A frequent doubt I notice from newcomers in analytical chemistry is:

“What is a good recovery ratio for my internal standard?”

Let’s use the following example with Iron (Fe) as the analyte and Yttrium (Y) as the internal standard:

Sample	Fe [cps]	Y [cps]	Ratio	Recovery
1	70287	76019	1.00	100%
2	71040	77285	0.98	98%
3	70533	75944	1.00	100%
4	72763	78583	0.97	97%
5	71080	76120	1.00	100%
6	71275	76862	0.99	99%
7	72351	77327	0.98	98%
8	70893	76597	0.99	99%
9	70901	76140	1.00	100%
10	68757	74204	1.02	102%

Table meaning:

• Fe [cps] – signal of analyte in counts per second.
• Y [cps] – signal of internal standard in counts per second.
• Ratio – (Y [cps] of Sample 1) / (Y [cps] of sample).
• Recovery – Ratio × 100.

---

Vendor Recommendations

Vendors often state that an acceptable ratio is between 0.8 – 1.2.
But—if no one explains why, take the advice with caution.

---

Why You Need Your Own Definition

When developing a method, your internal standard must show the expected behaviour:

• Does its signal vary in the same way as your analyte?
  Tip: plot a linear regression to check correlation.

  

• What recovery range works for your method?

  • Spike certified solutions with substances that interfere and observe bias.
  • Decide how much bias is tolerable before it impacts decisions—scientific or economic.

---

In Short, During Method Development:

• Find a suitable internal standard.
• Understand its behaviour in your real sample matrix.
• Define and justify your own acceptable recovery range.