πŸš€ NEW WHEATPASTING MARKETS Discover our wheatpasting campaigns in : πŸ‡¬πŸ‡§ London β€’ πŸ‡¨πŸ‡¦ Canada β€’ πŸ‡²πŸ‡½ Mexico City
Back to Articles
Street team handing out flyers during a guerrilla marketing campaign in a busy urban area

The hardest thing about reporting on brand ambassador work is that almost all of the data is self-reported by the people being measured.

That is not a cynical observation, it is a structural one. A poster is photographed by someone other than the poster. A truck is tracked by a device that does not care about the driver’s shift. But a brand ambassador’s output, how many people they spoke to, how many samples went out, what the reaction was, is recorded by the ambassador, usually on a phone, usually while doing something else.

Every reporting problem in this format comes back to that fact. Design around it and you get numbers you can trust. Ignore it and you get a spreadsheet full of round figures that tell you nothing.

I have built and read a lot of these reports. This page explains what should be in one, which metrics are actually meaningful, how capture works in the field, where the data goes wrong, and how to read a report critically enough to know whether the campaign did anything.

If you are commissioning ambassador work and the reporting conversation has not happened yet, have it before the first shift. Retrofitting it does not work.

Start by separating three different things

Most weak reports collapse these together, which is why they are so hard to interpret.

Delivery evidence. Did the campaign happen as agreed? Right people, right places, right hours, right materials. This is contractual and it should be verifiable independently of anyone’s opinion.

Activity output. What did the team actually do? Interactions, samples distributed, sign-ups captured, codes issued. This is the operational throughput.

Outcome. Did it achieve the objective? Footfall, redemptions, conversions, awareness shift in the market.

Those three answer different questions for different people. Delivery protects the buyer. Activity tells you whether the operation was efficient. Outcome tells you whether the investment was worth repeating.

A report that gives you a big activity number and nothing else is the most common failure. It proves effort, not effect, and effort is the easiest thing to inflate.

Benchmark I apply: a report should let a reader answer all three questions separately, and should never present an activity figure in a way that implies outcome.

Delivery evidence: the part that should be objective

This is where I would spend the first part of any report, and where buyers should be most demanding, because it is the only part that can be made genuinely hard to fudge.

Attendance and hours. Who worked, where, from when to when. Timestamped check-in and check-out, ideally location-stamped at the point of check-in. Not a roster filled in afterwards.

Position coverage. Which agreed positions were staffed, during which hours. If a plan called for four positions and one was unmanned for two hours, the report should say so plainly rather than averaging it away.

Photographic evidence. Timestamped photographs of the team working, in context, with enough surrounding detail to identify the location. Multiple captures spread through the shift rather than a single photo at the start when everyone is fresh and the stock is full.

Materials reconciliation. Stock issued, stock returned, stock distributed. This is the single most useful cross-check in the whole format, and it is underused. The difference between issued and returned is a hard number that constrains the claimed distribution figure. If an ambassador reports more interactions resulting in a sample than the stock count supports, you have found something.

Conditions log. Weather, foot traffic character, anything unusual. Written during the shift.

Benchmark: delivery evidence should be assembled from sources that are not the ambassador’s own count. Check-ins, photographs and stock reconciliation all qualify. Self-reported hours alone do not.

Activity metrics, and which ones are worth collecting

Here is where I would push back on a lot of standard practice, because several commonly reported metrics are close to meaningless.

Interactions or conversations. Reported constantly, and almost always soft. What counts as an interaction? Eye contact? A greeting? A conversation? Without a tight definition, this number is an estimate of an undefined thing, and it varies more between ambassadors than between campaigns. If you collect it, define it precisely and expect it to be approximate.

Samples or items distributed. Much better, because stock reconciliation constrains it. This is close to a hard number and I treat it as one of the anchors of the report.

Sign-ups, scans, code issues. The best activity metrics available, because they are captured by a system rather than by a person. Each one is a discrete, timestamped, verifiable event. If the campaign can be designed so that its primary activity metric is system-captured, do that.

Refusals. Rarely collected and genuinely useful. A high refusal rate at one position tells you the placement or the approach is wrong, and it is actionable during the campaign rather than after it.

Dwell or engagement time. Soft, and usually not worth the collection burden.

Qualitative feedback. The most undervalued output in this format. What people actually said, in their words, about the product, the price perception, the competition they mentioned, the confusion they expressed. A team speaking to many people a day is a research instrument nobody is reading. I ask for verbatim notes, and clients are routinely surprised by how useful they are.

Benchmark: pick a small number of metrics and define each one tightly. A team asked to track eight things tracks none of them well. Three well-defined metrics beat eight vague ones.

How capture actually works in the field

The collection method determines the data quality far more than the staff’s good intentions do.

Capture at the moment, not at the end. Anything logged retrospectively is reconstructed. People round, they estimate, they fill in what feels right. A tally captured as it happens is a different quality of data entirely.

Make it one tap. If logging an interaction takes more than a second, it will not happen during a busy period, which is exactly when the data matters most. The collection interface has to be faster than the activity it measures.

Prefer systems that capture themselves. A scanned code, a sign-up form, a redeemed voucher. These create their own record without the ambassador doing anything. Every metric you can move into this category is a metric you no longer have to trust anyone about.

Separate counting from selling. An ambassador in the middle of a good conversation should not be breaking off to log it. Either the logging is instant or it is batched at natural pauses with a short interval.

The lead owns the shift report. The on-site lead writes a short structured report within the day, covering positions, hours, output, conditions, issues and verbatim feedback. Written same day, not assembled at the end of the week from memory.

Benchmark: no shift report written more than a day after the shift. The decay in accuracy beyond that is steep.

Where the numbers go wrong

Being direct about this, because buyers should know what they are reading.

Round numbers. A report full of figures ending in zero or five was estimated, not counted. Real tallies are untidy.

Suspiciously consistent output across people and days. Human performance varies, weather varies, foot traffic varies. Identical output across a week is a sign of a template being filled in.

Interaction counts that outrun the stock. The reconciliation catches this.

Totals without position breakdown. Aggregating hides the fact that one position produced most of the output and another produced nothing. The breakdown is where the useful information is.

Sample photography presented as full coverage. A handful of good photographs is a sample. Ask what the capture standard was.

No conditions context. A wet Tuesday and a dry Saturday are not comparable, and a report that does not note the difference is unreadable a month later.

Missing hours quietly absorbed. A position unmanned for part of a shift should appear as a line, not disappear into an average.

None of this means the format cannot be measured well. It means the measurement has to be designed with the knowledge that the easiest thing to inflate is the number people are judged on.

Connecting activity to outcome

This is the question that decides whether the programme gets renewed, and it is the one most reports do not even attempt.

The mechanisms that work are the ones designed in before the campaign.

Unique codes. Issued in person, redeemed later. Gives you a hard, attributable chain from a street interaction to an action. The redemption rate is also a quality signal: a team handing out codes nobody uses is reaching the wrong people or pitching badly.

Distinct landing paths. A destination used only by this activity, so traffic to it is attributable. Keep it short enough that someone can remember it, because nobody types a long path from a leaflet.

Venue footfall. Where the objective is driving people to a nearby location, footfall data compared against baseline periods is strong evidence, particularly if the team worked specific approaches on specific days.

Market-level comparison. Run the activity in one market and compare against a similar market without it. Crude, but for awareness objectives it is often the most honest approach available.

Redemption and conversion data from the client side. Frequently the client holds the outcome data and never joins it back to the activity data. Agreeing who joins these two datasets, and when, is worth doing before the campaign.

Benchmark: at least one system-captured outcome mechanism per campaign, agreed and built before the first shift, with a named owner on the client side responsible for it. Campaigns where this is nobody’s job are the ones that end with an unanswerable question.

Reporting cadence and format

During the campaign. Daily or per-shift summaries, short, structured, same day. The point of in-campaign reporting is not record keeping, it is intervention. A position that is not working should be moved on day two, not noted in a final report in week five.

That is the single biggest argument for fast reporting. The data is only worth collecting if someone acts on it while the campaign is live.

Weekly. Consolidated view with position-level breakdown, trends, stock reconciliation, notable feedback and any changes made.

Final report. Delivery evidence in full, activity by position and by day, outcome data where available, qualitative themes, photographic documentation, and an honest section on what did not work.

That last part matters. A final report with no problems in it is not a report, it is a sales document. Every campaign has positions that underperformed and days that were poor. A report that names them is more useful and considerably more credible.

Benchmark: full photographic coverage of positions and shifts, not a curated selection.

Crew and conditions, the context that makes numbers readable

A report is uninterpretable without the operational context, and this is where a field-written note beats any dashboard.

Shift timing follows the audience. Transit entrances peak at commute hours, entertainment districts in the evening, campus edges on class timetables, venues on doors and let-out. A position worked at the wrong hour produces poor numbers that say nothing about the campaign.

Team structure affects output. Pairs are approached more readily than individuals and sustain better. A named lead has to be physically present, handling rotation, replenishment and early correction of approach problems. Shift length matters because performance degrades with time on feet, so a long shift shows a strong first half and a weak second, and the report should show that curve rather than an average that hides it.

Weather drives the numbers hard. Rain thins foot traffic and removes the eye contact the format depends on, and people will not take anything that gets wet. Cold shortens every interaction and degrades staff quickly, because standing still in cold is far harder than walking in it. Direct midday sun in a hot market is punishing and produces poor interactions as well as a welfare issue. Wind ruins printed material and anything freestanding.

None of that is in the numbers. All of it belongs in the report beside them.

Permission context belongs there too. Rules for distributing, occupying space and using sound vary by municipality and enforcement is inconsistent, so it stays a grey area rather than a settled question. Where a position had to be moved because a property controller asked, that should be in the report as a logged event rather than an unexplained gap. We work from positions where the controlling party has given documented consent, and when a position changes, the record says why.

What we talk clients out of

Making interaction count the headline KPI. It is the softest number in the report and the easiest to inflate, and making it the measure of success guarantees it gets inflated. Anchor on stock reconciliation and system-captured events instead.

Asking the team to track too much. Every extra field degrades all the others. Three tight metrics beat eight loose ones.

Real-time dashboards nobody acts on. A live feed is only worth building if somebody is empowered to move a position based on it. Otherwise it is an anxiety generator, and I would rather spend the effort on same-day shift reports with recommendations attached.

Judging ambassadors purely on output. Position quality dominates individual performance. Someone on a dead corner will underperform someone mediocre on a busy one. Compare like with like, and rotate positions so the comparison is fair.

Weekly reporting on a two-week campaign. By the time you read it, it is over. Daily, or do not bother.

Reporting that hides the bad days. It destroys the credibility of everything else in the document.

Running ambassadors with no supporting presence and then measuring them harshly. A team working a corridor cold has to build recognition from nothing in a three-second approach. The same team in a corridor that has been carrying Wheatpasting & Poster Campaigns for a fortnight is talking to people who have already seen the brand several times that week, and the difference in reception shows up clearly in the numbers. If the reporting is going to judge the team, give the team a fair setup first.

A reporting standard worth asking for

Short list to put in front of any operator before booking.

Timestamped, location-stamped check-in and check-out per person per shift.

Position-level breakdown of hours staffed against hours planned.

Stock issued, returned and reconciled per shift.

At least one system-captured activity or outcome metric, agreed in advance.

Tightly written definitions for any human-counted metric.

Full photographic coverage, timestamped, spread through the shift.

A same-day written shift report from the on-site lead, including conditions and verbatim feedback.

A final report that names the positions and days that underperformed.

An operator who can commit to all eight is running a real measurement process. One who offers a big interaction number and a photo gallery is not.

Frequently asked questions

What metrics actually matter in brand ambassador reporting?

Anchor on things that are hard to inflate: stock reconciliation, which constrains distribution claims, and system-captured events like code issues, scans and sign-ups. Treat interaction counts as soft estimates of a loosely defined thing. Collect refusals, which are rarely gathered and highly actionable. Qualitative verbatim feedback is the most undervalued output in the format.

Why is self-reported data a problem in this format?

Because almost all activity data is recorded by the people whose performance it measures, usually on a phone while they are busy. That is structural rather than a character issue. The fix is designing capture so it is instant, moving as many metrics as possible into systems that record themselves, and cross-checking claims against stock reconciliation and timestamped photography.

How often should we receive reports during a campaign?

Same day, per shift. The purpose of in-campaign reporting is intervention, not record keeping. A position that is not working should be moved on day two rather than noted in a final report weeks later. Weekly reporting on a short campaign arrives after the campaign is effectively over and cannot change anything.

How do we connect street activity to actual business outcomes?

Design the mechanism before the first shift and give it an owner on the client side. Unique codes issued in person and redeemed later give an attributable chain. Distinct short landing paths work if they are memorable. Venue footfall against a baseline suits footfall objectives. Market-level comparison against a similar untreated market suits awareness objectives.

What are the warning signs of an unreliable report?

Round numbers, since real tallies are untidy. Output that is suspiciously consistent across different people, days and weather. Interaction counts that exceed what the stock reconciliation supports. Totals with no position-level breakdown. A small curated photo selection presented as coverage. No conditions context. And a final report containing no problems at all.

Should we judge individual ambassadors on their numbers?

Only when comparing like with like. Position quality dominates individual performance, so someone on a dead corner will underperform a weaker ambassador on a busy one. Rotate positions through the shift so comparisons are fair, and read output against the conditions log, since weather and timing move the numbers substantially on their own.

Request a campaign quote

If you are commissioning brand ambassador activity and want reporting you can actually interrogate, tell us the market, the locations, the objective, the dates and what decision the reporting needs to support.

We will come back with a measurement design, metric definitions, a capture method, a documentation standard, a reporting cadence built for intervention rather than record keeping, and an honest view of which numbers will be solid and which will only ever be estimates.

Request a campaign quote and we will agree how it gets measured before anybody stands on a corner.

Ready to Run Your Campaign?

Call us or email us. We’ll tell you exactly what we can do in your market and what it costs.

Related Services

Explore similar advertising solutions offered by American Guerrilla Marketing.

American Guerrilla Marketing β€” Los Angeles

β˜…β˜…β˜…β˜…β˜… 5.0 Β· 34 Google reviews

Street-level campaigns in Los Angeles and nationwide. Wheatpasting, LED trucks, street teams, and more.

(646) 776-2770