# What changed our advice

Keep the results that challenged the original idea visible.

Author: Murali Sid (https://linkedin.com/in/muralisid)
Source: https://www.agenticarchitectureskills.com/research/what-changed
Published: 2026-09-17
Updated: 2026-10-15
Licence: CC BY-SA 4.0, https://creativecommons.org/licenses/by-sa/4.0/
Cite as: Murali Sid. What changed our advice. Agentic Architecture Skills. https://www.agenticarchitectureskills.com/research/what-changed

---


A useful research program records when a more elaborate approach did not help. These changes come from the existing author-reported context experiments; their scope matters.

## More representations were not a universal improvement [#more-representations-were-not-a-universal-improvement]

The reported experiments found settings where multiple views helped and settings where matched chunks performed better. The advice is to start with a matched baseline and test additional views on the actual questions. See [real-prose comparisons](/patterns/real-prose).

## The objective did not reliably improve the index [#the-objective-did-not-reliably-improve-the-index]

The tested forms of objective-conditioned representation did not support the expected improvement. The guide therefore does not recommend treating the business objective as an automatic index-design signal. See [the comparisons and limits](/patterns/why-not-to-condition-on-the-objective).

## Retrieval is only part of the outcome [#retrieval-is-only-part-of-the-outcome]

The consuming model and the composition of the retrieved set affect what happens after search. A retrieval score is not a complete measure of task success. See [diversity and the consumer](/patterns/diversity-and-consumer).

## New media are new hypotheses [#new-media-are-new-hypotheses]

These text results neither prove nor disprove a particular satellite, image, or time-series representation. Evaluate those on their own data and questions, with exact source metadata preserved.

Read [what we got wrong](/patterns/what-we-got-wrong) and [the research method](/research/method).