Measuring brand without inventing numbers

Brand measurement is where small businesses either invent numbers or give up. The useful position is in between: a few things can be observed honestly, none of them is proof, and the direction over a year is enough to make decisions with.

What cannot be measured at this size

Worth saying first, so the rest is believable. Brand awareness, brand equity, share of voice and sentiment are all real concepts and all require survey panels or tracking studies that cost more than a small business spends on marketing in a year. Anybody offering you these cheaply is producing a number rather than measuring one.

Attribution has the same problem in a different form. A customer who saw you three times over two months and then searched your name is recorded as having arrived from search, and the three impressions that caused it are invisible. This is not a tooling gap you can close; it is the nature of the thing.

What can be observed

• Searches for your own name. People looking for you specifically rather than for your category. The single most useful signal available, it is free, and it is in your search console.
• The share of orders arriving direct, with no referrer and no campaign attached. Imperfect, because direct is also where untracked traffic lands, but the trend is meaningful.
• Cost per customer in paid channels at constant spend, over a year. If being known is working, this should resist the usual upward drift.
• Repeat purchase rate, which is the compounded verdict on whether the experience matched the expectation.
• What people say when asked how they heard of you. Unreliable individually, informative in aggregate, and it costs one optional field at checkout.
• Unprompted mentions: somebody writing about you without being asked. Rare enough to count individually when you are small.

The one question worth asking customers

Not how satisfied they are, which produces agreeable answers and no information. Ask what they would have done if you had not existed.

The answers sort into three, and each means something different. Gone to a specific competitor means you are in a comparison set and competing on terms. Done it themselves means your real rival is their own time, which changes what every page should argue. Nothing, means you created the purchase, which is the strongest position available and the one most businesses do not realise they are in.

Ten answers to that question are worth more than any dashboard, and they take a fortnight of asking.

What to write down, and when

1. 01Once a quarter, record the six observable figures above with the date. Fifteen minutes.
2. 02Once a year, compare each with the same quarter the previous year and write one sentence about the direction.
3. 03Keep the customer answers verbatim rather than categorised, because the wording is the part that is useful later.
4. 04Do not change the measures when they are inconvenient. A measure you abandon in a bad quarter was never a measure.

And resist building a dashboard. The value here is in a short record kept consistently for years, which a spreadsheet does perfectly and a reporting tool makes look more rigorous than it is.

What to actually do with the answers

Measurement that changes nothing is a hobby. Each of the observable figures points at a specific decision, and it is worth writing the decision down in advance so the number has somewhere to go.

• Searches for your own name flat or falling while sales hold: you are buying every customer. That is the argument for spending on being known, and it is the clearest one you will get.
• Paid cost per customer rising at constant spend: you are reaching past your cheapest audience. Either accept a lower volume at a sustainable cost, or build the recognition that makes the same spend convert better.
• Repeat rate low: the experience is not matching the expectation, and no amount of acquisition fixes that. This is a product and service problem wearing a marketing problem's clothes.
• People saying they were recommended: the most valuable finding on the list. Find out what the recommender actually said, because that sentence is your positioning working, and it is usually better than the one on your website.

That last one is worth chasing properly. A customer who arrived on a recommendation can usually tell you the words that were used, and those words are free market research into what you are actually good for.