Track Citation Counts for Research Impact Over Time

TL;DR — The Google Scholar Scraper returns citedBy per paper as JSON. Query your paper list on a schedule, store snapshots, and citation growth becomes a measured series — for tenure files, grant reports, and lab dashboards.

Why citation tracking beats checking Scholar by hand

Impact evidence arrives on a deadline: a tenure packet, a progress report, a grant renewal. "I checked Scholar" isn't a series — a dated count per paper is.

The job-to-be-done is refreshable snapshots: same queries, same fields, new counts — joined by title into a growth table.

How does this compare to the alternatives?

Scopus and Web of Science count fewer, cleaner citations. Scholar counts more, sooner — for trend reporting, breadth is the feature.

How to track citations in 4 steps

How do I snapshot a paper list?

import os
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
titles = [
    '"Attention Is All You Need"',
    '"BERT: Pre-training of Deep Bidirectional Transformers"',
]
run = client.actor("thirdwatch/google-scholar-scraper").call(
    run_input={"queries": titles, "maxResults": 5}
)
rows = list(client.dataset(run["defaultDatasetId"]).iterate_items())

Quote the title — position 1 is almost always the target paper.

How do I store the series?

import json, datetime, pathlib
snap = {"date": datetime.date.today().isoformat(),
        "counts": {r["title"]: r["citedBy"] for r in rows}}
pathlib.Path("citations.jsonl").open("a").write(json.dumps(snap) + "\n")

Weekly appends build the series with zero rework.

How do I chart growth?

import pandas as pd
snaps = pd.read_json("citations.jsonl", lines=True)
growth = pd.json_normalize(snaps.set_index("date")["counts"]).T
growth.columns = snaps["date"]
print(growth.assign(delta=growth.iloc[:, -1] - growth.iloc[:, 0]))

How do I scope by year?

fromYear/toYear bound the search — useful when the same title exists across editions and you want the canonical recent version.

Sample output

{"title": "Attention Is All You Need",
 "url": "https://arxiv.org/abs/1706.03762",
 "authors": "A Vaswani, N Shazeer", "venue": "NeurIPS",
 "year": "2017", "citedBy": 130000,
 "pdfUrl": "https://arxiv.org/pdf/1706.03762",
 "query": "\"Attention Is All You Need\"", "position": 1}

Common pitfalls

Scholar merges versions — citedBy is the union across arXiv, proceedings, and journal forms, which is exactly why trends work well. Homonymous titles occasionally return the wrong paper at position 1; sanity-check authors. Counts update on Scholar's schedule, not yours — compare weekly, not daily. The actor reads public result pages; volume modesty keeps it reliable.

Related use cases

• Scrape Google Scholar for a literature review
• Monitor new papers in your research field
• Build an arXiv research dataset
• Blog hub