When P-Scales ended in 2020, many SENCOs breathed relief. Finally, a framework that acknowledged what we'd always known: pupils with SEND don't develop in neat boxes. But the space that opened up was also the problem. Without a shared language for tracking progress below National Curriculum expectations, schools defaulted to whatever they'd always done, or bought software and hoped for the best. B Squared solved the naming problem. What it doesn't do is tell you how to use it well. Research on SEND assessment consistency (Carpenter, 2015) shows that without structured frameworks, teacher judgement varies by 40 per cent or more. B Squared doesn't eliminate this variance, but it makes it visible and manageable.
This guide walks you through the reality of B Squared assessment: not as a replacement for good teaching, but as a discipline that forces honesty about what pupils can actually do, what they should learn next, and whether we're fooling ourselves about progress.
What Is B Squared and Why Is It Used for SEND Pupils?
B Squared is a curriculum assessment tool designed specifically for pupils working below the National Curriculum. It works where the curriculum doesn't, which is a polite way of saying: it tracks learning in smaller steps for the pupils who need smaller steps.
The DfE's Pre-Key Stage Standards (2020) sit at the top of B Squared. These are six broad levels (PKS1 through PKS4 roughly map to age five through age fourteen). Underneath each standard, B Squared divides learning into "Connecting Steps": individual competencies that add up to the broader standard. One step might be "Pupil responds to their own name in quiet conditions," another "Pupil recognises common objects in photos," another "Pupil initiates communication without adult prompt." This isn't the same as a tidy learning progression. It's more like a map of actual skills that get in the way.
The truth SENCOs live with: you cannot use National Curriculum levels for pupils with significant SEND because the gaps are too large. You can't jump from "cannot reliably identify basic classroom objects" to "sorts objects by colour." That's not differentiation. That's a different curriculum entirely. B Squared names the steps between.
Why it matters. Before B Squared, schools used spreadsheets, their own frameworks, or (this happened more than anyone admits) didn't track SEND pupil progress systematically at all. This left SENCOs vulnerable: unable to evidence progress in EHCP annual reviews, unable to challenge assumptions about what pupils could learn, and unable to spot when "no progress" was actually "no good teaching." The Ofsted SEND review (2023) found that 35 per cent of schools lacked consistent assessment systems for SEND pupils. B Squared forced standardisation. Not perfection, but a shared vocabulary.
A classroom example. In Year 3, Maya has significant language and learning needs. Her previous school's data said "not making progress in communication." B Squared forced specificity: she could request a drink by pointing. She could follow one-step instructions with visual support. She could not initiate communication. Within ten months on Connecting Steps, focused work on this last barrier shifted her to initiating requests. The emphasis on observing actual scaffolding techniques showed where she was stuck. The old spreadsheet would have said "still in Level 1." B Squared said "made two important steps." That difference is why your EHCP panel takes you seriously when you present B Squared data.
After P-Scales: How Connecting Steps Meets DfE Requirements
P-Scales are gone. The Department for Education withdrew them in 2020 specifically because schools were tracking pupils down to P2 and then... freezing them there for years, teaching them nothing new. The assumption was that smaller steps meant smaller expectations. It didn't.
Connecting Steps (the main module in B Squared for pupils below National Curriculum) was designed explicitly to replace P-Scales. The DfE's Pre-Key Stage Standards sit above Connecting Steps as the target framework. If you're tracking a pupil in B Squared, your annual review paperwork should reference which Pre-Key Stage Standard they're working towards, not how many steps they've completed.
What changed in practice. Under P-Scales, a pupil "at P7" was thought to be nearly at Level 1. Under Pre-Key Stage Standards, you say "working towards PKS2 in Reading." This sounds like a tiny shift in language. It completely changes the conversation. A parent, a teacher, an EHCP officer all hear "working towards this recognized standard" instead of "in some weird in-between place."
The compliance problem. Your local authority almost certainly has a B Squared subscription already. Half the schools in the local area are on it. This means: (1) you must be using it for pupils on EHCPs, (2) you cannot claim significant progress without showing step-level movement in B Squared, and (3) audits will check whether your B Squared data matches your annual review narrative. Beating Bureaucracy in SEN (Gross, 2015) advises against parallel systems precisely because they create duplication without clarity. If you're running a parallel assessment system, you're creating work.
Where it breaks. Early development is messier than B Squared assumes. An EYFS class with three pupils with significant needs doesn't fit the "Connecting Steps" framework easily. They're learning to sit, to tolerate transition, to notice other people. This isn't in Connecting Steps because it's theoretically not SEND, it's child development. But these pupils aren't developing typically. We'll come back to how the Engagement Profile solves this.
B Squared acknowledged this problem by creating Pre-Baseline data for very young children and very low-attaining pupils. You can track "Pupil makes eye contact with a preferred adult" before any Connecting Step makes sense. Use it. Your EYFS and early primary SEND pupils won't fit the normal steps anyway.
The Modules: Connecting Steps, SOLAR, and Engagement Profile
B Squared has three main modules. Most SENCOs know this. Most use them inconsistently.
Connecting Steps is the workhorse. These are small learning steps organized by curriculum area and Pre-Key Stage Standard level. A pupil might be working on ten different Connecting Steps simultaneously: maybe "Points to body parts" in Communication, "Grasps a range of objects" in Physical Development, "Responds to simple classroom routine" in Personal Development. Each step has observable criteria, which is the hard part. More on that in a moment.
SOLAR (Structured Observational Learning and Assessment Record) is the monitoring module. You take pupils off Connecting Steps and put them on SOLAR when you want to monitor a specific behaviour or skill in detail without committing to a full Connecting Step. This is where many schools make a mistake: they use SOLAR as a "pending" zone for pupils who haven't been assessed yet. SOLAR should be used when you're intensively observing something. A pupil learning to use visual timetables. A pupil you suspect is dysregulated in large group times. A pupil trying a new communication system. Not "we haven't done baseline yet."
Engagement Profile is where the real insight lives, and where most schools don't dig deep enough. This is not a step-by-step tracker. It's a profile of how a pupil learns, where they are most engaged, what barriers they face, and what support is working. Questions cover: Does the pupil show sustained engagement? With what type of activity? What physical environment helps? What type of adult interaction works best? Are there sensory sensitivities? Does the pupil initiate activity or mostly respond? The Engagement Profile is worth more than twenty Connecting Steps because it tells you whether your teaching is actually landing or whether the pupil is just going through motions.
When to use each. Connecting Steps = "We're teaching this and tracking progress with formative assessment." SOLAR = "We're intensively observing this for two to four weeks." Engagement Profile = "We need to understand this pupil before we write their learning plan." Many schools loop through: Engagement Profile first (baseline), Connecting Steps for teaching and monitoring, SOLAR for troubleshooting when a pupil plateaus, back to Engagement Profile to reassess why.
Setting Meaningful Targets: How to Avoid the Baseline Trap
This is where B Squared assessment meets bad teaching.
Most schools start B Squared well: baseline assessment in the first few weeks, careful observation, thoughtful selection of steps. Then target-setting. And this is where it falls apart, because SENCOs inherit target-setting practices from mainstream assessment that don't work in SEND.
The baseline trap looks like this: A Year 5 pupil is assessed on Connecting Steps. They're able to do 12 out of 30 steps at their relevant standard level. So the target is set: "Complete 15 steps by the end of the year." Three more steps. That's the plan. Eleven months to move through three steps, and somehow this count is a success metric.
What went wrong? The school confused "steps available" with "steps the pupil is ready to learn." A pupil might be unable to do 18 of the available steps, but that doesn't mean they need to learn all of them. Maybe the pupil needs to learn communication first, but 8 of the 18 uncompleted steps are about numeracy. Maybe the pupil can only manage one new concept at a time, so three steps is too ambitious anyway.
Meaningful targets work differently. Choose maybe two curriculum areas to focus on this year. Within one curriculum area, identify three to four Connecting Steps that are genuinely next for this pupil, not just available. Build the teaching plan around those steps. If the pupil completes them, the target is met. If they complete three steps in literacy but plateau in numeracy, that's useful data. It doesn't mean failure; it means "numeracy was too ambitious, try again next year."
A primary example: Ella is in Year 4, globally developmentally delayed. B Squared shows she has mastered 8 steps in Physical Development but only 3 in Communication. The temptation: set targets for Physical Development (she's ready to move) plus Communication (she's behind). Better plan: focus the year on Communication. Why? Because the barrier to almost everything else is that she can't tell anyone what she wants, can't ask for help, can't participate in group activities. Research on growth mindset and target-setting shows that pupils with multiple learning barriers benefit from focused, sequential progression rather than parallel development. Crack that barrier, and next year Physical and Academic Development move faster. The target: "Adds three new steps in Communication (Requests help, Initiates interaction with peer, Uses 10+ words functionally)." That's one focused, defensible target that actually shapes teaching.
The Engagement Profile: Measuring Participation Beyond Curriculum
Standard teaching looks like this: teacher delivers input, pupils practice, pupils show learning. For pupils with SEND, this sequence doesn't work unless the pupil is first engaged. An engaged pupil who struggles with numeracy will have a chance. A disengaged pupil with no learning barriers at all will learn nothing.
The Engagement Profile measures this. It asks: How long does the pupil sustain focus on an activity? Does their engagement differ by subject, setting, or adult? What type of materials hold their attention? Are they engaged with the learning or just compliant? Does the pupil show persistence or avoid challenge?
These questions sound soft. They're not. A pupil who is engaged for three minutes, then avoids, learns at 20% of the rate of one who engages for fifteen. This is why Engagement Profile data should shape your timetable, room layout, and adult deployment at least as much as academic targets do.
An example of what Engagement Profile catches. Amir, Year 6, SLCN and probable autism. His Connecting Steps data is reasonable for his age: he's making progress in most areas. His engagement profile reveals the hidden issue: he's engaged consistently only in structured turn-taking activities. In free-choice time, large-group time, and transition between tasks, he avoids and becomes anxious. His teachers thought Amir was coping. The profile showed he was coping within a narrow band and spent the rest of the day minimizing discomfort, not learning. The evidence for metacognition in self-regulation shows that pupils need explicit instruction in managing transitions, not just natural exposure. The TA's redeployment (from group support to Amir-specific scaffolding through transitions) followed this insight, not from lesson planning, but from the Engagement Profile.
B Squared has a set of engagement descriptors: Early Engagement (brief, reactive focus), Focused Engagement (sustained 5+ minutes), Deep Engagement (shows curiosity and persistence). The profile asks which conditions produce each level. Most SENCOs never scroll down to this section. If you're not looking at which contexts engage your pupils, you're assessing steps on pupils who are barely paying attention. That's not failure to progress; that's failure to create conditions for learning.
Using B Squared Data in EHCP Annual Reviews
Your EHCP annual review document must reference progress in special educational needs pupils. B Squared is the only defensible source for pupils working below National Curriculum. Here's how to make it meaningful instead of a data dump.
What panels actually want to know: Is the pupil making progress at an expected rate? If not, why? Is the provision working? Should it stay the same, increase, or change direction?
What most schools say: Pupil completed 6 new steps this year, compared to 3 last year. Progress is accelerating.
What's actually useful: Pupil completed 6 steps, concentrated in Communication (teacher changed approach from group targets to individual AAC system; this led to faster progress). Plateau in Numeracy (more investigation needed; suspect conceptual barrier rather than motivation). Engagement Profile shows pupil now initiates peer interaction consistently, which is new and important. Revised target: maintain Communication momentum, redirect numeracy teaching.
The panel reads the second version and understands what's happening. They see thinking, not just numbers.
Using B Squared in the review meeting. Bring a one-page summary, not the B Squared printout. Three columns: Area, Progress This Year, Plan for Next Year. Under Progress, use Connecting Steps data but always translate: "3 steps in Literacy = now reads CVC words; previously only read initial letters." Under Plan, be specific about what will change. "Continue current Speech and Language approach (which is working) plus add visual structure support (which Engagement Profile shows helps)." TA deployment research (Webster and Blatchford, 2014) shows that data-driven decisions about support change outcomes significantly more than default practice.
Parent officers see a hundred reviews. They trust the ones that: (1) use assessment data but don't hide behind numbers, (2) explain what the pupil has actually learned in plain language, (3) give a reason for the targets, and (4) connect targets to provision. B Squared does all of this if you translate it.
Writing Progress Reports That Parents Can Understand
Here's something SENCOs rarely hear: parents don't care about Connecting Steps. They care about their child, whether their child is happy, and what the child can do now that they couldn't do before.
A B Squared-based progress report can speak both languages, or it can be technical and useless. Most choose useless by accident.
What not to do: "Jackson has completed Connecting Steps 2, 5, and 9 in Literacy. He is working towards PKS2. This represents continued progress within the Connecting Steps framework."
What actually tells a parent something: "Jackson is learning to recognize and name common objects (his school, his teacher, a cup). This term, he started pointing to the object when you say the word, which is new. Before Christmas, he could only look when you pointed. This is important because he's showing he understands what words mean, which is the foundation for everything else in learning to read and talk."
The shift: from "completed steps" to "can do X, which is new because previously they couldn't, and here's why that matters."
B Squared data enables this report because you have the before-and-after. A step like "Recognises three familiar objects when named" is specific and clear. A parent understands it. You can photograph it happening. You can say "when you ask 'Where's the dog?' he points every time now." That's a progress report that lands.
Many schools still write prose reports for SEND pupils with no reference to actual learning outcomes, just opinions: "Jackson is a lovely boy, engaged and happy, trying hard." Parents of SEND children have heard this a thousand times. They want to know: Did he learn something? What will he learn next? B Squared gives you the evidence to answer these questions honestly. The DfE Pre-Key Stage Standards (2020) explicitly require this shift from descriptive to evidence-based assessment language, not because it sounds nicer, but because it demonstrates real progress.
Giving Pupils a Voice in Their B Squared Targets
Here's what everyone says: pupil voice is important. Here's what most schools do: exclude pupils from target-setting because it's "administrative."
B Squared targets don't need to happen in a staffroom. Much of it can happen with the pupil.
For pupils with communication barriers. Use their communication system. Show them Connecting Steps using visual supports. "This is your current learning. You can do these four things now. This is what we're learning next." For pupils with AAC devices, even very low-tech communication boards, target-setting can include their confirmation. "Are you working on communication next?" (Point to yes/no). "Are you working on numbers?" (Yes, head nod, verbal confirmation, or eye gaze). This practice aligns with research on pupil voice in SEND contexts (Carpenter, 2015): pupils learn best when they understand their own targets, even in simple formats.
For pupils with more developed understanding. Involve them in identifying learning barriers. Many pupils with SEND can name what's hard: "I can't remember the days," "I get confused when there's too much," "I don't know what to do." These observations often map onto Connecting Steps or Engagement Profile findings. A joint conversation ("What are you learning about?", "What's hard?", "What should we practice?") helps pupils invest in their targets and helps teachers see the pupil's perspective on their own learning.
An example. Sasha, Year 5, developmental speech/language needs and mild learning delay. In spring term target review, she was shown her Communication targets on a visual chart: one completed, two in progress, one being introduced. Asked "Which one are you happy about?", she pointed to the completed one and said "I can do it now." Asked "Which is hard?", she pointed to a target about "using sentences with who and what" and demonstrated by saying "I confused. I don't know when." The SENCO rewrote the plan: drop one target (too much), focus the emerging target with sentence stems ("Who are we talking about? What is happening?"), and included Sasha's own words in the review: "Sasha knows which target is hard and is practising every day." Parents read that report, saw their daughter knew what she was learning, and engagement at home improved because there was specificity to support.
Pupil voice in targets doesn't mean pupils choose curriculum. It means pupils know what they're learning and contribute to the plan. B Squared is opaque enough without also hiding it from the pupils involved.
Data Reliability: Managing Subjectivity in Teacher Assessment
B Squared is based on teacher observation. This is its strength and its catastrophic weakness.
Teacher observation is reliable when teachers are trained, consistent, and scrutinized. Most schools have none of these things.
A Connecting Step like "Pupil grasps a range of objects" requires judgment. One teacher marks this as achieved when a pupil grasps three objects in a structured session. Another marks it achieved when the pupil grasps any object at any time, including accidentally grabbing. By the end of the year, the first teacher's data shows the pupil made two steps. The second teacher's data shows five. Same pupil, same teaching, same progress. Different data. This inconsistency is well documented in assessment research (Carpenter, 2015).
Where inconsistency hides. Across year groups (Year 1 teacher is stricter than Year 2), across staff (TAs assess differently than teachers), across time (everyone's standards slip by summer term), and across contexts (pupil shows the skill in one-to-one but not in group).
This is why B Squared data needs moderation.
Moderation done well. A group of teachers from your school (or your local authority if you're tiny) gather with the same pupil's observations and decide: does this pupil meet this step or not? The conversation surfaces hidden standards. "I would have marked that as achieved, but actually the pupil did it once in ten tries." "I'm counting scaffolded success as independent. Should I be?"
Moderation also builds confidence. Teachers realize their standards are roughly aligned and start feeling secure in their assessments.
Moderation done badly. Schools run moderation as a checkbox: "Everyone agrees she's on step 5." No conversation, no scrutiny, just consensus. You now have aligned-but-wrong data, which is worse than nothing because it's confidently wrong.
If you're a SENCO in a school of fewer than a hundred pupils, you probably can't moderate internally. Join your local authority's moderation network, or do paired moderation with another school's SENCO. Once a term, spend two hours on B Squared consistency. You'll uncover things (hidden standards, unclear step criteria, TAs assessing too generously) that improve both assessment and teaching.
A hidden problem: digital drift. Once a pupil is marked as having achieved a step in B Squared, the system shows that as done. Except nothing prevents a pupil from losing a skill. A pupil learns to use a cup with two hands; six months later, due to decreased tone, they're back to needing to use both hands gripping. Has the pupil lost the step, or is this temporary? B Squared doesn't capture regression well. You need a teaching conversation in addition to your data system: "Has this pupil maintained skills over time, or slipped on anything?" Write it down somewhere your successor will see it.
B Squared in Transition Planning: Primary to Secondary
Transition is when inconsistent B Squared data becomes dangerous.
A Year 6 pupil who is assessed at a higher level than they actually perform will arrive at secondary and immediately fail. Their new teachers will assume they can do things they can't, set targets they can't meet, and the pupil's confidence crashes. Alternatively, a pupil assessed conservatively will arrive with low expectations and won't be challenged. Both happen because the secondary school trusts the data without checking.
Before transition. Do moderation moderation across primary and secondary. Invite the secondary teachers to your summer term data meetings. Show them the pupil's Connecting Steps and Engagement Profile. More importantly: show them video or photos of the pupil doing the skill. If you mark "Uses single words to request" without showing what "request" means in your school, the secondary teacher might think the pupil can do something they can't.
Transition visits should always include a B Squared discussion. Not "here's your data" but "here's what this pupil can do, here's how we've been supporting them, here's what they find hard."
What to transfer. Don't transfer last year's targets. The secondary teacher will have different priorities, different curriculum, different approaches. Do transfer the Engagement Profile. This is the gold: "This pupil learns best with visual structure and one-to-one check-ins," or "This pupil engages consistently with peer activities, less with adult-led work." The Engagement Profile is curriculum-independent. The secondary teacher can't teach without knowing how this pupil engages.
Transfer high-priority Connecting Steps for areas where the pupil is close to progress. "They've completed 4 of 6 communication steps at this level" is more useful than twenty steps of background data.
An example of what goes wrong: Jamal, Year 7, SLCN. Primary school data shows he's been "working on" a communication step for two years: "Uses 5-10 words functionally." In primary, he did use about eight words, mostly with visual support. Secondary teacher sees data, assumes he's stuck on a low target, doesn't push. Jamal stays at eight words all year. If the primary data had said "Uses 8 words in structured settings, 3-4 in unstructured; target this year: expand to more contexts," secondary might have approached it differently.
Getting Started: Training Your Team and Onboarding
Most of this guide assumes your school already uses B Squared, or at least has a subscription. If you're just starting, the first mistake is uploading pupil data without training anyone.
Week 1: Admin foundations. Someone (you, probably) needs to set up the B Squared system correctly. Pupils entered with correct year group, starting date, and baseline assessment completed before teachers start recording steps. This sounds basic. Many schools create a thousand pupils in B Squared and then nobody knows whose class is whose.
Then run a whole-school training on what B Squared is and isn't. Not a two-hour flog through the software. A one-hour conversation about: What are Connecting Steps? Why do we track below National Curriculum? What's the Engagement Profile for? How will this change your day-to-day?
Week 2-4: Subject moderation. Bring teachers together by subject (Literacy people, then Numeracy, then Physical, etc.) to look at actual steps and ask: What does this step actually mean? When have we seen this? What are we looking for? This isn't rote learning. This is building shared understanding. A Literacy teacher might think "Recognises own name" means the child reads it unprompted in any context. Another might count recognising it on a label in their own tray. Clarify this before everyone starts assessing.
Month 2: First baseline. Teachers start entering baseline observations. The temptation: rush this. Resist. A slow, careful baseline with well-considered evidence (observation notes, photos, even short videos) takes longer but prevents six months of corrections. If you rush baseline, you'll spend October and November hunting down teachers for missing details.
Month 3: First review. Look at data not to judge progress (it's too early) but to audit for mistakes. Did everyone understand what a step is? Are there steps nobody has ever entered data for (probably means the step is confusing)? Are there year groups where almost everyone's marked the same step achieved (suggests either the step is too easy or everyone's being generous)?
Ongoing: Monthly SENCO supervision with class staff. Not "check in on data," which is soul-crushing. Have a real conversation about one pupil: "This pupil's Connecting Steps look good, but their Engagement Profile worries me. What are we seeing?" That conversation is the point. The data is just the starting point.
Training materials. B Squared provides video training. Watch it once. Then forget it. Instead, create your own one-page guide to your school's standards. "When we say a pupil has 'completed' a Connecting Step, we mean they've shown the skill on at least three occasions, in more than one context, with minimal adult support (or whatever your standard is)." Every teacher gets a copy. Every September, it's reviewed. This document is worth more than the software manual.
TAs need specific training. Teaching Assistants often carry out observations and record steps. They need to understand that their observations should be evidence-based and specific, not inferences. "Requested help three times independently this week" is evidence. "Pupil is understanding they can ask for help" is an inference that might follow from it, but they're not the same. TAs who are trained to observe precisely will give you data you can actually use. This separation of observation and interpretation is fundamental to reliable assessment (NASEN guidance on SEND assessment).
Get buy-in from the deputy or other leader. B Squared success requires that someone (probably not the SENCO, who has seventy other things) champions it. This person checks in termly, celebrates examples of good data being used to change teaching, and holds the line on moderate (don't let it become a data-entry burden). Without a champion, B Squared becomes your solo project and it won't scale.
---
The honest truth about B Squared: it's only as good as you make it. The software is neutral. It doesn't make anyone teach better. What it does is force specificity. You can no longer say "SEND progress is hard to measure." You have to say what the pupil learned, when they learned it, and what they're learning next. That conversation is where better teaching happens.
Start with one thing: meaningful targets that actually reflect what pupils should learn, not what steps exist. Get that right, and the rest of B Squared becomes useful instead of burdensome.
