Reading a competitor's brand is mostly an exercise in restraint. The instinct is to compile a swipe file, list the things you like, and quietly absorb them. The work is the opposite — to read carefully enough that you can name what the competitor is deliberately doing, distinguish it from what they are accidentally doing, and refuse to import either into your own brand without a strategic reason. The interesting part is the refusal.

Why most competitor brand analysis becomes derivative

Most teams that audit competitor brands end up looking more like their competitors a quarter later, not less. The mechanism is straightforward. The team gathers a slide of competitor screenshots, talks through what each one is doing, and identifies a handful of patterns. The patterns are good ideas. The team adopts some of them. By the next quarterly review, the company's website, deck, and messaging have all drifted towards the category mean, and the brand has lost some of the specificity that distinguished it. The drift happens because competitor analysis without a strong filter is a force towards the average. Each individual borrowed element is defensible. The aggregate is generic. The pattern is well known in product strategy and design — the same dynamic operates on brand. The teams that resist it are the ones whose competitor analysis produces refusals as well as discoveries.

The two questions worth asking of a competitor's brand

A serviceable read of a competitor brand answers two questions and treats the rest as noise. First: what is this competitor's strategic positioning, expressed in the brand work — the audience they have chosen, the problem they have framed, the kind of company they are claiming to be? Second: where does the brand work succeed at expressing that positioning, where does it fail, and what does the difference tell you about the competitor's operating constraints? Both questions are about positioning, not aesthetics. The colour palette is downstream of positioning. The wordmark is downstream of positioning. The tone of voice is downstream of positioning. A competitor analysis that gets pulled into aesthetics before it has resolved the positioning question is reading the second draft, not the first. The team will absorb surface decisions without understanding the strategic decisions underneath, and the result will be derivative.

How to read positioning from brand work

The cleanest read of competitor positioning comes from three artefacts in combination — the homepage hero, the most recent product or campaign launch page, and a public-facing piece of founder or executive communication (a keynote, a long-form post, an investor letter). Each one shows the competitor's positioning under different constraints. The homepage shows the official version. The launch page shows the operational version — the language used when something needs to convert. The founder communication shows the version the principals actually believe. The interesting work is in the gaps between the three. When the founder voice is sharper than the homepage, the company is leadership-led but the marketing function has compromised the positioning. When the launch page is sharper than the homepage, the company has learned in operation what it has not yet codified in brand. When all three are aligned and specific, the competitor is well-positioned. When all three are aligned and generic, the competitor has positioned itself in a place that does not require differentiation — usually because it is competing on price or distribution rather than brand.

What the visual system is actually telling you

Visual systems express two strategic decisions: where the brand is sitting in the category's familiar visual language, and where it is deliberately departing from it. The departures are the signal. A SaaS competitor using a serif typeface in a category dominated by sans-serifs is making a positioning statement. A consumer competitor using a muted, editorial palette in a category that defaults to high-saturation is making a positioning statement. The default choices tell you about the team's resources; the departures tell you about the team's beliefs. The mistake teams make in competitor analysis is to record the visual choices without separating the departures from the defaults. Both end up in the swipe file. When the team imports a default — a category-standard sans-serif, a category-standard saturated palette — it gains nothing distinctive. When it imports a departure without understanding why the competitor made it, it imports the surface and discards the reasoning. A useful competitor audit names each visual choice as either default or departure, and asks of each departure: what is this saying, and would I want to say it about my own brand?

What the tone of voice is actually telling you

Tone analysis runs into the same trap. The team reads competitor copy, identifies a tone they like, and starts writing in something close to it. The result is brand voice convergence across the category. A more useful read separates three layers of the competitor's voice: the register (formal versus conversational), the structural patterns (how arguments are built, how facts are presented), and the lexical choices (the specific vocabulary, including the terms the competitor avoids). Of the three, the lexical layer is the most diagnostic. A competitor that consistently refuses certain category clichés is making a positioning argument; a competitor that adopts them all is conceding the positioning conversation. A competitor whose vocabulary is precise and consistent has a brand voice document that is actually being used; a competitor whose vocabulary drifts across surfaces does not. Reading the lexicon tells you whether the competitor's brand voice exists as a system or as a personality, and which surfaces the voice has reached and which it has not.

The refusal layer

The most important output of a competitor brand audit is the refusal list — the things the competitor is doing that you have decided you will not do, and the reasons. The refusal list is harder to write than the swipe file because it requires a position. It is also where the audit becomes strategic rather than decorative. A serviceable refusal list names at least three categories of refusal. First, defaults the competitor has imported that you do not want to import — the standard sans-serif, the standard saturated palette, the standard tone of voice — because doing so would erase your distinctiveness. Second, departures the competitor has made that you do not want to copy — interesting choices that belong to their positioning, not yours, and would look borrowed if you adopted them. Third, things the competitor is doing badly that you might be tempted to do badly in the same way — usually because the failure is in service of a goal you share, and the failure mode is contagious. The refusal list is not a permanent document. It gets revisited each time a meaningful competitor change is observed. But it is the artefact that prevents drift towards the category average. Without it, the audit feeds the drift.

Common errors to watch for

Auditing in screenshots. The audit consists entirely of static images pasted into a deck. The team never reads the competitor's content, never tries the product, never reads the founder voice. The audit captures surface and misses positioning. Auditing the wrong competitors. The competitor list contains category leaders and direct competitors but not adjacent companies whose brand the company is actually being compared to in customer minds — the higher-end consultancy the founder used to work for, the agency the prospect just rejected, the open-source alternative. The audit reads the wrong field. Auditing once. The audit is a one-off project, completed at a moment in time, then filed. Competitor brands evolve. The audit should be re-run at least annually and ideally on a trigger — a competitor's rebrand, a major launch, a leadership change. Auditing without naming. The patterns observed are described in generic language — "modern, friendly, professional" — that captures nothing. A useful audit names patterns specifically enough that another team would recognise them in a blind test.

What This Looks Like in Practice

In the work with BGR, the competitor audit at the start of the brand engagement initially looked like a swipe file. The team had collected screenshots of seven category competitors and a list of liked elements. The work was to take each of those elements and ask two questions — what is this saying about the competitor's positioning, and is it true of BGR? Roughly half of the liked elements turned out to belong to the competitors' positioning, not BGR's, and went onto the refusal list. The remaining elements were not borrowed; they were the prompts for a separate conversation about what BGR's equivalent answer should be. The output of the audit was a refusal list of nine items and a list of fourteen open questions the brand strategy needed to answer in its own terms. The audit had stopped being a swipe file and started being a positioning document.

Closing

A competitor brand audit produces three outputs if it is doing its job: a clear read of the competitor's positioning, expressed in a few sentences; a small list of things the competitor is doing that point to questions your brand needs to answer in its own terms; and a refusal list of things you have decided not to import. The swipe file is the absence of the work, not the work itself. Reading well enough to refuse is the discipline. If you are about to start a competitor audit and want to avoid the drift towards the category average, we are happy to walk through how to structure the read so it produces a positioning document rather than a swipe file.