Ronnie Winsted applied for disability insurance benefits and supplemental security *468income claiming disability based on numerous conditions, including degenerative disc disease, osteoarthritis, and anxiety. An administrative law judge denied benefits, finding that Winsted could work with certain limitations. After the district court upheld this denial, Winsted appealed, arguing the ALJ did not consider his difficulties with concentration, persistence, and pace. We agree-the ALJ did not adequately explain how the limitations he placed on Winsted's residual functional capacity accounted for the claimant's mental difficulties, so we remand to the agency.
I. Background
Winsted was 42 years old when he applied for benefits, asserting an onset date of October 2010. Although he initially alleged he became disabled in 2005, two prior applications alleging this onset date were denied and deemed administratively final.
Winsted suffers from multiple physical impairments, mostly associated with his previous work in hard labor as an industrial truck driver, a highway maintenance worker, and an operating engineer. MRIs taken in 2010 and 2011 showed he had focal, isolated degenerative disc disease. Other tests revealed osteoarthritis, mild carpal tunnel syndrome in his hands, and cavus (high-arched) foot that he treats with special shoes.
Winsted complained of shortness of breath in May 2011 and was diagnosed with acute bronchitis and chronic obstructive pulmonary disease ("COPD"). Although he wheezed at times, he often responded well to medication. Throughout the relevant period, Winsted sometimes complained of wheezing, but often his lungs were clear. A pulmonary function test in 2013, however, showed Winsted had moderate obstructive lung disease and possibly restrictive lung disease.
Winsted began seeing an internist, Dr. Nedu Gopala, for back pain in August 2013. The doctor prescribed medication for Winsted's breathing, chest pain, back pain, and anxiety. At appointments throughout 2013 and into March 2015, Winsted's range of motion in his arms and legs alternated from full, to limited. He maintained a chronic cough, mild shortness of breath, and wheezing, though a 2014 pulmonary function test did not show any evidence of lung obstruction.
To address stress-related heart issues, Winsted sought mental-health treatment in 2012. A therapist diagnosed him with a panic disorder, post-traumatic stress disorder, and major depressive disorder. Winsted had a guarded attitude, "very little insight," "below average" intellect, and was assigned a Global Assessment of Function ("GAF") of 51, indicating he had moderate difficulty in social and occupational functioning.1 AM. PSYCHIATRIC ASS'N, DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS 32 (4th ed. 1994). In his therapy appointments, Winsted regularly complained about altercations with neighbors.
Later that year, Winsted sought treatment from a psychiatrist, who diagnosed major depressive disorder and assigned a GAF of 45, indicating a serious impairment in social or occupational functioning. AM. PSYCHIATRIC ASS'N , supra . The psychiatrist reported that Winsted was tense, anxious, "very restless," and moderately depressed.
*469He prescribed medication for anxiety and depression and continued to treat Winsted.
Between September 2013 and February 2015, Winsted's mental health fluctuated. In September 2013, Winsted's psychiatrist reported that his affect was appropriate, his mood was not depressed, and "on the whole [he was] doing better." But two months later, Winsted's affect was anxious, his mood was depressed, he was "feeling more irritable, anxious, and restless," and he suffered panic attacks. At a therapy session in July 2014, a therapist reported Winsted's "symptoms of depression and worry impair overall functioning," and in August and November 2014, he was "mildly depressed." But in February 2015, Winsted presented with an appropriate affect and a not-depressed mood. The same was true in April 2015, though Winsted reported he sometimes felt "tense and anxious" and stress continued to cause him to "become overwhelmed."
The disability application also triggered an examination in 2013 from an agency psychologist, Dr. Steven Marlow, who diagnosed Winsted with major depressive disorder, a generalized anxiety disorder, and a panic disorder. Specifically, he reported Winsted "has a[n] avoidant, hostile, and easily distracted attitude." Dr. Marlow determined Winsted had below average levels of mental control, understanding and memory, and concentration; poor levels of persistence; and he did not do well in social situations.
A state-agency physician, Dr. George Siderys, also examined Winsted in 2013 and opined he had a mild functional impairment. This included: "mild decrease in range of motion," pain that would be "expected to cause him problems with prolonged standing, walking, or heavy lifting," and a history of heart difficulties that would cause him to "wear out if he participated in prolonged walking or lifting."
In connection with Winsted's disability claim, treating physician Dr. Gopala completed a physical residual functional capacity ("RFC") questionnaire in early 2015 and reported Winsted suffered from hypertension, COPD, and back pain, and described Winsted's prognosis for back pain as "poor." He determined Winsted had a "painful range of movement" and was incapable of performing even "low stress" work. Dr. Gopala also wrote that Winsted's symptoms would affect his attention and concentration frequently; he could walk only about one block; and he could sit or stand for only 15 minutes at a time.
In April 2015, treating therapist Jessica Nevill filled out a mental RFC questionnaire. She opined Winsted had marked impairments in his abilities to: relate to other people, respond to supervision, respond to work pressures, and respond appropriately to changes in the work setting. She wrote Winsted would miss work three to four days per month because of his impairments.
After the Social Security Administration denied Winsted's application, he had a hearing before an ALJ. Winsted testified he used an inhaler twice a day, slept with a CPAP machine, used a nebulizer for breathing every three months, and continued to smoke a half-a-pack of cigarettes per day. He said he could not grip a two-liter bottle with his left hand. Due to the pain in his knees and feet, he said he could stand for only a few minutes and, even then, he could not stand still. He also said he could walk only a few blocks before needing to stop and catch his breath, and experienced chest pain three to four times per week. He noted he has trouble getting along with people and does not like to be around groups.
After Winsted testified, the ALJ asked the vocational expert ("VE") three hypothetical *470questions. First, the ALJ asked the VE to consider an individual of the same age, education, and work experience as Winsted. He continued:
This hypothetical individual would be capable of light work, but four hours maximum standing and walking in an eight hour day, only occasional climbing of ramps, and stairs, but no ropes, ladders, or scaffolds, only occasional balancing, stooping, kneeling, crouching, and crawling. Frequent, but not constant handling and fingering bilaterally. This individual would need to avoid concentrated exposure to breathing irritants, such as fumes, orders, dust, and gasses, as well as wet, slippery surfaces, and unprotected heights and would further be limited to only simply reaching, repetitive tasks, with few workplace changes, no team work, and no interactions with the public.
The expert determined such a person could work as a bench assembler, electronics worker, or production assembler. In the second hypothetical, the ALJ asked about an individual with the same limitations as in the first hypothetical, but who also "due to impair-related symptoms, such as the need to lay down during the day to relieve pain would be off task 20% of the work day." The VE replied that such an individual could not sustain employment. Finally, the ALJ asked about a person with all the same limitations as provided in the first hypothetical, "but due to the frequency of bad days versus good days, this individual would have two unscheduled absences per month." Again, the VE answered, "there would be no jobs."
The ALJ conducted the Administration's 5-step analysis, see 20 C.F.R. § 404.1520(a), § 416.920(a), and found Winsted not disabled. At Step 1 the ALJ determined Winsted had not engaged in substantial gainful activity since October 22, 2010. At Step 2 the ALJ identified Winsted's severe impairments as degenerative disc disease of the lumbar spine, bilateral carpal tunnel syndrome, osteoarthritis in his left knee, bilateral cavus foot, COPD, obstructive sleep apnea, obesity, an affective disorder, and an anxiety disorder. At Step 3 the ALJ acknowledged Winsted had moderate difficulty with social functioning and concentration, persistence, and pace because of his mental-health issues, but concluded these severe impairments did not meet a listing for presumptive disability. Between Steps 3 and 4 the ALJ determined Winsted had the requisite RFC to perform light work with certain limitations (as provided in the first hypothetical, and including being limited to "simple, routine, repetitive tasks with few workplace changes, no team work, and no interaction with the public") but his limitations precluded him from performing his past relevant work (Step 4). At Step 5 the ALJ concluded, based on Winsted's age, education, work experience, and RFC, that he was capable of successfully changing to other work.
Winsted appealed to the agency's Appeals Council, which denied review. He then sought judicial review, and the parties agreed to have a magistrate judge adjudicate this case. See 28 U.S.C. § 636(c). That judge upheld the ALJ's decision.
II. Analysis
A. ALJ's Evaluation of Winsted's Limitations in Concentration, Persistence, and Pace
Winsted argues neither the ALJ's RFC nor his first hypothetical question properly accounted for the finding that he has "moderate" difficulties with concentration, persistence, and pace. He submits the ALJ's proposed limitations-that Winsted perform only "simple, routine, repetitive tasks with few workplace changes"-fail to *471address his concentration-functioning deficits because "both the hypothetical posed to the VE and the ALJ's RFC assessment must incorporate all of the claimant's limitations supported by the medical record." Varga v. Colvin , 794 F.3d 809, 813 (7th Cir. 2015).
Winsted's argument here is correct. Again and again, we have said that when an ALJ finds there are documented limitations of concentration, persistence, and pace, the hypothetical question presented to the VE must account for these limitations. Moreno v. Berryhill , 882 F.3d 722, 730 (7th Cir. 2018) ; Varga , 794 F.3d at 814-15 ; O'Connor-Spinner v. Astrue , 627 F.3d 614, 620 (7th Cir. 2010) ; Stewart v. Astrue , 561 F.3d 679, 684 (7th Cir. 2009) ; Kasarsky v. Barnhart, 335 F.3d 539, 544 (7th Cir. 2003) ; see also Young v. Barnhart , 362 F.3d 995, 1003 (7th Cir. 2004). We have also made clear that in most cases "employing terms like 'simple, repetitive tasks' on their own will not necessarily exclude from the VE's consideration those positions that present significant problems of concentration, persistence and pace," and thus, alone, are insufficient to present the claimant's limitations in this area. O'Connor-Spinner , 627 F.3d at 620 ; see Moreno , 882 F.3d at 730. Here, at Step 3 the ALJ found Winsted's moderate difficulties with concentration, persistence, and pace could cause problems with concentration and following written instructions, as well as stress with changes in his routine. And Winsted's psychiatrist and therapist both remarked that stress caused Winsted to "become overwhelmed" and his depression impaired his overall functioning.
But the first hypothetical the ALJ posed to the VE did not direct the expert to consider problems with concentration, persistence, and pace, which is the hypothetical the ALJ relied on for the RFC. Though particular words need not be incanted, we cannot look at the absence of the phrase "moderate difficulties with concentration, persistence, and pace" and feel confident this limitation was properly incorporated in the RFC and in the hypothetical question. See O'Connor-Spinner , 627 F.3d at 619. The ALJ may have thought, as the agency proposes, he was addressing Winsted's concentration difficulties by including limitations that would minimize social interaction. But that restriction could just have likely been meant to account for Winsted's moderate difficulty with social functioning-the ALJ acknowledged Winsted experiences anxiety, panic attacks, and irritability when he is around people . Nothing in the hypothetical question and the RFC, however, accounted for the ALJ's discussion of how Winsted's low GAF scores reflect serious mental-health symptoms or his mention that Winsted often "appeared tense, anxious, and/or restless" without interacting with other people . Additionally, where a claimant's limitations are stress-related, as Winsted's appear to be, the hypothetical question should account for the level of stress a claimant can handle. See Arnold v. Barnhart , 473 F.3d 816, 820, 823 (7th Cir. 2007) ; Johansen v. Barnhart , 314 F.3d 283, 285, 288-89 (7th Cir. 2002). But there was no restriction related to stress in the RFC or hypothetical question.
Notably, it appears the ALJ disregarded testimony from the VE about a person with limitations in concentration, persistence, and pace. The ALJ asked two additional hypothetical questions of the VE about an individual who would either be off task 20% of the workday or would have two unscheduled absences per month-presumably assuming someone with "moderate difficulties with concentration, persistence, and pace." The VE responded that neither individual could sustain employment. But these responses are not reflected *472in the ALJ's decision, which means it cannot stand.
B. ALJ's Evaluation of the Medical Opinion Evidence
Winsted also challenges the evidentiary weight the ALJ gave to four medical opinions, two from treating medical professionals (Dr. Gopala and Ms. Nevill) and two from state agency doctors (examining psychologist Dr. Marlow and consultative examiner Dr. Siderys).
Before reaching the merits of this argument, we must address the agency's contention that Winsted waived it. The agency is not correct on this; Winsted never "knowingly and intelligently relinquished" his claim, Wood v. Milyard , 566 U.S. 463, 470 n.4, 132 S.Ct. 1826, 182 L.Ed.2d 733 (2012). That he developed the argument poorly means at most he forfeited it. Brown v. Colvin , 845 F.3d 247, 254 (7th Cir. 2016).
Forfeited or not, this argument fails. In the decision, the ALJ adequately articulated why he gave each opinion the weight he did, entitling his decision, in this respect, to our deference. See Elder v. Astrue , 529 F.3d 408, 413, 416 (7th Cir. 2008). Starting with Dr. Gopala, the ALJ appropriately questioned the doctor's conclusion-that Winsted had a painful range of motion that made him incapable of engaging in "low-stress" work-in light of other record evidence. This included Dr. Gopala's own notes, which showed Winsted regularly had a full range of motion, no gross sensory or motor deficits, fine motor skills within normal limits, and lungs that "have often been clear." And though treating physician's opinions, like Dr. Gopala's, are usually entitled to controlling weight, see 20 C.F.R. § 404.1527(c)(2) ; SSR 96-2p,2 an ALJ may discredit the opinion if it is inconsistent with the record. See Loveless v. Colvin , 810 F.3d 502, 507 (7th Cir. 2016) ; Campbell v. Astrue , 627 F.3d 299, 306 (7th Cir. 2010) ; 20 C.F.R. § 404.1527(c)(2).
Next, Winsted argues the ALJ erred in giving little evidentiary weight to Ms. Nevill's mental RFC assessment. But the ALJ wrote he discounted her report because she was a non-medical professional, and thus not an "acceptable medical source" See 20 CFR § 404.1513(a), § 416.913(a). Also, her findings were "based solely on [Winsted's] subjective complaints"-an appropriate reason for an ALJ to discount an opinion, see Ketelboeter v. Astrue , 550 F.3d 620, 625 (7th Cir. 2008). Additionally, he found Ms. Nevill's report, like Dr. Gopala's, was inconsistent with Winsted's medical-health record as a whole. See Filus v. Astrue , 694 F.3d 863, 868 (7th Cir. 2012) (citing 20 C.F.R. § 404.1527(c)(2)-(3) ).
Finally, Winsted claims the ALJ gave short shrift to the two state examiners' 2013 opinions, asserting the ALJ erred in discussing the state psychologist's evaluation "in one sentence" and the state physician's opinion in a footnote. But as the agency points out, the ALJ discussed these opinions throughout the decision. The ALJ cited the state psychologist's findings when discussing Winsted's mental-health diagnosis, and referred repeatedly to the state physician's opinion throughout his discussion of Winsted's gait, grip strength, and scattered wheezing. The court applies a common-sense reading to the entirety of an ALJ's decision. Rice v. Barnhart , 384 F.3d 363, 369 (7th Cir. 2004) ; Shramek v. Apfel, 226 F.3d 809, 811 (7th Cir. 2000).
*473Here, the ALJ adequately articulated his reasons for discounting these two opinions-both reports were based on only one evaluation and largely reflected Winsted's subjective reporting. See Elder , 529 F.3d at 416 ; Rice , 384 F.3d at 371 (ALJs should rely on medical opinions "based on objective observations," not "subjective complaints."); 20 C.F.R. § 404.1527(c)(i) (ALJs should consider "frequency of examination" in weight it assigns opinion).
III. Conclusion
Because the ALJ's hypothetical question to the vocational examiner and the residual function capacity did not capture one of Winsted's most significant problems-his concentration-functioning deficits-we conclude further proceedings are necessary on that issue only. Therefore, we REVERSE the district court judgment and REMAND this case to the Social Security Administration.