HASHTRACKER

Brand reputation 3 min read

Competitor benchmarking on social media, done properly

Competitor benchmarking on social media can be rigged by picking the right month. Measure every brand identically and report share by reach, not only by volume.

Cover drawn as a line that climbs and falls

Competitor benchmarking on social media is easy to produce and easy to rig. Pick the right competitors, the right month and the right metric, and anyone can come first.

The useful version is built to survive being questioned.

Benchmarking social X Twitter
Benchmarking social X Twitter

Tell us the category you compete in and we will define the comparison before running it.

Tell us the category you compete in

What competitor benchmarking has to get right

One thing above all: every brand measured identically. Same period, same operator structure, same decision about reposts, same data source. A benchmark where one brand's figure came from an API and another's came from an estimate is a chart with no meaning, and it will still get presented because charts are persuasive even when they are wrong.

Choose the competitor set by a rule

The temptation is to pick competitors you beat. Everyone does it and everyone in the room knows it is happening. A defensible set has a stated rule behind it: the three closest by revenue, everyone in a named category list, or whoever the brief specified. Write the rule into the report. An unexplained set is the first thing questioned and the fastest way to lose the room.

Normalise before comparing anything

Absolute totals measure how much each brand posted, not how well. A competitor posting four times a day will beat one posting twice a week on every raw number, regardless of quality.

Use impressions per post and engagement per post instead. Both are independent of output, which is what makes the comparison say something about performance rather than about staffing.

Report share by volume and by reach separately

Divide your mentions by the category total, then divide your impressions by the category total. Report both. They frequently disagree, and the disagreement is usually the most useful finding in the whole exercise. A competitor mentioned half as often as you can be seen twice as much because one large account posted about them, and only the second calculation reveals it.

Split earned from owned

For each brand, separate mentions written by the brand itself from mentions written by everyone else. Then compare those two columns across the set. A brand whose share comes mostly from its own posting is buying its position. One whose share comes from other accounts has earned it, and that distinction predicts what happens when a brand stops posting for a fortnight.

Watch the direction, not the position

A single month's ranking is noise. Plot it monthly and read the slope instead. A competitor climbing steadily for four months matters more than whoever happens to be ahead today, and that pattern only appears with a consistent method applied over time. It is also the finding that tends to change budgets.

Check for events that distort the period

Before drawing conclusions, look at the underlying data for launches, crises and news cycles. A competitor's share jumping because they had an outage is not a win for them. Reporting it as one makes clear you did not read the mentions, and one instance of that undoes a lot of credibility. Note the event in the report and adjust the reading rather than the numbers.

How often to rerun it

Monthly is the right cadence for an active category and quarterly is enough for a stable one. What matters more than the frequency is that nothing changes between runs: same query, same set, same exclusions.

A benchmark rerun with a tweaked method is a new benchmark, and comparing it to the previous one produces a trend that exists only in the spreadsheet. If the method has to change, say so and treat that month as a new baseline rather than pretending the series continues.

Everything we gather stays within the Twitter compliance rules and we hand over the rows, so the comparison can be audited brand by brand. Tell us who to measure against and we will scope it.

Tell us who to measure against

Choosing a subscription

Every subscription carries the same product. The only decision is how many tweets you need to analyse.

See the subscriptions