging derived from it by audience, the evidence behind each claim, the interview material and any naming work with its source files. Everything created for you should be yours in editable form, including design files and font licences, since positioning work usually flows straight into identity work, which is normally bought next as brand identity services. If naming is in scope, insist on a clearance search before anyone falls in love with a name: the trademark register is public and searchable, and the USPTO publishes plain guidance on how registration and searching work. A name chosen without that check is a risk carried into every asset you print afterwards. Agree a term with a defined end rather than an open retainer, because positioning is a project and a project should finish.
Questions people ask about brand positioning agency
How long should a positioning engagement take?
Six to ten weeks is typical for a company with one product and a reachable customer base, most of which is interviews and scheduling rather than analysis. Faster is possible if the research is thin, which is exactly the tradeoff. Considerably longer usually means scope has expanded into identity, website and campaign work, which is fine but should be priced and scheduled as separate phases with their own decisions.
Can we do this in house?
Yes, and some companies do it well, but the difficulty is not method, it is candour. Internal teams struggle to interview lost deals honestly and struggle harder to write a statement that contradicts a founder's belief. An outsider's real value is often permission to say the uncomfortable thing. If you do it in house, have someone outside the leadership team run the interviews and report the findings unedited.
What if the positioning contradicts the product roadmap?
That is a finding, not a failure, and it is one of the more valuable outcomes an engagement can produce. Surface it explicitly rather than softening the statement until the conflict disappears. Either the roadmap changes or the positioning describes where the product will be in a defined timeframe, and both are legitimate. What is not legitimate is publishing a position the product cannot support, because sales will discover it first and customers second.
How do we judge two proposals that look identical?
Compare the research, not the language, since every proposal will describe a rigorous process. Count the external interviews included, check who conducts them, and ask whether lost customers are in scope. Then ask each agency for a client where the positioning changed something measurable and speak to the person who had to use it daily. The proposals stop looking identical very quickly.