Weekly

Turn failed searches into synonym ideas

Failed queries arrive grouped by intent, with suggested synonyms for each group.

Runs: Weekly
Apps: Algolia
AI: Yes, uses the built-in AI allowance
Access: analytics ACL

Send this to your assistant

Create a ClickRuntime automation: every Friday, take last week's Algolia no-result queries, group them by what the shopper meant and email me synonym suggestions for each group.

Paste it into your assistant. It tells you what it will send or change before it runs live.

Add to ChatGPT Add to Claude

First time? Adding ClickRuntime to your assistant takes about two minutes.

Why use it

Why teams use this automation

A site search that returns nothing is a customer telling you what they wanted and could not find. On a shop that runs on Algolia, the no-results list can have hundreds of different queries a week: typos, brand names, regional words, plurals, things you do not sell. Read as a flat list, it is hard to see which ones matter. Merchandisers skip it, and the same gaps cost sales for months because nobody decided that sneakers and trainers should be the same thing.

Every week, the automation pulls the searches that returned no results, and ClickRuntime's built-in AI groups them by intent, for example misspellings, synonyms of products you sell and requests for items you do not carry. For each group it suggests synonyms or redirects you could add to Algolia. The result arrives as an email.

It is useful for e-commerce managers, merchandisers and anyone who owns the search experience of a site or app.

What you get out of it

Search gaps turn into a clear to-do list
Instead of 400 queries, you see six groups with an idea for each. A merchandiser can add a synonym or a rule in minutes, which turns lost searches into found products.
Vocabulary differences are caught
The AI notices that customers say couch where your catalog says sofa. Matching the words people really use is one of the cheapest ways to lift conversion. Adding the missing synonym brings those shoppers to the right products.
Demand you do not serve becomes visible
Groups of queries for products you do not sell are evidence for what to stock next. It is customer research that costs nothing to collect. You learn what to stock without running a survey.
Typos are separated from real gaps
Misspellings are grouped on their own, so you can decide whether to rely on typo tolerance or add rules, and avoid treating them as missing products. Each kind of problem has a different remedy, so sorting them saves effort.
Weekly attention keeps search improving
A small number of changes each week adds up. Over a quarter, the share of searches that end in nothing can fall noticeably, with no big project. That is a real gain for very little work each week.

How it works

What happens, step by step

Every week, ClickRuntime collects the searches on your shop from the past 7 days that found nothing in Algolia, with how often each was searched.
AI groups the failed searches by what shoppers were looking for, such as rain gear.
For each group it suggests synonyms that would link those searches to products you already sell.
You get an email listing each group with its search count, example searches and the suggested synonyms.

Before it runs live

Connect Algolia

What you get

RunsWeekly

Turn failed searches into synonym ideastoday, 9:00Succeeded

Email · Synonym ideasRain gear (126 searches): "rain mac", "waterproof jacket" Synonyms: rain mac ↔ raincoat, waterproof jacket ↔ rain jacket

Build your own

Need it to work differently?

This template is a starting point. Copy the prompt and change it, or describe a different automation from scratch, with your own conditions, more apps or an AI step. Your assistant writes it, tests it and runs it on ClickRuntime.

Your own rules
Conditions, thresholds, filters and exceptions, written the way you'd explain them to a colleague.
Any number of apps
One automation can read from one app and update several others. Over 248 apps are ready, and anything with an API works too.
AI where it helps
Summaries, classification, extraction and drafts use ClickRuntime's built-in AI, with no extra key.
On a schedule or on each event
Runs as often as every 5 minutes, or the moment an app sends an event, and keeps track of what it already handled.

Changes you can ask for

Run it at a different time, more or less often, or only on weekdays
Change what the AI looks for, or how it writes the result
Add another app, for example also send it to Slack
Combine it with other Algolia steps into one automation

Under the hood

The code that runs this automation

This is the Python script that runs each time the automation starts. Your AI assistant writes it for you when you send the prompt, and you allow each step in your assistant. To change how the automation works, describe the change to your assistant and it updates the script for you. For every function it can use, see the script reference.

ai_grouped_zero_results_with_synonym_ideas.pyexample

from datetime import date, timedelta
from clickruntime import on_trigger, credentials, http, ai, notify

INDEX = "products"

@on_trigger
def run(event):
    searches = http.get("https://analytics.algolia.com/2/searches/noResults",
                        headers={"X-Algolia-Application-Id": credentials.secret("ALGOLIA_APP_ID"),
                                 "X-Algolia-API-Key": credentials.secret("ALGOLIA_ANALYTICS_KEY")},
                        params={"index": INDEX, "limit": 200,
                                "startDate": (date.today() - timedelta(days=7)).isoformat(),
                                "endDate": date.today().isoformat()}).json()["searches"]
    if not searches:
        return {"queries": 0}
    queries = "\n".join(f"{s['search']} ({s['count']})" for s in searches)
    report = ai.generate(f"Shop search queries that returned nothing last week:\n{queries}",
                         system="Group these queries by what the shopper meant. For each group give the queries, "
                                "total count and 1-3 synonym rules to add in Algolia. Plain text.",
                         max_output_tokens=1200).text
    notify.email_me("No-result searches grouped, with synonym ideas", report)
    return {"queries": len(searches)}

FAQ

Common questions

Does the AI change my Algolia settings?

No. It only suggests synonyms and groups, in an email. A person decides which ones to add in the Algolia dashboard, so your search index is not changed automatically. You stay in control of the index at all times.

How many queries does it look at?

It takes the no-results searches returned for the past week, which for most sites is a manageable list. Rare queries that appear once are included but are usually summarized, not listed one by one. Very rare queries are summarized, since a single search rarely justifies a rule.

What Algolia data does it need?

Access to the Analytics API, which provides the no-results searches for an index. The API key is stored as a secret in ClickRuntime, and the key only needs analytics read access. Nothing else in your Algolia account is touched.

What permissions does it need?

analytics ACL. Nothing else in your accounts is used.

Does it need AI?

Yes. It uses ClickRuntime's built-in AI, so you don't need an OpenAI or Anthropic key. Every plan includes a monthly AI allowance.

How many runs does it use?

One run each time it runs (weekly). Checks that find nothing to do and stop right away aren't counted.

Can I change how it works?

Yes. Describe the change to your AI assistant, for example a different schedule, channel or filter. It updates the automation, tests it and tells you what changed.

Weekly