The Camino Crowd Index

Name: The Camino Crowd Index
Creator: WalkSage
License: https://creativecommons.org/licenses/by/4.0/

Every guidebook tells you the Camino gets busy. None of them tell you where, in numbers you can check. So we built an index: 54 towns across 2 routes, ranked by relative visitor interest and computed from a 12-month window of Wikipedia pageviews. It is free to download and reuse.

Download CSV Download JSON

Released under CC BY 4.0 — reuse it anywhere, including commercially, with credit to WalkSage and a link to this page. Upstream sources and their licences are listed on data & sources.

What it shows

Crowding on the Camino de Santiago Francés is not spread along the route — it is concentrated in a handful of cities, and it collapses to almost nothing in between. The busiest town scores 1.000; the quietest measured town scores 0.003. That is a spread of more than two orders of magnitude between stops on the same walk.

Busiest

Santiago de Compostela 1.000
Pamplona 0.928
Burgos 0.910
León 0.824
Ponferrada 0.669

Quietest measured

O Pedrouzo 0.003
Hontanas 0.010
San Martín del Camino 0.011
Agés 0.013
Terradillos de los Templarios 0.017

Method

For every town on a route we take the town's Wikipedia article, pull its monthly pageview counts over a fixed 12-month window from the Wikimedia REST API, and average them. The Crowd Index is that average divided by the highest average on the same route, so the busiest town scores 1.0 and everything else is a fraction of it. Towns are reconciled to Wikidata items first, which is how we know we are counting the right article.

Camino de Santiago Francés: crowd_index.normalized = town monthly_avg pageviews / max monthly_avg across all 33-route towns (same 12-full-month window, agent=user, all-access; es.wikipedia preferred, else en, else fr — 'project' records which). Deterministic and reproducible from the monthly_views arrays stored per town. Proxy for relative visitor interest, not a measured footfall count. Window: 20250801-20260731. Computed 2026-08-02.

Camino Portugués: crowd_index.normalized = town monthly_avg pageviews / max monthly_avg across all 20 Camino Portugués towns (12-full-month window 20250801-20260731 — the same window as the Camino Francés run — agent=user, all-access). Project selection is country-local first: pt.wikipedia for Portuguese towns, es.wikipedia for Spanish towns, else en.wikipedia, else gl.wikipedia; 'project' records which was used. Deterministic and reproducible from the monthly_views arrays stored per town. Proxy for relative visitor interest, not a measured footfall count. Window: 20250801-20260731. Computed 2026-08-04.

The calculation is deterministic: the same window and the same articles produce the same numbers, and the per-town monthly series is stored so anyone can recompute it.

What this does not measure

Reading about a place is not visiting it. Pageviews measure attention, which correlates with visitor interest but is not the same thing. A town with a famous battle in its history draws readers who will never walk through it.
Big towns are big for reasons unrelated to the Camino. León and Burgos are cities with their own tourism, industry and news cycles. Their scores reflect all of it, not just pilgrims.
The index is normalised per route. Each route's busiest town is 1.0, so a 0.5 on one route is not a 0.5 on another. To compare routes, use the raw monthly pageview column, which uses an identical window and identical API settings everywhere.
Cross-language counts are not perfectly comparable. Where a route spans two countries we read the country-local Wikipedia first, and those projects have different-sized readerships. That is why a second index computed on English Wikipedia alone is published alongside, as a cross-check.
It has no seasonality. These are 12-month averages. They tell you which towns are busy relative to each other, not which month to walk.
Some towns cannot be measured at all. A hamlet with no Wikipedia article has no series. Those rows are kept and marked, never quietly dropped.

The full data

Every town, every measure, nothing hidden. Rows marked no data are towns the method could not reach.

Camino de Santiago Francés

34 towns measured. Most crowded first. See the Camino Francés guides.

#	Town	Crowd Index	Monthly avg. pageviews	Population
1	Santiago de Compostela	1.000	15,786.5	100,965
2	Pamplona	0.928	14,643.1	209,094
3	Burgos	0.910	14,360.5	177,402
4	León	0.824	13,000.7	123,446
5	Ponferrada	0.669	10,566.2	63,186
6	Logroño	0.620	9,781.7	152,150
7	Astorga	0.467	7,373.8	10,305
8	Santo Domingo de la Calzada	0.406	6,417.2	6,441
9	Estella	0.305	4,821.6	14,317
10	Carrión de los Condes	0.271	4,285.2	1,997
11	Nájera	0.246	3,889.2	8,336
12	Sarria	0.230	3,626.4	13,726
13	Frómista	0.193	3,041.3	727
14	Saint-Jean-Pied-de-Port	0.176	2,776.0	1,479
15	Roncesvalles	0.164	2,595.2	25
16	Villafranca del Bierzo	0.152	2,394.3	2,591
17	Puente la Reina	0.142	2,246.4	2,972
18	Belorado	0.126	1,991.6	1,823
19	Castrojeriz	0.124	1,959.2	794
20	Arzúa	0.089	1,401.7	5,916
21	Palas de Rei	0.085	1,342.2	3,227
22	Portomarín	0.073	1,158.8	1,277
23	Mansilla de las Mulas	0.071	1,115.7	1,706
24	O Cebreiro	0.059	937.2	125
25	Los Arcos	0.049	774.3	1,177
26	Zubiri	0.035	549.3	491
27	Triacastela	0.029	452.6	589
28	El Burgo Ranero	0.026	415.2	697
29	Rabanal del Camino	0.018	283.9	74
30	Terradillos de los Templarios	0.017	263.4	56
31	Agés	0.013	198.2	56
32	San Martín del Camino	0.011	169.1	354
33	Hontanas	0.010	162.2	69
34	O Pedrouzo	0.003	49.7	—

Camino Portugués

19 towns measured, 1 with no data. Most crowded first. See the Camino Portugués guides.

NORMALISATION IS CAMPAIGN-LOCAL: crowd_index is scaled to the busiest town of THIS route (Santiago de Compostela ~1.0), so a Camino Portugués index is NOT comparable to a Camino Francés index. To compare the two campaigns, use raw_monthly_avg_pageviews, which is stored for every town and uses an identical 12-month window (20250801-20260731) and identical API settings on both routes. SECOND CAVEAT: this route spans two countries, so primary counts come from different Wikipedia projects (pt vs es), whose reader populations differ in size; counts are not perfectly comparable across projects. Each town therefore also stores wikipedia_pageviews.all_projects (per-project monthly averages) and crowd_index.cross_check_enwiki, a second index computed on en.wikipedia alone — one common project for every town, which removes the cross-project bias. Where the two indices disagree, say so.

#	Town	Country	Crowd Index	Monthly avg. pageviews	en.wiki cross-check	Population
1	Santiago de Compostela	Spain	1.000	15,786.5	0.751	100,965
2	Vigo	Spain	0.761	12,008.3	0.413	294,489
3	Pontevedra	Spain	0.651	10,283.6	0.131	83,316
4	Porto	Portugal	0.569	8,983.8	1.000	231,800
5	Baiona	Spain	0.267	4,208.2	0.028	12,361
6	Tui	Spain	0.262	4,129.1	0.031	17,468
7	Viana do Castelo	Portugal	0.255	4,017.6	0.098	88,725
8	Barcelos	Portugal	0.170	2,684.0	0.071	116,752
9	Ponte de Lima	Portugal	0.168	2,648.6	0.047	43,498
10	Padrón	Spain	0.146	2,307.7	0.025	8,334
11	Póvoa de Varzim	Portugal	0.145	2,292.6	0.060	63,408
12	Vila do Conde	Portugal	0.140	2,217.2	0.071	79,533
13	Redondela	Spain	0.134	2,118.3	0.017	28,874
14	O Porriño	Spain	0.132	2,088.9	0.018	20,965
15	Caldas de Reis	Spain	0.127	2,011.8	0.013	9,615
16	A Guarda	Spain	0.127	1,998.8	0.009	9,970
17	Esposende	Portugal	0.086	1,356.6	0.015	34,254
18	Caminha	Portugal	0.064	1,002.9	0.019	16,684
19	Rubiães	Portugal	0.006	88.9	0.002	493
—	Vilarinho	—	no data	—	—	—

Towns shown as no data have no Wikidata item, so no pageview series exists for them. They are listed rather than dropped — an unmeasurable town is a finding, not a blank. The reasons travel with each row in the downloadable data.

Citing this

If you use the index, please credit WalkSage and link to this page — that is the whole licence condition. A suggested form:

Journalists and researchers: the per-town monthly pageview series behind these averages exists and can be shared. The method is deliberately simple so that anyone can reproduce or challenge it.