[Part 1] Implementing "Effectiveness Measurement" for Municipal Branding Projects: A Practical Record from the Tachiarai Town Advisor

Introduction

Hello. I am Ryu Yoshida, and I provide the data-driven consulting service "Engine".

From November 2025 to March 2026, I served as a "Branding Project Effectiveness Measurement Advisor" for Tachiarai Town, Fukuoka Prefecture.

My involvement was fully remote, consisting of one-hour online meetings per week, totaling four hours per month. Even so, over five months, I was able to support the entire process from KPI design to questionnaire design and analysis, as well as the development of a KPI management manual.

I am writing this article for municipal DX and data utilization personnel as a record of the practical work involved in launching "branding project effectiveness measurement" from scratch. In this first part, I will cover everything from project launch and inventorying existing data to the philosophy behind questionnaire design. In the second part, I plan to write about the details of the questionnaire analysis, the KPI management manual, and operational design.

I will write as concretely as possible, rather than focusing on abstract theory. I hope this will be a useful reference for municipal staff facing similar challenges.

I have written about my personal reflections in this article, so please be sure to read it as well.

1. The Starting Point: "It's a Good Initiative, But Hard to Explain"

Tachiarai Town has set "Improvement of Regional Brand Power and Promotion of Town Promotion" as Measure 32 of its 5th Comprehensive Plan (2019–2028). The desired state is clear.

This has been promoted through three pillars: "Awareness," "Involvement," and "Attachment/Pride." There are initiatives that have been built up over about 10 years, such as SNS outreach, media exposure, PR events, the Support Ambassador system, the Edamame Harvest Festival, and the Lettuce Festa.

In fact, results have been achieved. In the 2023 "Town Happiness Ranking," it was number one in Kyushu. The population is currently hitting record highs.

Even so, the challenge the staff on the ground faced was this:

"We feel the impact, but it's hard to explain with numbers."

Reporting to the town council, explaining to residents, and justifying the budget for the next fiscal year. In all these situations where you cannot fight with "gut feelings," the data backing is weak. When asked about cost-effectiveness, the situation was such that they had no choice but to rely on subjective explanations.

I don't think this is unique to Tachiarai Town. Many municipalities face the same structural challenges.

They are doing "good initiatives," but they cannot structurally explain that they are "good initiatives." If the person in charge changes, it becomes unclear what the achievements were.

My role was to introduce a"measurement mechanism"here.

2. The first thing I did was 'organizing before measuring'

When I received the request, there was a trap I was likely to fall into at the beginning.

'Just listing KPIs'

SNS followers, number of event participants, hometown tax donation amounts... if you collect indicators that look plausible, the appearance is tidy. But if you do that, you end up 'just measuring what can be measured.' The numbers keep increasing while the purpose of measuring and what constitutes success remain ambiguous...

So, in the first few meetings, before deciding on KPIs, I started by sharing questions like these.

• 'What is the purpose' of this branding project?

• What does a 'successful state' look like?

• Which objective is each current measure linked to?

Fortunately, the objectives and the three pillars of the 32 measures were already documented. So, it wasn't starting completely from scratch. However, when I tried to organize 'which indicators can measure each of the three pillars''which indicators can measure each of the three pillars', there were quite a few gaps.

I spent about the first month getting this aligned. It's steady work, but if you set KPIs without doing this, you will definitely go astray later.

3. The next thing I did was 'taking inventory of existing quantitative data'

Once the premises were aligned, the next thing I did wastaking inventory of existing data.

Before introducing a new system, you cannot decide on the next action without grasping 'what is already being measured and what is not.'

Specifically, I organized data such as the following.

• Number of event attendees: Participation records for events like the Edamame Harvest Festival and Lettuce Festa

• Population trends in Tachiarai Town and comparisons with surrounding areas: Social increase/decrease, age composition, and in-migration/out-migration

• Trends in SNS followers: Instagram, YouTube, Official LINE

• Furusato Nozei (Hometown Tax) amounts and number of registered supporters: Indicators of related population

• Media coverage and exposure track record: Press, TV, Web media

When I took stock of everything, two things became clear.

1. There were more "things captured as numbers" than I imagined

SNS numbers, event participant counts, and Furusato Nozei amounts. The records were properly maintained. This was data that the staff had been accumulating day by day. This was a significant asset.

2. The "crucial part" was blank

On the other hand, the essence of the branding project, which is 'how town residents and people in the related population actually feel'. Data on this was almost entirely blank.

Degree of attachment, intention to recommend, experience of introducing the town to others, and pride in the town. We were in a state where we had no material to speak in numbers about the condition that Measure 32 aims for: 'everyone takes pride in Tachiarai Town, and the number of people who have a connection to and attachment to the town is increasing.'

I don't think this is a problem unique to Tachiarai Town. In many municipalities, 'behavioral numbers' are captured, but 'sentiment numbers' are not. I feel this structure is quite common.

4. Deciding to design a survey

Since the stocktaking of quantitative data revealed that the 'sentiment numbers' were blank, as the next step, I proposed conducting a wide-area survey.

What I kept in mind here was that 'a survey must have a purpose'.

You cannot get meaningful data by 'just asking for the sake of it.' You need to decide first 'what you are asking in order to make what decision.'

The points I kept in mind during the design were as follows:

1. Asking about the "three pillars" from various angles

The three pillars of Measure 32 are 'awareness,' 'connection,' and 'attachment/pride,' but even if you ask residents using just these words, it is difficult to grasp them in a three-dimensional way.

Therefore, you need to translate them into questions that residents can understand in their own language. For example,

• "Pride" → "How likely are you to recommend Tachiarai Town to your family or friends?"

• "Interest" → "Do you feel that more people are involved with (or interested in) the town compared to 10 years ago?"

• "Awareness" → "How did you first learn about Tachiarai Town?"

If you do not make the questions natural for the respondents, the reliability of the data will decrease.

(2) Ask about each stage of the funnel separately

You cannot analyze it by simply asking, "What do you think of the town?" "I know it," "I have participated," "I have recommended it," "I have visited"—you should break down and ask about the stages from awareness to recommendation.

This allows you to later calculate the "conversion rate from awareness to action" and the "conversion rate from action to recommendation." Whether or not you can perform funnel analysis is almost entirely determined at the question design stage.

(3) Enable segment analysis by residence and age group

Ask by categorizing into "Town residents," "Within Fukuoka Prefecture," "Outside Fukuoka Prefecture," and "Overseas." Also, divide age groups from teens to 70s and older.

Because of this, you can analyze from perspectives such as "visit intentions of non-residents" or "awareness channels for teenagers." Even if the questions are the same, what you see changes completely when you slice the data by attributes.

(4) Always include NPS (Net Promoter Score)

Ask "Would you recommend Tachiarai Town to your family or friends?" on an 11-point scale.

With this, the difference in NPS between the 'contact group' and the 'non-contact group'based on whether they follow social media or have participated in events will become clear. You will be able to discuss the effectiveness of public relations and event measures using the most essential indicator: recommendation intention.

5. Conducting the survey

Once the design was finalized, we conducted a broad survey both inside and outside the town from the end of December through January. Thanks to the public relations efforts of the town office staff, we were ultimately able to collect responses from 1,288 people.

The breakdown of respondents is as follows.

• By place of residence: 60.1% living in Tachiarai Town, 26.4% within Fukuoka Prefecture, 12.7% outside Fukuoka Prefecture, and 0.9% overseas

• By age group: Teens were the largest group at 29.7% (due to collection via schools), followed by those in their 50s, 40s, and 60s

The high ratio of teenagers is due to cooperation from schools, so it is necessary to keep the bias in age composition in mind when interpreting the results. However, excluding teenagers, the data forms a pyramid centered on those in their 50s, meaning the opinions of the core generation in the region are strongly reflected.

A composition of 60% from within the town and 40% from outside (including outside the prefecture) can be said to be a composition that can reflect both internal town awareness and external evaluation to a certain extent.

I will write about the detailed content of the survey in the second part.

Summary of Part 1: What I kept in mind up to this point

In the first part, I wrote about the project launch, the inventory of quantitative data, and the survey design and implementation. To summarize, the flow is as follows.

1. Confirmation of premises — Aligning what the project is for, what success looks like, and linking measures to objectives

2. Inventory of existing quantitative data — Understanding what is being measured and what is not

3. Identification of blank areas — In the case of Tachiarai Town, "emotional metrics" were the blank area

4. Survey design — Creating questions with a clear purpose and a structure that allows for funnel analysis

5. Survey implementation — Collecting responses from 1,288 people

This is what we call the "data preparation phase."

In the second part, I will write about the current status of Tachiarai Town as revealed by analyzing the 1,288 survey responses, future challenges, and the KPI design and management manual development to address those challenges.