Competitor research

In depth

Sources and what each is good for

Reviews are the easiest source to over-read. The people who post them form a self-selected sample, weighted towards the delighted and the furious, so they show what can go right or wrong, not how often. Count themes in them as you would code open answers, and never report a share of reviewers as a share of customers.

For listed companies, filings are the richest public source: the US Securities and Exchange Commission’s EDGAR system gives free access to the documents public companies file, with full-text search, and the secondary-research lesson describes what such filings contain. Most of a small business’s rivals are private, though, and for them the shelf, the trade and the customer matter more.

Customers see a rival’s offer from the side that matters. A win–loss analysis interviews buyers soon after they decide, both those who chose you and those who chose a competitor, about what decided it. Asking only the buyers who chose you invites survivorship bias, the flaw in M01’s client interviews; the prospects who went elsewhere hold the evidence you lack.

Store checks and shelf-price audits

A store check is a visit to a set of outlets to record what is actually on the shelf: which products, at what prices, with how much space, what is on promotion and what is out of stock. A shelf-price audit is a store check focused on prices. Its value is that it measures what shoppers face, not what a price list says. Five design decisions make an audit worth analysing.

1. Purpose. Start from the decision. Ana’s is whether $3.49 fits, so she needs single cans of the rival brands, in the channels of her plan and, for comparison, convenience stores.
2. Which stores. Cover each channel, and say how the stores were chosen. Only a probability sample of stores, drawn at random from a list of all of them, supports an estimate for the whole city with a margin of error (the sampling-methods lesson); otherwise the audit describes the stores visited.
3. When. Visit every store in the same week, so that prices are compared under the same promotions; repeat the audit to learn how often rivals promote.
4. What to record. The shelf price as shown, and when a promotion is on, the regular price too; the pack size; facings and out-of-stocks; the store and the date.
5. Like with like. Compare the same pack size, or convert to a price per unit of volume, and compare within a channel before comparing across channels.

* on promotion that week. Store names withheld.

Includes the two promoted prices at S2.

Summit Brew is the dearest brand in every channel and Daily Grind RTD the cheapest wherever it is sold. Convenience stores charge most for both brands they stock ($4.06 and $2.52 on average), and natural grocers a little more than supermarkets. Across all stores the ranges are $3.59 to $4.19 for Summit Brew, $1.99 to $2.59 for Daily Grind RTD and $2.79 to $3.09 for Arcadia Coffee.

Be careful with the “All stores” column. Daily Grind RTD isn’t sold in natural grocers and Arcadia Coffee isn’t sold in convenience stores, so each brand’s overall average mixes different channels, and the comparison between them partly measures where they are sold. Compare within a channel first.

Two of the 24 prices were promotions, both at S2: Summit Brew at $3.59 and Daily Grind RTD at $1.99. Against M02’s usual prices ($3.79 and $2.29) those are cuts of 5.3% and 13.1%, the promotion depth. Without them the supermarket averages are $3.76 for Summit Brew and $2.26 for Daily Grind RTD. One week can’t show how often the rivals promote, and the audit missed something it should have recorded: the regular price beside each promoted one.

Finally, what the audit can’t say. Its ten stores were not drawn at random (four of Harbour City’s 120 supermarkets, for instance), so its averages describe those stores in that week, with no margin of error for the city. A simple average of shelf prices also ignores which stores sell most. It is good evidence of where the rivals’ prices sit; it is not evidence of what shoppers will do at $3.49, which is M02’s price-response question.

Mystery shopping and its limits

Mystery shopping means sending a trained researcher, posing as an ordinary customer, to experience and record a service. When a company shops its own outlets to check service standards, MRS’s guidelines require that its staff have been told it may happen. Shopping competitors is different, because their staff can’t be told, and the professional codes allow it only with care.

The MRS Code of Conduct (2023) allows recording only in a company’s own mystery shopping, where staff have been told it may happen, and refers members to separate MRS guidelines. Those guidelines (2011) say that a competitor’s staff must not be identified or recorded in any way, results must be reported in aggregate, the time taken should be as short as possible and no more than a normal customer enquiry, and the staff must suffer no detriment.

ESOMAR’s guideline on mystery shopping studies (2005) adds that such studies should be carried out only where no substantial disadvantage will result, with the time and other demands reasonable and kept to a minimum, since for a self-employed professional an enquiry can literally cost lost income; any booking should be cancelled as quickly as possible. Results must not be used to discipline or otherwise disadvantage individuals.

Competitive-intelligence professionals have a stricter rule for interviews. The code of ethics of the Strategic Consortium of Intelligence Professionals (SCIP) requires members to comply with all applicable laws and to disclose all relevant information, including their identity and organisation, before every interview. A sound working rule follows from all three: observe what any customer would see and buy what any customer would buy; don’t make a rival do work for a sale that will never happen (a detailed quote, a booking you will cancel) beyond what an ordinary enquiry takes; and never use a false identity to obtain information a real customer wouldn’t be given, or to question a rival’s staff.

The legal line: trade secrets

The law draws the outer line at trade secrets. In US federal law a trade secret is business information whose owner has taken reasonable measures to keep it secret and which has economic value because it is not generally known or readily found out by proper means (18 U.S.C. § 1839). Acquiring one by “improper means” is misappropriation under the same section, and improper means include theft, bribery, misrepresentation and inducing a breach of a duty to keep it secret; reverse engineering, independent derivation and other lawful means are expressly excluded.

The EU’s Trade Secrets Directive (2016/943) draws a similar line. In the UK its rules on lawful and unlawful acquisition take effect through the law of breach of confidence, with the Trade Secrets (Enforcement, etc.) Regulations 2018 filling gaps, as their explanatory note explains. Under the directive, observing, studying, taking apart or testing a product that has been made available to the public is lawful acquisition; unauthorised access to or copying of a holder’s documents, and any other conduct contrary to honest commercial practices, is not.

From findings to implications

Benchmarking lays out your offer or performance beside competitors’ on the same measures. The table is only half the work: each row should end in an implication or a question for your own offer. A gap matters only if customers care about it, and a difference customers can’t see is no difference to them.

Read down the last column and a research plan appears. The price gap is known; what isn’t known is whether the buyers who chose Brewline would pay $1.50 more for baristas if they saw them, which is a question for win–loss calls with the five prospects. Brewline’s second van is a threat to watch, and its job advertisements will say when it arrives. None of this needs anything but public information and honest questions.